Advanced Euchre QUIZ - PART 2

Ask questions, discuss and debate your strategies, euchre polls and more
irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Advanced Euchre QUIZ - PART 2

Unread post by irishwolf » Mon Mar 11, 2019 12:17 pm

Here is a link to the original post titled "Advanced Euchre QUIZ" by Wes on 2/24/19

viewtopic.php?t=111

===========================================================
Response to Answers on The QUIZ:

Realize this is not grade school where there is ONE right answer. The correct answer is - where you achieve your objective according to what occur from your partners the Majority of the time (statistically). And that is severely lacking for each or at least most of the answers. And I probably agreed with Wes (at least in part) than most who did the Quiz. As my answers are, it is one person's opinion even though may be a highly skilled person. Various answers based on opponents. Euchre is a statistical game and what players do is more variable than the cards themselves. So how you play each hand also has to change accordingly. Realize these answers by WES are by an aggressive player. Knowing that, he is vulnerable to BAGGING. Set back and wait for some aggressive calls.

For example #19: His answer is statistically incorrect for many hands played - # 19. WHY- leading a bower first is not the Best way to get a euchre. Eldest has two bowers and MUST have a trick from your partner. Leading a bower will strip your partner of his one trump but more importantly - it will shut down any further trump lead from the dealer. You have to first try to hit your partner for an ace or to trump your lead. If the dealer trumps, he/she will now lead trump and you get both hits by your bowers AND another off suit lead to your partner for a possible Ace. In addition, your partner can sort his hand to save an off suit doubleton to catch the dealer's off suit.
This set up occurs a often - I have studied it in detail and confident in my statements above.

#9 - score is 9 to 6 - better to go for two points and IT 'S YOUR DEAL NEXT. The answer almost assumes the opponents are UNAWARE of what a loner in next would do. I would agree only if opponents were weak but that is not what was given to ASSUME for answering the questions.

#24. The answer is another one. 9D is the upcard - score is 0 to 0. One is you are playing as if you alone and not considering your partner has five cards. What do you think are the odds of the dealer having a loner, or his partner with a 9 up? VERY LOW - statistically conceding one point - 10% advantage to the opponent many hands played is the answer. And if assuming you believe Natty, he would never donate at 0 t0 0. If the Jack were up then statistically it goes up (to 1 in 6 hands). Is that good enough to donate as well.

#25 - score is 0 - 3. Going alone - if opponents lead to your void - clubs you have to use your ace and lead the left. Opponent will have the right 56% of the time. Another lead of clubs you have no answer. And one of those opponents will have another trump for a euchre. Eldest leading from 3 of an off suit to one of those aces, Pone has a good chance to trump. A euchre now yo0u are down 5 to 0. You will lose most games that way. Assist and go for 2 pts - now only down by 1.

Off hand I take serious issue with 9., 19., 23., 24., 25., 28. There are others but quick review these stand out.

Oh there is much more but enough for now.

Irishwolf



irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Mar 11, 2019 1:24 pm

Question 1 - said not to go alone? However it depends on your two opponents. It said nothing about where in games you were or if either of the two opponents may be top competitors. If top competitors it is just as important to keep both at ZERO.

NO guarantee you will end up with 11, 12, or 13.

RedDuke
Posts: 460
Joined: Sun Dec 02, 2018 2:22 am

Unread post by RedDuke » Mon Mar 11, 2019 1:59 pm

16) The score is 6-6. You're in the 3rd seat. The dealer upcard is the (Card_A-S)

You hold (Card_J-C), (Card_K-S), (Card_Q-S), (Card_A-D), (Card_J-H)

ANSWER: Pass. Too many good things can happen if you pass here. 1) If the dealer picks up you have a great chance to set him (all he needs is the Right bower to justify an order). And 2) If the dealer passes you hit your partner no what he calls, and those times your partner goes alone you'll also be glad you passed. And the worse case scenario, the dealer and your partner both passing isn't that bad as you have all suits blocked. Ordering this hand up in the first round from the 3rd spot and getting set would be a small tragedy since it was a risk you didn't need to take given the upside to passing--a real risk to always respect since 3rd seat orders are the toughest ones to make.
This one could even be a great pass second round if you're playing stick-the-dealer. No matter what the dealer calls, you've got a decent chance to euchre him.

You've got a guaranteed trick in any suit. You probably have at least two and likely three no matter what trump is. Figure:

1. If dealer passes, Ace-Spades is now out of play. Your King is now boss in spades and will probably take a trick.
2. If dealer calls clubs, you have the right.
3. If dealer calls hearts (reverse next), you have the right. If he calls diamonds, you have a guarded left. Either one is good for a trick.
4. The Ace of Diamonds will probably take a trick.

Thus, you've probably got two or three tricks here no matter what any other player does. That's a hand that you're always going to want to pass.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Mar 11, 2019 2:50 pm

RedDuke wrote:
Mon Mar 11, 2019 1:59 pm
16) The score is 6-6. You're in the 3rd seat. The dealer upcard is the (Card_A-S)

You hold (Card_J-C), (Card_K-S), (Card_Q-S), (Card_A-D), (Card_J-H)

ANSWER: Pass. Too many good things can happen if you pass here. 1) If the dealer picks up you have a great chance to set him (all he needs is the Right bower to justify an order). And 2) If the dealer passes you hit your partner no what he calls, and those times your partner goes alone you'll also be glad you passed. And the worse case scenario, the dealer and your partner both passing isn't that bad as you have all suits blocked. Ordering this hand up in the first round from the 3rd spot and getting set would be a small tragedy since it was a risk you didn't need to take given the upside to passing--a real risk to always respect since 3rd seat orders are the toughest ones to make.
This one could even be a great pass second round if you're playing stick-the-dealer. No matter what the dealer calls, you've got a decent chance to euchre him.

You've got a guaranteed trick in any suit. You probably have at least two and likely three no matter what trump is. Figure:

1. If dealer passes, Ace-Spades is now out of play. Your King is now boss in spades and will probably take a trick.
2. If dealer calls clubs, you have the right.
3. If dealer calls hearts (reverse next), you have the right. If he calls diamonds, you have a guarded left. Either one is good for a trick.
4. The Ace of Diamonds will probably take a trick.

Thus, you've probably got two or three tricks here no matter what any other player does. That's a hand that you're always going to want to pass.
Yep exactly. Funny thing is this hand comes from a hand I played in my local tournament. I had the above hand in the 3rd spot and correctly passed. Then the dealer went alone with:

(Card_J-S) (Card_A-S) (Card_10-S) (Card_9-S) (Card_A-C)

She ended up getting euchred. Notice you can't get euchred on her hand unless you play it wrong. She took the first trick with a low spade, then sent out the Js, and then on the next play she lead the As instead of the Ac, allowing me to clean her out. Strictly speaking, her play is not wrong, but score dependent. If she were down 9-6, she should play the As before the Ac. Maximizing her chances at a 4 point sweep is worth the small risk of getting euchred.

Anyways, she was in utter shock after the hand, even during the hand. When I overtrumped her As and sent the Ks her exact words were: "Oh my god, Oh my god, I'm actually gonna get euchred on this hand. This is unreal." She was despondent, in shock for a solid 2 minutes afterward. I did my best to console her trying to explain to her that there is no shame in getting set by a legend. The "F U" look on her face was classic :)
Last edited by Wes (aka the legend) on Mon Mar 11, 2019 4:06 pm, edited 1 time in total.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Mar 11, 2019 3:03 pm

irishwolf wrote:
Mon Mar 11, 2019 1:24 pm
Question 1 - said not to go alone? However it depends on your two opponents. It said nothing about where in games you were or if either of the two opponents may be top competitors. If top competitors it is just as important to keep both at ZERO.
Sure there's other variables to consider. Point is if you're already up 6-0, you're more likely to end the tournament with a higher score if you just call with that perfect loner. Generally speaking, Going up 8-0 has more value than winning the game 10-0, so calling should be your default strategy unless you feel you have a strong reason to deviate. Also I intentionally made the scenario for game 1 so you actually had no idea how your night was gonna go. Most likely you're gonna have a sub-par game or two score-wise, so when you have a great chance to grab some extra points it's worth going for.
irishwolf wrote:
Mon Mar 11, 2019 1:24 pm
NO guarantee you will end up with 11, 12, or 13.
I don't care about guarantees, as long as the odds are on my side.

Based on the probability chart I linked to, we have a 96% chance of getting 10, 11, 12, or 13 points and a 4% chance of only getting 8 or 9 points. The risk is worth it especially when you consider the fact that the 4% of the time you get 8/9 is not a doomsday scenario. That's still a quality loss points-wise which keeps you firmly in the hunt to win the tournament.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Mar 11, 2019 4:05 pm

irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
Response to Answers on The QUIZ:

Realize this is not grade school where there is ONE right answer. The correct answer is - where you achieve your objective according to what occur from your partners the Majority of the time (statistically). And that is severely lacking for each or at least most of the answers.
I'm ok with this. As long as there is not some euchre simulator out there where we can run a scenario a millions times and compare it to other scenarios, there will be some uncertainty we all have to live with (for now).
irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
RAnd I probably agreed with Wes (at least in part) than most who did the Quiz. As my answers are, it is one person's opinion even though may be a highly skilled person. Various answers based on opponents. Euchre is a statistical game and what players do is more variable than the cards themselves. So how you play each hand also has to change accordingly.


There are adjustments you have to make for bad partners, highly skilled partners and the same for opponents. That is outside the scope of the quiz, but the quiz can be used as a tool to stimulate that kind've discussion. For example, in hand 11) I advocated a pass from the 3rd seat. If your partner is a bad player who passes a lot and doesn't go alone enough, then one can argue that's a good reason to call. Conversely, if seat 1 is highly skilled, calling from the 3rd seat with that holding is even more likely to be a mistake.
irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
Realize these answers by WES are by an aggressive player. Knowing that, he is vulnerable to BAGGING. Set back and wait for some aggressive calls.
I think worrying about bagging is just being paranoid, and there's often nothing you can do about it anyways. If the enemy turns down a black card and you have nothing in red, you gotta call thin in Next no matter what the nature of the enemy is. Yes vs conservative opponents you'll go set more, but you'll also get to call more. The latter outweighs the annoyance of the former in the long run.

Altho I never worry about bagging, there is a time I get concerned, and that's when there is a pro in the 1st seat. If a pro's in the first seat, I have to tighten up on my reverse next calls. Almost by definition a pro is not gonna pass in the 2nd rd from the 1st seat unless he has reverse next blocked, which means calling reverse next too weak isn't gonna work out that well. But even this is kindve a mythical problem. I play on an app where around 10k people play per day, and only like 5-10 players, I.E. .01% play at that top level.

That said, there is certainly room for bagging strategically. E.G. you're in the dealer seat and a pro is in first seat. If you have a biddable hand that blocks all suits and has a strong chance of setting a next call then pass.

Upcard: (Card_K-S)

Your hand: (Card_A-H) (Card_J-D) (Card_J-S) (Card_10-C) (Card_9-C)

Another example.

Upcard: (Card_9-C)

Your hand: (Card_A-H) (Card_J-D) (Card_K-S) (Card_J-S) (Card_10-C)
irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
For example #19: His answer is statistically incorrect for many hands played - # 19. WHY- leading a bower first is not the Best way to get a euchre. Eldest has two bowers and MUST have a trick from your partner. Leading a bower will strip your partner of his one trump but more importantly - it will shut down any further trump lead from the dealer. You have to first try to hit your partner for an ace or to trump your lead. If the dealer trumps, he/she will now lead trump and you get both hits by your bowers AND another off suit lead to your partner for a possible Ace. In addition, your partner can sort his hand to save an off suit doubleton to catch the dealer's off suit.
This set up occurs a often - I have studied it in detail and confident in my statements above.
I don't feel strongly one way or another about this hand. I posted it becuz of the good discussion on it at another site. I am fully aware of the conventional strategy here. I know we are all taught not to lead a bower in this spot. To me, just sharing the fact that other strong players--who take the game very seriously--have a different opinion on something we all tend to take for granted as true (along with a well thought out argument), has value in itself.
irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
#9 - score is 9 to 6 - better to go for two points and IT 'S YOUR DEAL NEXT. The answer almost assumes the opponents are UNAWARE of what a loner in next would do. I would agree only if opponents were weak but that is not what was given to ASSUME for answering the questions.
Not sure what you mean by "weak" but 99% of euchre players just play their cards. Either way, I understand if people disagree with me on this one. This "loner" has worked out often enough for me to give it a try, and it still scores a point a decent amount of times too. Everyone's free to disagree, but down 9-6 I'm going for the win. I couldn't live with myself otherwise. That said, if that mythical euchre simulator says I shouldn't then I would change up.
irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
#24. The answer is another one. 9D is the upcard - score is 0 to 0. One is you are playing as if you alone and not considering your partner has five cards. What do you think are the odds of the dealer having a loner, or his partner with a 9 up? VERY LOW - statistically conceding one point - 10% advantage to the opponent many hands played is the answer. And if assuming you believe Natty, he would never donate at 0 t0 0. If the Jack were up then statistically it goes up (to 1 in 6 hands). Is that good enough to donate as well.
I can't prove I'm right. Just saying what I believe is right. Also, I'm not playing as if I'm alone. I have 6 exposed cards of information to work with here, and my job as seat 1, the most important seat in the game, is to utilize that information and do what's best for my team. The best evidence I have here is I have literally played in some of the toughest games in world. Remember those 5-10 pros on the app. It's rare, but I have played in games on the app where a pro is sitting in all 4 seats, and they all insta-donate in that scenario. Again, that doesn't prove anything, but it is some evidence. Math is still the ultimate answer, but we don't have a euchre simulator so we just have to accept the possibility we are wrong, especially in controversial spots. I've laid out why I think it's a donate, but others are free to disagree.
irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
#25 - score is 0 - 3. Going alone - if opponents lead to your void - clubs you have to use your ace and lead the left. Opponent will have the right 56% of the time. Another lead of clubs you have no answer. And one of those opponents will have another trump for a euchre. Eldest leading from 3 of an off suit to one of those aces, Pone has a good chance to trump. A euchre now yo0u are down 5 to 0. You will lose most games that way. Assist and go for 2 pts - now only down by 1.
Yes things can go wrong, but this hand has a lot of good things going for it too. I'm fine with people disagreeing with me on this one. If they at least consider this hand as a loner candidate then I've accomplished my goal even if they end up rejecting the idea.
Last edited by Wes (aka the legend) on Thu Mar 14, 2019 2:49 am, edited 3 times in total.

irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Mar 11, 2019 5:03 pm

You were playing against a dumb ASS. You can't euchre me in a MILLION years with me having JS AS 10S 9S AC going alone - only if I am dumb enough to lead AS on trick 3. Why do this when only third seat played a trump to my JS. I lead my AC on trick 3 you have to use a trump and now only have the Left and I have AS 10S 9S.

Laughable!


YOU SAID:
Yep exactly. Funny thing is this hand comes from a hand I played in my local tournament. I had the above hand in the 3rd spot and correctly passed. Then the dealer went alone with:

(Card_J-S) (Card_A-S) (Card_10-S) (Card_9-S) (Card_A-C)

She ended up getting euchred. Notice you can't get euchred on her hand unless you play it wrong. She took the first trick with a low spade, then sent out the Js, and then on the next play she lead the As instead of the Ac, allowing me to clean her out. Strictly speaking, her play is not wrong, but score dependent. If she were down 9-6, she should play the As before the Ac. Maximizing her chances at a 4 point sweep is worth the small risk of getting euchred.

Anyways, she was in utter shock after the hand, even during the hand. When I overtrumped her As and sent the Ks her exact words were: "Oh my god, Oh my god, I'm actually gonna get euchred on this hand. This is unreal." She was despondent, in shock for a solid 2 minutes afterward. I did my best to console her trying to explain to her that there is no shame in getting set by a legend. The "F U" look on her face was classic :)

irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Mar 11, 2019 10:34 pm

QUIZ # 10 - Order the dealer from 1st seat - this hand is weaker than 11 & 16.

irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Mar 11, 2019 10:46 pm

Again, these answers are opinions without statistics or hands played to back up the answers to those that are marginal. Proof comes by way of statistical analysis and or several random hands set up and played. This is lacking for a concrete strategy for play. Those I suggested as marginal for starters.

Verbiage is not proof, called Gut probability and it is frequently incorrect.

WES - Your QUIZ AND ANSWERS!

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Mar 11, 2019 11:26 pm

irishwolf wrote:
Mon Mar 11, 2019 10:34 pm
QUIZ # 10 - Order the dealer from 1st seat - this hand is weaker than 11 & 16.
You're absolutely correct. That hand is weaker than 11 & 16, and yet those answers still stand. That's the wonderful thing about Euchre. It's not just about playing your cards, the game is more complex than that. It's about playing the situation. In fact, to be even more bold, if one doesn't understand why the above apparent contradiction is actually no contradiction at all, they will never be a great euchre player.

irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Tue Mar 12, 2019 10:57 am

WES,
You do a lot of chess beating, I am not so sure it's justified. You have not proven your self proclaimed top of the heap. i take it you believe you have absolute authority on your answers as being correct. I say otherwise!

Hand #16 - what is your EV of being euchred? What is your EV on dealer making trump? What is your EV of Eldest in calling (and of what)?

You have to have some strategy with specific EV?

Irishwolf


You're absolutely correct. That hand is weaker than 11 & 16, and yet those answers still stand. That's the wonderful thing about Euchre. It's not just about playing your cards, the game is more complex than that. It's about playing the situation. In fact, to be even more bold, if one doesn't understand why the above apparent contradiction is actually no contradiction at all, they will never be a great euchre player.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Tue Mar 12, 2019 1:46 pm

irishwolf wrote:
Tue Mar 12, 2019 10:57 am
WES,
You do a lot of chess beating, I am not so sure it's justified. You have not proven your self proclaimed top of the heap. i take it you believe you have absolute authority on your answers as being correct. I say otherwise!

Hand #16 - what is your EV of being euchred? What is your EV on dealer making trump? What is your EV of Eldest in calling (and of what)?

You have to have some strategy with specific EV?

Irishwolf
I've said it before and I'll say it again. We don't have that theoretical euchre simulator. We can't just run a spot a million times and compare different strategies. In poker, we have large databases available played by experts online and we can see how they played certain spots. As far as I know, we don't even have that with euchre. So, yes, there is going to be a degree of uncertainty here. All I can do is present my answers and tell people why I think they are correct. If my reasons aren't compelling, anyone is free to reject them. If you or anyone else thinks I suck at euchre that's fine. I don't care. It's just a game man.

PS: And yes, I strongly think #16 is a clear pass. I've made my case. I think RedDuke explains it even better. Calling in the first round from the 3rd seat and fighting for a point is a very poor way to play a euchre hand in my expert, esteemed, prestigious, holy opinion.

irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Tue Mar 12, 2019 5:35 pm

You failed to answer my Questions on Your EVs? Let's face it - you use SWAG and are not open to any analysis - ANOTHER CLOSED MIND. Self appointed expert.

Let me assure you the answer is by ordering you will score a point 73% +/- 5% OF THE TIME. That is the best strategy on this hand. I could break this down statistically. For example the JS will be buried or with my partner - 44.5%. Additionally, the JS will be at 2nd seat and unguarded 50% of the time and that AS will be the only trump he has 35% of the time. Your known statistically is in shambles - spot a million times, times? Where did you get that idea?

Tired of wasting my time with you. So much for a Legend. You get about 75% on your own test - C+.

IrishWolf

I've said it before and I'll say it again. We don't have that theoretical euchre simulator. We can't just run a and compare different strategies. In poker, we have large databases available played by experts online and we can see how they played certain spots. As far as I know, we don't even have that with euchre. So, yes, there is going to be a degree of uncertainty here. All I can do is present my answers and tell people why I think they are correct. If my reasons aren't compelling, anyone is free to reject them. If you or anyone else thinks I suck at euchre that's fine. I don't care. It's just a game man.

PS: And yes, I strongly think #16 is a clear pass. I've made my case. I think RedDuke explains it even better. Calling in the first round from the 3rd seat and fighting for a point is a very poor way to play a euchre hand in my expert, esteemed, prestigious, holy opinion.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Tue Mar 12, 2019 7:52 pm

irishwolf wrote:
Tue Mar 12, 2019 5:35 pm
You failed to answer my Questions on Your EVs? Let's face it - you use SWAG and are not open to any analysis - ANOTHER CLOSED MIND. Self appointed expert.

Let me assure you the answer is by ordering you will score a point 73% +/- 5% OF THE TIME. That is the best strategy on this hand. I could break this down statistically. For example the JS will be buried or with my partner - 44.5%. Additionally, the JS will be at 2nd seat and unguarded 50% of the time and that AS will be the only trump he has 35% of the time. Your known statistically is in shambles - spot a million times, times? Where did you get that idea?

Tired of wasting my time with you. So much for a Legend. You get about 75% on your own test - C+.

IrishWolf
You have not come close to proving your case. You'd have to run a simulation, with a very large sample size comparing the outcome of calling vs passing with this hand. For now, whether we like it or not we'll have to rely on "swag" if that's what you wanna call it. And my Swag-O-Meter says trying to fight it out for a point in the toughest seat in the game with a strong euchre hand--I.E. a hand that blocks all suits with nearly 2 tricks in every suit--is poor level zero play that conveys a facile understanding of the game.

PS: Just found out that Swag-O-Meter is actually a real word on urbandictionary rendering this discussion more fruitful than anticipated.

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Thu Jan 06, 2022 6:15 pm

So I have started running these hands on my program to see how it fares, and already have some cool results.

Problem #1: You're in a Euchre tournament. The structure is very simple. You play 6 games with random partners. The person with the most points in the end wins. Each game is up to ten with the max points possible being 13. You are in game one.

You are the dealer, and your team is up 6-0

The upcard is the (Card_K-C)

You hold (Card_J-S), (Card_J-C), (Card_9-D), (Card_A-C), (Card_Q-C)

My gut tells me it is a mistake to pass on a gift like a loner hand, but I can't think of a way to prove or even demonstrate which strategy is better. Giving points to your adversaries just has to be wrong (so much luck of the hands dealt, even if these opponents aren't on your radar as "experts" they aren't out of the running). I'll keep noodling on this, hopefully will think of something.

Problem #2: The score is 0-0. You are the dealer.

The upcard is the (Card_10-C)

You hold (Card_A-D), (Card_K-C), (Card_10-H), (Card_9-H), (Card_9-S)

My program passed this hand. So I tried calling Clubs instead, and found a slightly better EV (-0.65 vs. -0.73; 100,000 hands simulated, 61,842 made it to dealer for a decision)

I then tested the other scenarios Wes suggested (and a couple of my own):
a) replace 9H with 9D (create doubleton green Ace):
bid EV = -0.55; pass EV = -0.73; 61,933 of 100,000 hands
10H now discarded, so better EV than base hand; still better to call

b) replace AD with AS and 9S with 9D (create singleton next Ace):
bid EV = -0.67; pass EV = -0.85; 61,663 of 100,000 hands
10D discarded, but Ace in shorter suit, so slightly worse EV than base hand; still better to call

c) replace KC with 9C (worse trump holding):
bid EV = -0.93; pass EV = -0.85; 60,021 of 100,000 hands
now better to pass!

d) replace KC with QC:
bid EV = -0.79; pass EV = -0.78; EV between base case and previous case is bid (intermediate trump holding); too close to say which strategy is better

I introduced a new rule in my program for S4 bidding R1:
if have:
- no jacks
- 2 trump led by A or K
- at least 1 off-suit Ace
then bid.

I then tested this new rule for 1,000,000 random hands. It was applied 16,013 times, and the net extra points earned by S2/S4 was 1,989, for an incremental EV gain or +0.12. For those who are interested, here is the change in the # of hands won/lost by each bidder:

S4R1: 0 / 0 / 0 || +324 / +8,060 / +7,627 [so actually lose points here]
S1R2: -1,019 / -2,340 / -210 || -891 / -5380 / -2030 [keep S1 from scoring lots of points here!]
S2R2: -96 / -349 / -42 || -249 / -1,756 / -877
S3R2: -29 / -61 / -1 || -77 / -398 / -99
S4R2: 0 / -1 / 0 || -5 / -54 / -49

Overall, both teams score more points when S4 bids R1, but S4 improves by more.

As one final check I added one more rule: that S4 has 0 or 1 voids. This means that S1 does NOT have 2 trump + A-x-x. I again ran this simulation over 1,000,000 hands. It was invoked 13,223 times, with S2/S4
earning an incremental 497 more points vs. S1/S3, for a positive incremental EV of +0.04. So the restriction of not counting A-x-x was detrimental.

I ran this last simulation once again, for 1,000,000, and got the same result. So I tested the following hand: A-10-9D + KC + 9S, 10C turned. I tested this over 100,000 hands (61,636 made it dealer for bidding).
bid C: EV = -0.27
pass: EV = -0.70

So it would seem that even being two-suited (i.e, having a tripleton Ace), it's better to bid. In fact, it may even make more of a difference than having a singleton or doubleton Ace (I only tested this one example, but that's what my overall results seem to show).

So in the end, my new rule for S4R1 bidding is to BID with:
- no jacks
- 2 trump led by A or K
- at least 1 off-suit Ace (next or green; singleton, doubleton of tripleton)

I'm working on question #3 now, with similar interesting results...

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Thu Jan 06, 2022 7:36 pm

irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
Response to Answers on The QUIZ:

Realize this is not grade school where there is ONE right answer. The correct answer is - where you achieve your objective according to what occur from your partners the Majority of the time (statistically). And that is severely lacking for each or at least most of the answers. And I probably agreed with Wes (at least in part) than most who did the Quiz. As my answers are, it is one person's opinion even though may be a highly skilled person. Various answers based on opponents. Euchre is a statistical game and what players do is more variable than the cards themselves. So how you play each hand also has to change accordingly. Realize these answers by WES are by an aggressive player. Knowing that, he is vulnerable to BAGGING. Set back and wait for some aggressive calls.
I agree with the above I think. Yes none of these hands have statistical backing. All of these hands are based on my expert intuition. Expert intuition is a nice thing but it doesn't come close to actually analyzing a hand statistically. The probability that every answer is correct is pretty much zero. What's more important to me is 1) this quiz is a great learning tool as it did bring to the forefront a lot of concepts that can help fine-tune one's game and 2) it was fun!
irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
For example #19: His answer is statistically incorrect for many hands played - # 19. WHY- leading a bower first is not the Best way to get a euchre. Eldest has two bowers and MUST have a trick from your partner. Leading a bower will strip your partner of his one trump but more importantly - it will shut down any further trump lead from the dealer. You have to first try to hit your partner for an ace or to trump your lead. If the dealer trumps, he/she will now lead trump and you get both hits by your bowers AND another off suit lead to your partner for a possible Ace. In addition, your partner can sort his hand to save an off suit doubleton to catch the dealer's off suit.
This set up occurs a often - I have studied it in detail and confident in my statements above.
My answer was based on a 50 hand sample done by Natty. A laughably small sample size and yeah there's a decent chance it's wrong. That said, I liked this one becuz Natty's answer ran counterintuitive to what most people think. Most people feel strongly about not leading a bower in this spot becuz that's what they were taught. I would go with the results of a 50 hand sample over "feels" any day but sometimes "feels" is right and the 50 hand sample is wrong. That's why we need a bigger hand sample. If you say leading a bower is statistically incorrect I would trust your answer as I know you've done the work. I'll also be interested in what Ray's simulation pumps out on this.
irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
#9 - score is 9 to 6 - better to go for two points and IT 'S YOUR DEAL NEXT. The answer almost assumes the opponents are UNAWARE of what a loner in next would do. I would agree only if opponents were weak but that is not what was given to ASSUME for answering the questions.
There's some confusion here becuz I can't tell if you're attacking my original answer of go alone in Next or my updated answer of go alone in hearts, lead the Right, and sell out and lead the Ace of trump even if the Left has not revealed itself. Also, I don't think this statement as a valid criticism: "The answer almost assumes the opponents are UNAWARE of what a loner in next would do". At least in 95% of game textures that will indeed be the case and I think I'm being conservative when I say 95%. I think it's closer to 99%. Whenever one plays euchre against randoms like on the app/online I think one should always assume their opponents are weak until proven otherwise. If you're playing in some home game with experts adjust accordingly but that game texture is exceedingly rare. The vast majority of euchre players simply don't care enough to take this game seriously, and I can't blame them. Their outlook is more rational imo. :)
irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
#24. The answer is another one. 9D is the upcard - score is 0 to 0. One is you are playing as if you alone and not considering your partner has five cards. What do you think are the odds of the dealer having a loner, or his partner with a 9 up? VERY LOW - statistically conceding one point - 10% advantage to the opponent many hands played is the answer. And if assuming you believe Natty, he would never donate at 0 t0 0. If the Jack were up then statistically it goes up (to 1 in 6 hands). Is that good enough to donate as well.
I think there's enough statistical evidence now where I have to agree with you on this. But still I wonder.....what if our team has a large advantage over our opponents. Perhaps it may be worth donating with this hand at 0-0 burning around a half a point in the long run knowing our opponents will burn way more than that throughout the game. I propose this becuz when a team has a large advantage in skill over their opponents often loner luck is the factor that allows the worse team to win. That said, if up 2 pts I'm still donating in this spot every time with this holding. Is it worth burning a half a point in the long run when up 2 pts to control some variance? I doubt we'll ever be able to answer that question.

irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
#25 - score is 0 - 3. Going alone - if opponents lead to your void - clubs you have to use your ace and lead the left. Opponent will have the right 56% of the time. Another lead of clubs you have no answer. And one of those opponents will have another trump for a euchre. Eldest leading from 3 of an off suit to one of those aces, Pone has a good chance to trump. A euchre now yo0u are down 5 to 0. You will lose most games that way. Assist and go for 2 pts - now only down by 1.
The problem with this argument is EVEN IF one proves that going alone is -EV that still doesn't refute my claim that when down 0-3 we should go for it. When down 3 points in a race to 10 game it is theoretically possible that a -EV play could lead to more wins in the long run. That's the conundrum of euchre, and it basically makes my hypothesis unfalsifiable rendering it useless.
irishwolf wrote:
Mon Mar 11, 2019 12:17 pm
Off hand I take serious issue with 9., 19., 23., 24., 25., 28. There are others but quick review these stand out.

Oh there is much more but enough for now.

Irishwolf
viewtopic.php?t=111

9) See above.

19) See above.

23) I stand by that one and feel very strongly about that one down 5-0. IMO there isn't a top player in the world that wouldn't go for it in that spot, but once again we're in unfalsifiable hypothesis territory so there's no where to go.

24) See above.

25) See above.

28) I said pass that L+1+A hand from the 2S-R1 becuz we have a euchre hand in the 2nd round. I still believe that's correct but one never knows unless it's put to the test!

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Thu Jan 06, 2022 7:52 pm

irishwolf wrote:
Mon Mar 11, 2019 1:24 pm
Question 1 - said not to go alone? However it depends on your two opponents. It said nothing about where in games you were or if either of the two opponents may be top competitors. If top competitors it is just as important to keep both at ZERO.

NO guarantee you will end up with 11, 12, or 13.
We're in agreement on this now. Excluding myself, there are only 4 players in my tournament that have a chance to win player of the year. I would call them all experts. Whenever I'm against any of them, I would go alone and hand them that ZERO. To score tournament points for player of the year in any given tournament you have to finish in the top 5. Handing one of those guys a zero effectively prevents them from scoring any player of the year points for that session. That's valuable to me. Anyone else, I'm just calling.

Moving on, when up 8-0 on our opponents deal we have 96% equity. In a tournament where the most points wins I'd rather have a 96% chance of getting 10,11,12,13 than a 100% chance of getting just 10. Another way to look at it is when this strategy backfires it's not a big deal. Let's say around 1.5% of the time we lose the game with just 8 pts and 2.5% of the time we lose the game with 9 pts. Those are still good scores. We can still win this tournament if we start with an 8. Hell I've won the tournament with a 2 in one game before. 8 pts or 9pts is still a quality loss. So that's the other side I think people aren't fully appreciating. There is very little downside to the strategy I'm proposing (except for maybe pissing your partner off but that's fun too!), but getting that 11, 12, or 13 is pretty huge.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Thu Jan 06, 2022 7:55 pm

raydog wrote:
Thu Jan 06, 2022 6:15 pm
So I have started running these hands on my program to see how it fares, and already have some cool results.

Problem #1: You're in a Euchre tournament. The structure is very simple. You play 6 games with random partners. The person with the most points in the end wins. Each game is up to ten with the max points possible being 13. You are in game one.

You are the dealer, and your team is up 6-0

The upcard is the (Card_K-C)

You hold (Card_J-S), (Card_J-C), (Card_9-D), (Card_A-C), (Card_Q-C)

My gut tells me it is a mistake to pass on a gift like a loner hand, but I can't think of a way to prove or even demonstrate which strategy is better. Giving points to your adversaries just has to be wrong (so much luck of the hands dealt, even if these opponents aren't on your radar as "experts" they aren't out of the running). I'll keep noodling on this, hopefully will think of something.
I wouldn't even waste your time on this hand. It's so rare of a situation to be almost irrelevant. I mean if you regularly play in a tournament where most points wins then yeah there's something to think about here. But this hand is still kind of a waste of time imo. Either way, I did give another explanation of it in the above post.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Thu Jan 06, 2022 8:07 pm

raydog wrote:
Thu Jan 06, 2022 6:15 pm
Problem #2: The score is 0-0. You are the dealer.

The upcard is the (Card_10-C)

You hold (Card_A-D), (Card_K-C), (Card_10-H), (Card_9-H), (Card_9-S)

My program passed this hand. So I tried calling Clubs instead, and found a slightly better EV (-0.65 vs. -0.73; 100,000 hands simulated, 61,842 made it to dealer for a decision)
Awesome.
raydog wrote:
Thu Jan 06, 2022 6:15 pm
I then tested the other scenarios Wes suggested (and a couple of my own):
a) replace 9H with 9D (create doubleton green Ace):
bid EV = -0.55; pass EV = -0.73; 61,933 of 100,000 hands
10H now discarded, so better EV than base hand; still better to call
Nice! I would not have predicted that but it makes sense as we can double lead diamonds during the hand and force out enemy trump.
raydog wrote:
Thu Jan 06, 2022 6:15 pm
b) replace AD with AS and 9S with 9D (create singleton next Ace):
bid EV = -0.67; pass EV = -0.85; 61,663 of 100,000 hands
10D discarded, but Ace in shorter suit, so slightly worse EV than base hand; still better to call
Love it! That's what I would've predicted. A little worse but still +EV vs passing.
raydog wrote:
Thu Jan 06, 2022 6:15 pm
c) replace KC with 9C (worse trump holding):
bid EV = -0.93; pass EV = -0.85; 60,021 of 100,000 hands
now better to pass!
Nice. Makes sense. It's close though where we should probably make this call to manage loner variance at least when we are up 3+ pts.
raydog wrote:
Thu Jan 06, 2022 6:15 pm
d) replace KC with QC:
bid EV = -0.79; pass EV = -0.78; EV between base case and previous case is bid (intermediate trump holding); too close to say which strategy is better
Statistical ties go to the strategy that prevents loners imo, so I would call with that. Very good stuff!
raydog wrote:
Thu Jan 06, 2022 6:15 pm
I introduced a new rule in my program for S4 bidding R1:
if have:
- no jacks
- 2 trump led by A or K
- at least 1 off-suit Ace
then bid.

I then tested this new rule for 1,000,000 random hands. It was applied 16,013 times, and the net extra points earned by S2/S4 was 1,989, for an incremental EV gain or +0.12. For those who are interested, here is the change in the # of hands won/lost by each bidder:

S4R1: 0 / 0 / 0 || +324 / +8,060 / +7,627 [so actually lose points here]
S1R2: -1,019 / -2,340 / -210 || -891 / -5380 / -2030 [keep S1 from scoring lots of points here!]
S2R2: -96 / -349 / -42 || -249 / -1,756 / -877
S3R2: -29 / -61 / -1 || -77 / -398 / -99
S4R2: 0 / -1 / 0 || -5 / -54 / -49

Overall, both teams score more points when S4 bids R1, but S4 improves by more.

As one final check I added one more rule: that S4 has 0 or 1 voids. This means that S1 does NOT have 2 trump + A-x-x. I again ran this simulation over 1,000,000 hands. It was invoked 13,223 times, with S2/S4
earning an incremental 497 more points vs. S1/S3, for a positive incremental EV of +0.04. So the restriction of not counting A-x-x was detrimental.

I ran this last simulation once again, for 1,000,000, and got the same result. So I tested the following hand: A-10-9D + KC + 9S, 10C turned. I tested this over 100,000 hands (61,636 made it dealer for bidding).
bid C: EV = -0.27
pass: EV = -0.70

So it would seem that even being two-suited (i.e, having a tripleton Ace), it's better to bid. In fact, it may even make more of a difference than having a singleton or doubleton Ace (I only tested this one example, but that's what my overall results seem to show).

So in the end, my new rule for S4R1 bidding is to BID with:
- no jacks
- 2 trump led by A or K
- at least 1 off-suit Ace (next or green; singleton, doubleton of tripleton)

I'm working on question #3 now, with similar interesting results...
Awesome stuff man.

Question: Does your findings say that this club hand is a call (S4-R1):

(Card_K-C) (Card_9-C) (Card_A-S) (Card_10-S) (Card_9-S)

I pass that to bag Next but I've always wondered about one.

Also what about:

(Card_K-C) (Card_9-C) (Card_A-S) (Card_9-S) (Card_9-H)

I usually pass that too. Sorry to give you more work but I'm really curious about those holdings.

User avatar
LeftyK
Posts: 226
Joined: Wed Jun 10, 2020 9:45 am
Location: North Carolina

Unread post by LeftyK » Thu Jan 06, 2022 10:39 pm

it's only a game (Card_A-D)
Lefty

irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Fri Jan 07, 2022 1:04 pm

Ray,

Can't wait until you get to the one with both Bowers that sugg

I am interested in Hands 10, 11 & 19.

#10: I have Hand #10 AS up S1 has QS 10S 9S KD QH. I disagree that this is an Order hand even though it blocks nothing but spades. I say PASS EV will be negative if you order, with a high euchre rate EV > -.40 is my estimate. Quiz has order.

And it raises a very important question: Should one always order if you can't block two or three suits? Or conversely, pass even if you have a hand that has a + EV but you can block two or three suits? This is the acid test and essence of Euchre! Know when to say when and your win percentage will go up!

#11: I have Hand #11 10H up S3 has JD KH QH KC QC. ORDER This is a strong hand with a + EV. But I wouldn't go alone just too many key cards have to with S1 or buried. Order, doubtful S1 can do better! My philosophy, if you have a positive EV, go for it!

#19: Both bowers, suggests you lead a Bower. I disagree unless you have An ace or two.

If this is a QUIZ it suggests to me, IMO, definitive strategy. I think there should have a DISCLAIMER, that should say: Player be aware, this is ONE PERSON's ASSESSMENT AND ADVICE.

There are numerous approaches to strategy of how to play a hand and game, etc. etc. And we can always agree to disagree on the situation and play.

There is always a defense to any strategy!

We shall wait to see you simulator results suggests.

Overall, the Quiz was/is Great as it sparks debate and conversation. So good job Wes.

IRISHWOLF

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Fri Jan 07, 2022 2:20 pm

I ran the scenarios you suggested, Wes - not too much effort, and I was curious as well.

Just to be explicit, I used 9C as the turn card, and added the 9H to dealer's hand, so after calling C and discarding the 9H they ended up with the hands you showed.

S4R1, A-xS (doubleton next Ace); 100,000 hands, ~60,000 make it to S4 for a decision:
bid: EV = -0.69 / -0.69
pass: EV = -0.65 / -0.64
So slightly better to pass. But I note that this is the absolute worst hand that passes the criteria, so in the interest of minimizing complexity I'm going to let the new rule stand.

S4R1, A-x-xS (tripleton next Ace); 100,000 hands, ~60,000 make it to S4 for a decision:
bid: EV = -0.29 / -0.28 [again, I ran it twice]
pass: EV = -0.37 / -0.39
So here it is better (less damage) to bid, as per the rule. I presume that the best asset in S4's hand are the voids in both green suits, and the AS only occasionally a late trick (when trump have been largely depleted) or serves tp draw out opponents' trump.
As for being a good bagging hand, I counted S1R2 bids: 25,100 times a red suit (a "green" suit) and 18,348 spades ("next"). Perhaps against more aggressive opponents, or against "Hoyle-ists", S1 would make more unfavorable next calls, R2, but it would have to be A LOT to make passing more favorable.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Sat Jan 08, 2022 2:15 am

raydog wrote:
Fri Jan 07, 2022 2:20 pm
I ran the scenarios you suggested, Wes - not too much effort, and I was curious as well.

Just to be explicit, I used 9C as the turn card, and added the 9H to dealer's hand, so after calling C and discarding the 9H they ended up with the hands you showed.

S4R1, A-xS (doubleton next Ace); 100,000 hands, ~60,000 make it to S4 for a decision:
bid: EV = -0.69 / -0.69
pass: EV = -0.65 / -0.64
So slightly better to pass. But I note that this is the absolute worst hand that passes the criteria, so in the interest of minimizing complexity I'm going to let the new rule stand.

I agree. Let the rule stand. This is still close enough to a tie and ties go to the caller cuz calling stops enemy loners. Also I suspect it's ok to burn a little EV against weaker opponents when the payoff is controlling some loner variance and given your presumed skill level, your opposing team will be the weaker team most of the time. I cant prove that. All I can say is my whole game revolves around this notion, I get euchred more than anyone in my tournament by far and I willingly take more -EV hits than anyone I know and I just wrapped up my 2nd Las Vegas euchre championship (2019 & 2021). I'm also very happy with with my results on the app. This idea may remain just a working hypothesis forever but I'm ok with that.

If I am onto something here it may even suggest that calling with just the Tc9c of trump + A from this configuration can be justified since this call is still a very small -EV hit.

raydog wrote:
Fri Jan 07, 2022 2:20 pm
S4R1, A-x-xS (tripleton next Ace); 100,000 hands, ~60,000 make it to S4 for a decision:
bid: EV = -0.29 / -0.28 [again, I ran it twice]
pass: EV = -0.37 / -0.39
So here it is better (less damage) to bid, as per the rule. I presume that the best asset in S4's hand are the voids in both green suits, and the AS only occasionally a late trick (when trump have been largely depleted) or serves tp draw out opponents' trump.
As for being a good bagging hand, I counted S1R2 bids: 25,100 times a red suit (a "green" suit) and 18,348 spades ("next"). Perhaps against more aggressive opponents, or against "Hoyle-ists", S1 would make more unfavorable next calls, R2, but it would have to be A LOT to make passing more favorable.


This is pure gold man. And since approx 95% of the euchre population doesn't even call next we can feel really good about those numbers! God I love getting better at this game! Good work Ray

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Sat Jan 08, 2022 2:53 am

irishwolf wrote:
Fri Jan 07, 2022 1:04 pm
#10: I have Hand #10 AS up S1 has QS 10S 9S KD QH. I disagree that this is an Order hand even though it blocks nothing but spades. I say PASS EV will be negative if you order, with a high euchre rate EV > -.40 is my estimate. Quiz has order.
Wolf we already tested this hand in that one thread--or at least I did-- where I was trying to get Edward to see that you gotta call with 3 in that spot when you have no 2nd rd hand. It was a call at a greater than 95 confidence interval. I essentially proved calling with any 3 trump vs a non bower upcard is the correct strategy from S1-R1 when we have no where to go or no defense in the 2nd rd. There is a catch tho. I only tested a first rd call vs pass-call next. I never tested it against pass-pass.
irishwolf wrote:
Fri Jan 07, 2022 1:04 pm
And it raises a very important question: Should one always order if you can't block two or three suits? Or conversely, pass even if you have a hand that has a + EV but you can block two or three suits? This is the acid test and essence of Euchre! Know when to say when and your win percentage will go up!
That's my general approach from S1-R2 and S2-R2. There are a few "exception to the rule" hands tho. One example: score is 0-0, my partner turns down the Kc. It's passed to me in the 2 seat and I have:

(Card_J-S) (Card_A-D) (Card_9-D) (Card_A-H) (Card_9-H)

I think passing is best with that configuration despite the fact we only block 1 out of 3 remaining suits.
irishwolf wrote:
Fri Jan 07, 2022 1:04 pm
#11: I have Hand #11 10H up S3 has JD KH QH KC QC. ORDER This is a strong hand with a + EV. But I wouldn't go alone just too many key cards have to with S1 or buried. Order, doubtful S1 can do better! My philosophy, if you have a positive EV, go for it!
At a non-close out score I think passing is best but I am not very confident in that. Wouldn't be surprised if I am wrong on this. In fact I think there's a great chance I'm wrong those times I have an amateur P who doesn't play the 2nd rd well which means I could be wrong up to 95%+ of the time!! I would be very interested in Rays results on this one.
irishwolf wrote:
Fri Jan 07, 2022 1:04 pm
If this is a QUIZ it suggests to me, IMO, definitive strategy. I think there should have a DISCLAIMER, that should say: Player be aware, this is ONE PERSON's ASSESSMENT AND ADVICE.
Relax bro. This quiz is made by a living legend!
irishwolf wrote:
Fri Jan 07, 2022 1:04 pm
There are numerous approaches to strategy of how to play a hand and game, etc. etc. And we can always agree to disagree on the situation and play.
I strongly disagree with the above in spirit. I hate the "agree to disagree stuff" and the "there are numerous approaches" stuff. I mean sure once in a blue moon you'll have two or perhaps even three strategy paths with equal EVs but the vast majority of time there's going to be a clear right answer/best strategy for any particular spot. We just gotta find it.
irishwolf wrote:
Fri Jan 07, 2022 1:04 pm
There is always a defense to any strategy!
True. We all have arguments for why we do what we do. But as you already know actually putting a hand to the test can destroy the very best arguments because the very best arguments are still inherently weak without statistical backing. As I always say, math is the ultimate answer.
irishwolf wrote:
Fri Jan 07, 2022 1:04 pm
Overall, the Quiz was/is Great as it sparks debate and conversation. So good job Wes.

IRISHWOLF
Yep and that was the whole point. Spark debate and make people better including myself.

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Sun Jan 09, 2022 5:59 pm

Question #3:The score is 3-3. You're in seat 1, and the dealer turned down the (Card_9-S)

You have (Card_J-C), (Card_A-D), (Card_K-D), (Card_Q-H), (Card_J-H)

Bid or pass?

My program would bid this hand, but when I compared to passing I discovered that passing was indeed the better play. But under what exact circumstances? What should be the new rule I introduce into my program?

I first identified S1 hands which have all suits blocked, R2 (not too hard). I tried the simplest rule of just passing these hands (which my program almost always bids currently), but that gave me unfavorable results - a worse EV. So I needed more precise criteria.

I ended up creating thousands of random hands, grabbing 20-30 of them which had all suits blocked S1R2, comparing bidding vs. passing for all those hands and figuring out what characteristics made a hand still worth bidding (which turned out to be a minority of hands).

What I concluded was this: (all suits blocked for all the following hands)
- if have 3 trump in next suit, BID
- if have 2 trump including R in next suit, BID
- if have 3 trump including R in green suit, BID
- else, pass

I then tested this rule for 1,000,000 random hands. Of those, 1,747 were instances where the bidding reached S1R2 and that player had all suits blocked. Applying the new rule in those instances resulted in a +241 point differential for S1/S3 (more points for them or less points for S2/S4), or an EV boost of +0.14.

I have now incorporated this rule in my program.

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Sun Jan 09, 2022 8:50 pm

Question #4:
Your team is up 7-0. Your partner, the dealer turned down the (Card_9-H)

First seat passes and it's on you.

You hold (Card_K-C), (Card_10-D), (Card_9-D), (Card_Q-H), (Card_10-H)

Simulating this hand is problematic, because out of 100,000 random hands (just S2 cards and the turn card fixed), only 9,414 of them made it around to S2R2 for a decision. That's less than 10%, and invites all sorts of speculation as to which subset of hands would really be passed this far.

With that caveat, when I look at those 9.4% of hands and compare bidding C with passing, I find the following:

bid: [0, .03, .35, 0, .62, 0] EV = -0.83
euchred 62% of the time, sweep 3% of the time, 1 pt for the rest
pass: [.004, .146, .07, .595, .15, .035] EV = -0.67
a better EV; score a pt far less often, but also give up 2 pts. far less often; and the opponents only score 4 pts 3.5% of the time.

Is that 3.5% successful loners by S3 accurate? I believe it is. Analyzing totally random deals, I find that S1R1 scores the highest % of 4 pt games (15%), followed by S4R1 (12.5%), then S1R2 (7.5%), and so on...down to S3R2 at 2.7%. The overall percentage of successful loners comes in at 9.3% [certain seats get a greater weight because they bid more]. So for S3R2 to score 3.5% successful loners is indeed a 30% increase, but it remains quite small, and nothing to be overly worried about.

Furthermore, when I take those two point probability distributions above and use them to compare the chances of winning the game to 10 pts., I arrive at a 91.5% chance when bidding and a 91.8% chance when passing.

In my opinion, you have a better chance of winning the game by passing this garbage hand, and not being unduly afraid of your opponents scoring 4 pts.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Sun Jan 09, 2022 9:55 pm

raydog wrote:
Sun Jan 09, 2022 8:50 pm
Question #4:
Your team is up 7-0. Your partner, the dealer turned down the (Card_9-H)

First seat passes and it's on you.

You hold (Card_K-C), (Card_10-D), (Card_9-D), (Card_Q-H), (Card_10-H)

Simulating this hand is problematic, because out of 100,000 random hands (just S2 cards and the turn card fixed), only 9,414 of them made it around to S2R2 for a decision. That's less than 10%, and invites all sorts of speculation as to which subset of hands would really be passed this far.

With that caveat, when I look at those 9.4% of hands and compare bidding C with passing, I find the following:

bid: [0, .03, .35, 0, .62, 0] EV = -0.83
euchred 62% of the time, sweep 3% of the time, 1 pt for the rest
pass: [.004, .146, .07, .595, .15, .035] EV = -0.67
a better EV; score a pt far less often, but also give up 2 pts. far less often; and the opponents only score 4 pts 3.5% of the time.

Is that 3.5% successful loners by S3 accurate? I believe it is. Analyzing totally random deals, I find that S1R1 scores the highest % of 4 pt games (15%), followed by S4R1 (12.5%), then S1R2 (7.5%), and so on...down to S3R2 at 2.7%. The overall percentage of successful loners comes in at 9.3% [certain seats get a greater weight because they bid more]. So for S3R2 to score 3.5% successful loners is indeed a 30% increase, but it remains quite small, and nothing to be overly worried about.

Furthermore, when I take those two point probability distributions above and use them to compare the chances of winning the game to 10 pts., I arrive at a 91.5% chance when bidding and a 91.8% chance when passing.

In my opinion, you have a better chance of winning the game by passing this garbage hand, and not being unduly afraid of your opponents scoring 4 pts.
Interesting to note, I would no longer play this hand the way I recommended it. Up 7-0, blocking nothing with no 2nd rd hand I would call hearts with 2 trump in the first round. This is one of those spots where S2 should want to make sure the 2nd rd never happens.

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Mon Jan 10, 2022 9:38 am

Question #5:
The score is 5-5. You're in the 2 seat and the dealer upcard is the (Card_9-D)

You have (Card_J-D), (Card_10-D), (Card_10-C), (Card_9-S), (Card_9-H)

Clearly need to bid here. Based on 100,000 hands, here are the EVs:
bid: EV = +0.50 [euchre rate = 22%]
pass: EV = -0.13

I think this hand looks deceptively weak, and you need to realize that:
- partner in S4 has a 91% chance of having 2+ trump, giving you the majority of trump;
- partner in S4 has a 65% chance of having at least 1 off-suit Ace;
- partner in S4 has a void;
- you have a sure trick with the R.

Just so many ways to get the 3 necessary tricks, even if it doesn't appear so at first glance.

Question #6:
The score is 0-0. You're the dealer.

The upcard is the (Card_9-H)

You hold (Card_J-H), (Card_A-C), (Card_10-C), (Card_A-D), (Card_9-S)

Here I ran 10,000 hands and found the following EVs:
bid alone: EV = +0.93
bid w/p: EV = +0.85
As Wes suggested, you do get euchred a bit more often when going alone, but the extra points earned when successful more than compensate for that. Not much more to say.

irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Jan 10, 2022 10:37 am

Ray,
Results agreed for #5. But it should be noted:

I don't agree with your S4 has 91% 2+ trumps with 4 unknown trumps with 18 unknown cards. 4c1 = 46.7%, 4c0 = 23.4% (only the upcard & no off suit ace is when get euchre), 4c2 = 25%. Even without S2 ordering, the majority of the time the Dealer will pick it up anyway (2 + an Ace). S2 hand is rated as 1.5 tricks, he needs help big time!

So that number should ~76.6% (100 - 23.4). You said 91%, where did that come from(91%)?

You said, "I think this hand looks deceptively weak, and you need to realize that:
- partner in S4 has a 91% chance of having 2+ trump, giving you the majority of trump;
- partner in S4 has a 65% chance of having at least 1 off-suit Ace;
- partner in S4 has a void;"

IRISH

irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Jan 10, 2022 10:40 am

Ray,

There are different ways to Defensively to play this hand. If S1 leads trump to the first trick (Hand #5) vs leading an Ace. I think the euchre rate will be higher. Not sure how your program does that? If not it would be interesting to see how that turns out.

Irish

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Jan 10, 2022 2:23 pm

raydog wrote:
Sun Jan 09, 2022 5:59 pm
What I concluded was this: (all suits blocked for all the following hands)
- if have 3 trump in next suit, BID
I think I like this rule. Assume score is 0-0 and dealer turns down the (Card_9-S)

That would mean we should call next with:

(Card_Q-C) (Card_10-C) (Card_9-C) (Card_K-H) (Card_J-D)

Does your simulator support that?

A real mindf**k hand would be:

(Card_Q-C) (Card_10-C) (Card_9-C) (Card_J-D) (Card_J-H)

I would pass that hand. This hand is categorically different from the above hand as we now have 2 guaranteed tricks against a reverse next call making this hand a pretty powerful bagging hand. Btw I'm always assuming an STD format as that is the most popular form of the game today. Not sure if that matters. Anyways I would interpret your rule as: if I only have 1 trick in reverse Next I'm calling Next when I have 3 trump.
raydog wrote:
Sun Jan 09, 2022 5:59 pm
- if have 2 trump including R in next suit, BID
This is a very interesting claim. Again let's assume score is 0-0 and (Card_9-S) was turned down.

I have (Card_J-C) (Card_9-C) (Card_J-H) (Card_K-D) (Card_10-S)

I would pass. Is it really better to call clubs?
raydog wrote:
Sun Jan 09, 2022 5:59 pm
- if have 3 trump including R in green suit, BID
I agree with this rule.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Jan 10, 2022 4:44 pm

raydog wrote:
Fri Jan 07, 2022 2:20 pm
I ran the scenarios you suggested, Wes - not too much effort, and I was curious as well.

Just to be explicit, I used 9C as the turn card, and added the 9H to dealer's hand, so after calling C and discarding the 9H they ended up with the hands you showed.

S4R1, A-xS (doubleton next Ace); 100,000 hands, ~60,000 make it to S4 for a decision:
bid: EV = -0.69 / -0.69
pass: EV = -0.65 / -0.64
So slightly better to pass. But I note that this is the absolute worst hand that passes the criteria, so in the interest of minimizing complexity I'm going to let the new rule stand.

S4R1, A-x-xS (tripleton next Ace); 100,000 hands, ~60,000 make it to S4 for a decision:
bid: EV = -0.29 / -0.28 [again, I ran it twice]
pass: EV = -0.37 / -0.39
So here it is better (less damage) to bid, as per the rule. I presume that the best asset in S4's hand are the voids in both green suits, and the AS only occasionally a late trick (when trump have been largely depleted) or serves tp draw out opponents' trump.
As for being a good bagging hand, I counted S1R2 bids: 25,100 times a red suit (a "green" suit) and 18,348 spades ("next"). Perhaps against more aggressive opponents, or against "Hoyle-ists", S1 would make more unfavorable next calls, R2, but it would have to be A LOT to make passing more favorable.


Ray, your work has established that

(Card_K-C) (Card_9-C) (Card_A-S) (Card_10-S) (Card_9-S)

Is a +EV dealer club call (by +EV I mean compared to passing).

Another club call I wonder about is:

(Card_A-C) (Card_9-C) (Card_A-S) (Card_10-S) (Card_9-S)

Even tho this hand is better than the above hand I think there's a chance passing may be better. The idea is the cost of passing will be lower with this holding as we have a euchre hand in Next and we have two off aces to fight a red call. So question is will this dynamic make the cost of passing low enough for passing to be +EV vs calling.

Ray, when you get the chance can you test this one out.

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Mon Jan 10, 2022 10:11 pm

I'll address Irish first, hand #5.

1) I made a math error, indeed 77% chance that dealer has 2+ trump.
2) My program does consider having S1 lead trump when S2 calls trump, R1 (hoping to void S4 of trump).
Will do so if:
- has both bowers (not the case here)
- has 3+ trump AND R not turned (will lead low trump)
- has 2 trump AND 1 or 2 voids AND no off-suit Ace (will lead low trump, unless R turned)

So trump is sometimes led, but not automatically.
_______________________________________

Now Wes, hand #3.

With (Card_Q-C) (Card_10-C) (Card_9-C) (Card_K-H) (Card_J-D), 9S turned down, approx. 10-11% of hands make it to R2. In this case, all suits are not stopped (C not stopped) so I didn't actually look at these sorts of hands (but do want to at some point). Anyway, my program currently passes this hand, and rates H as the better bid if forced to bid, but I found that bidding C is actually the best course of action:
bid C: EV = +0.05 (9C led)
pass: EV = -0.13
bid H: EV = -0.39 (9C led)

So I have a little work to do to determine how to better bid these sorts of hands.

With (Card_Q-C) (Card_10-C) (Card_9-C) (Card_J-D) (Card_J-H), 9S turned down, once again C is not stopped. My program currently passes, but I found a better result bidding H (or D):
bid H: EV = +0.42 (9C led)
pass: EV = +0.31

Finally, with (Card_J-C) (Card_9-C) (Card_J-H) (Card_K-D) (Card_10-S), 9S turned down, about 35% of hands make it R2. I find it a toss-up between bidding and passing - EV = -0.07 either way.

Then you asked about another hand: (Card_A-C) (Card_9-C) (Card_A-S) (Card_10-S) (Card_9-S)
I set this up having 9C as the turn card, which dealer picks up, discarding the 9D. My program would currently pass this hand.
I tested 100,000 hands; 59,372 made it to S4 for a decision.
bid: EV = +0.12
pass: EV = +0.01

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Tue Jan 11, 2022 3:21 am

raydog wrote:
Mon Jan 10, 2022 10:11 pm
Now Wes, hand #3.

With (Card_Q-C) (Card_10-C) (Card_9-C) (Card_K-H) (Card_J-D), 9S turned down, approx. 10-11% of hands make it to R2. In this case, all suits are not stopped (C not stopped) so I didn't actually look at these sorts of hands (but do want to at some point). Anyway, my program currently passes this hand, and rates H as the better bid if forced to bid, but I found that bidding C is actually the best course of action:
bid C: EV = +0.05 (9C led)
pass: EV = -0.13
bid H: EV = -0.39 (9C led)
Very nice! I will call clubs from this configuration from now on.
raydog wrote:
Mon Jan 10, 2022 10:11 pm
With (Card_Q-C) (Card_10-C) (Card_9-C) (Card_J-D) (Card_J-H), 9S turned down, once again C is not stopped. My program currently passes, but I found a better result bidding H (or D):
bid H: EV = +0.42 (9C led)
pass: EV = +0.31
I can definitely see that being true as having both bowers is still a very good holding. Good stuff !
raydog wrote:
Mon Jan 10, 2022 10:11 pm
Finally, with (Card_J-C) (Card_9-C) (Card_J-H) (Card_K-D) (Card_10-S), 9S turned down, about 35% of hands make it R2. I find it a toss-up between bidding and passing - EV = -0.07 either way.
The fact that this is a toss up actually supports your rule to call with R+1+0 in Next even if we have a stopper hand. The above hand is pretty much the worst R+1 from this configuration and yet calling with it still wasn't worse than passing. Awesome stuff man. I have two more hand types from this configuration that I think is worth testing (in both hands assume score is 0-0 and dealer turned down the 9s):

(Card_A-C) (Card_10-C) (Card_9-C) (Card_K-D) (Card_J-H)

I think there's a decent shot passing beats out calling Next here given that we have a euchre hand vs a reverse next call.

Other hand:

(Card_J-C) (Card_9-C) (Card_A-S) (Card_K-D) (Card_J-H)

This hand is a euchre hand in all suits. Is it better to pass this hand or still call next?
raydog wrote:
Mon Jan 10, 2022 10:11 pm
Then you asked about another hand: (Card_A-C) (Card_9-C) (Card_A-S) (Card_10-S) (Card_9-S)
I set this up having 9C as the turn card, which dealer picks up, discarding the 9D. My program would currently pass this hand.
I tested 100,000 hands; 59,372 made it to S4 for a decision.
bid: EV = +0.12
pass: EV = +0.01
Wow. This is great stuff man. I will start calling with this hand.

Another hand I'm curious about--this is getting real marginal but I still wonder--Say we end up with this club call from the dealer spot:

(Card_A-C) (Card_K-C) (Card_K-D) (Card_Q-D) (Card_10-D)

I wonder if there's a chance calling beats out passing. We still have two voids and thus a chance to take 2 tricks and the cost of passing should be pretty high as we block no suits and have no Next cards.

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Tue Jan 11, 2022 12:16 pm

(Card_A-C) (Card_10-C) (Card_9-C) (Card_K-D) (Card_J-H), 9S turned, score 0-0.

I tested 100,000 hands; only 12,602 made it to R2.

bid C (lead 9C): EV = +0.27
pass: EV = +0.31
Not very definitive, especially given the low percentage of hands that make to R2.

(Card_J-C) (Card_9-C) (Card_A-S) (Card_K-D) (Card_J-H), 9S turned, score 0-0.

I again tested 100,000 hands, 36,389 made it to R2.

bid C (lead JC, followed by AS): EV = +0.42
pass: EV = +0.33
bid C (lead JH): EV = +0.21
Decent EV if pass, but still better if bid - but need to lead correctly!

(Card_A-C) (Card_K-C) (Card_K-D) (Card_Q-D) (Card_10-D) in dealer's hand (picked up KC, discarded 9H).

10,000 hands (6,421 make it to S4 for a decision)

bid: EV = -0.43
pass: EV = -0.61
You are correct, better to bid, simply because you get so slammed if you pass (S1R2 calls trump 75% of the time, for an EV of +0.87!)

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Tue Jan 11, 2022 12:41 pm

Question #7:
Your team is up 4-3. You're in the 2 seat and the dealer upcard is the (Card_K-H)

You hold (Card_A-H), (Card_J-H), (Card_A-D), (Card_J-S), (Card_J-C)

bid H w/p, R1: EV = +1.05
pass, then bid S alone, R2: EV = +1.34 [includes hands bid by S3R1, S4R1 and S1R2]
So better to pass and bid alone, R2. Only 20% of hands make it back around for S2 to bid, R2, but those other 80% of hands include lots of euchres of S3 and S1, and great support of S4 when partner bids.
Just confirming Wes' answer and adding a bit of detail about what actually happens.

Question #8:
Your team is up 6-0. You're in the 1 seat. The dealer just turned down the 9s.

You have (Card_J-S), (Card_9-D), (Card_10-D), (Card_10-H), (Card_9-H)

I did a simulation over 100,000 hands, just shy of 40% make to R2.

bid C: [0, .68, 0, .30, .02, 0] EV = -1.01
pass: [.05, .19, .50, .13, .12, .001] EV = -0.65

So if you bid you get euchred 68% of the time, sweep 2% of the time and score a point 30% of the time.
If you pass, opponents score 4 pts. 5% of the time and score 1 pt. 50% of the time, while you eke out a point (or more) 26% of the time.

I plugged these point probability distributions into my "who wins the game" program, and found that when bidding you can expect to lose 13.8% of games, but when passing you can expect to lose 12.6% of games. I think the fear of opponents scoring 4 pts on a loner here is irrationally high, and it is simply better - statistically - to pass.

Put another way: if you bid, the MOST LIKELY outcome is that you give your opponents 2 pts. (68% of the time), and you have virtually no chance of scoring more than a point (2%). If you pass, the MOST LIKELY outcome is that you give your opponents 1 pt. (50% of the time), and you retain a 13% chance of getting 2 or even 4 pts - more than 5 times better than if you bid. And again, the odds of the opponents scoring 4 pts - the only downside - is only 5%.

The only other factor here, as I see it, is if you believe in streaks, or you think your opponents will be reinvigorated to play MUCH better (and you perhaps play much worse) after they score 4 pts. Some sort of momentum theory. But I don't buy it; I pass this hand.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Tue Jan 11, 2022 12:58 pm

raydog wrote:
Tue Jan 11, 2022 12:16 pm
(Card_A-C) (Card_10-C) (Card_9-C) (Card_K-D) (Card_J-H), 9S turned, score 0-0.

I tested 100,000 hands; only 12,602 made it to R2.

bid C (lead 9C): EV = +0.27
pass: EV = +0.31
Not very definitive, especially given the low percentage of hands that make to R2.
Good stuff. I agree that your rule of "don't pass 3 trump in next even when you have a stopper hand" is a good rule. This hand barely contradicts that. And when we factor in that amateurs are typically playing in S2-S4 meaning they will be passing biddable hands unintentionally bagging next calls, that could easily sway the numbers to a clear call. Either way this is one of those edge hands.
raydog wrote:
Tue Jan 11, 2022 12:16 pm
(Card_J-C) (Card_9-C) (Card_A-S) (Card_K-D) (Card_J-H), 9S turned, score 0-0.

I again tested 100,000 hands, 36,389 made it to R2.

bid C (lead JC, followed by AS): EV = +0.42
pass: EV = +0.33
bid C (lead JH): EV = +0.21
Decent EV if pass, but still better if bid - but need to lead correctly!
Awesome. Makes sense. R+1+A is a good Next call even with a euchre hand. Really good stuff man. And yep, definitely gotta lead that Right followed by the Ace.
raydog wrote:
Tue Jan 11, 2022 12:16 pm
(Card_A-C) (Card_K-C) (Card_K-D) (Card_Q-D) (Card_10-D) in dealer's hand (picked up KC, discarded 9H).

10,000 hands (6,421 make it to S4 for a decision)

bid: EV = -0.43
pass: EV = -0.61
You are correct, better to bid, simply because you get so slammed if you pass (S1R2 calls trump 75% of the time, for an EV of +0.87!)
Wow!! This is a big discovery man! I always suspected this was a call but wasn't sure. Now the question is are all hands from this configuration a call. We probably need to test these edge hands to nail this down (dealer club calls):

(Card_A-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D)

(Card_K-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D)

(Card_Q-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D)

(Card_10-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D)

We basically need to find that point of indifference btw call and pass from this configuration. The hand you already tested was the best possible hand from this configuration. I chose JdTd9d as our worst case call scenario for our offsuit as that is actually worse than QdTd9d since the cost of passing will be lower with the former given that we then block 1 suit.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Tue Jan 11, 2022 1:32 pm

raydog wrote:
Tue Jan 11, 2022 12:41 pm
Question #7:
Your team is up 4-3. You're in the 2 seat and the dealer upcard is the (Card_K-H)

You hold (Card_A-H), (Card_J-H), (Card_A-D), (Card_J-S), (Card_J-C)

bid H w/p, R1: EV = +1.05
pass, then bid S alone, R2: EV = +1.34 [includes hands bid by S3R1, S4R1 and S1R2]
So better to pass and bid alone, R2. Only 20% of hands make it back around for S2 to bid, R2, but those other 80% of hands include lots of euchres of S3 and S1, and great support of S4 when partner bids.
Just confirming Wes' answer and adding a bit of detail about what actually happens.
Nice! Obviously not surprised by the results. Unfortunately this spot is so rare it's almost strategically irrelevant but I knew it would be a great quiz question as it was pretty predictable that most people would get it wrong. :)
raydog wrote:
Tue Jan 11, 2022 12:41 pm
Question #8:
Your team is up 6-0. You're in the 1 seat. The dealer just turned down the 9s.

You have (Card_J-S), (Card_9-D), (Card_10-D), (Card_10-H), (Card_9-H)

I did a simulation over 100,000 hands, just shy of 40% make to R2.

bid C: [0, .68, 0, .30, .02, 0] EV = -1.01
pass: [.05, .19, .50, .13, .12, .001] EV = -0.65

So if you bid you get euchred 68% of the time, sweep 2% of the time and score a point 30% of the time.
If you pass, opponents score 4 pts. 5% of the time and score 1 pt. 50% of the time, while you eke out a point (or more) 26% of the time.

I plugged these point probability distributions into my "who wins the game" program, and found that when bidding you can expect to lose 13.8% of games, but when passing you can expect to lose 12.6% of games. I think the fear of opponents scoring 4 pts on a loner here is irrationally high, and it is simply better - statistically - to pass.

Put another way: if you bid, the MOST LIKELY outcome is that you give your opponents 2 pts. (68% of the time), and you have virtually no chance of scoring more than a point (2%). If you pass, the MOST LIKELY outcome is that you give your opponents 1 pt. (50% of the time), and you retain a 13% chance of getting 2 or even 4 pts - more than 5 times better than if you bid. And again, the odds of the opponents scoring 4 pts - the only downside - is only 5%.

The only other factor here, as I see it, is if you believe in streaks, or you think your opponents will be reinvigorated to play MUCH better (and you perhaps play much worse) after they score 4 pts. Some sort of momentum theory. But I don't buy it; I pass this hand.
This certainly challenges the idea that one should be willing to take a -EV hit to control some variance. If it's not worth making a -.36 EV call up 6-0 one must wonder is it ever worth taking a -EV hit to control some variance?!?! Cuz this spot is about as good as it gets to take that -EV hit. And, if one concludes it's not worth it to make that call in this spot then imo they should also conclude that it's not worth reflexively donating up 9-6/9-7 when unguarded. Because in the latter case one will be burning more EV on average.

I'm not sure what to think about all this. To be honest I'm still gonna make the call I recommended. This philosophy of controlling variance has just been too good to me. I may be a prisoner of my own success. But then again, excluding myself, the 4 experts who play in my weekly tournament all play this way. They all make these type of dubious calls to control variance. And a non-expert has never won Player of the Year going back to at least 2011 (at that point I don't know who the players are). It's hard not to think there's gotta be something to this when you have a tournament with roughly 30 people and the same 4 people keep winning the championship every year. One could play devils advocate and say maybe they are winning because of superior play in other areas, IE maybe they are winning DESPITE these dubious variance controlling calls not because of them. IDK man.

Either way if the experts are making these "bad" calls and still having predictable consistent success we can at least conclude that even in the worst light these "bad" calls are probably not strategically relevant.

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Tue Jan 11, 2022 3:24 pm

I think the best players keep winning consistently because they are best in all (other) areas, not because they are making this one, infrequent, marginally unfavorable call.

More importantly, you cannot compare this to donating from S1 when you are winning 9-6 or 9-7, because the game ends at 10. Extra points don't count [looking at win/loss stats, putting aside for the moment the award in your tournament for total points scored].

The easiest way to think about this is to consider a player in S4 (dealer) whose team is tied 9-9. It is not hard to find hands where bidding presents a worse EV than passing, BUT where bidding is what must be done, because they win AT LEAST A POINT (usually 1 pt) more often than they lose AT LEAST A POINT (most often 2 pts, from being euchred).

bid: [0, 2, 38, 0, 60,0] EV = -0.78
pass: [1, 10, 15, 56, 14, 4] EV = -0.61

If you bid, you sweep 2% of the time, earn 1 pt 38% of the time and get euchred 60% of the time. This is the last hand of the game, so you win the game 40% of the time (you almost always win 10-9, but when you lose it's by a score of 9-11. Doesn't matter.)

If you pass, you win 4 pts 1% of the time, earn 2 pts 10% of the time, and earn 1 pt 15% of the time, mostly on partner's bids (euchring the opponents also gives you 2 pts). The opponents earn 1, 2 or 4 pts,
56%, 14% and 4% of the time, respectively. This is the better EV play, BUT you only win the game 26% of the time. The difference is that when you win, you often win 11-9, and when you lose, you often lose 9-10. But that does't matter!

Similarly, there are a small fraction of very bad hands (I haven't calculated this, but I would guess less than 5% of total hands) which S1 should donate (bid with the expectation of losing) if the score is 9-6 or 9-7 in their favor, because it actually gives them a BETTER chance of winning the game, even though it presents a WORSE EV than passing. Because losing 4 pts to the opponents ends the game there and then, AND wins of 2 or 4 pts are irrelevant - they boost the calculated EV but do not make winning the game any more likely (S1/S3 just need 1 pt.)

In conclusion, EV may become an inaccurate metric when superfluous points are at stake. When either team has more than 6 points, and the closer to 10 points they are, the more care must be given to interpreting - or even caring about - EV. At a score of 6-0, EV is perfectly relevant.

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Tue Jan 11, 2022 4:26 pm

Wes:

(Card_10-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D): EVcall = =0.99; EVpass == -0.84

(Card_Q-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D): EVcall = -0.87; EVpass = -0.86

(Card_K-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D): EVcall = -0.65; EVpass = -0.75

(Card_A-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D): EVcall = -0.48; EVpass = -0.48

and I tried:
(Card_Q-C) (Card_9-C) (Card_Q-D) (Card_10-D) (Card_9-D) : EVcall = -0.86; EVpass = -1.12

All tested for 10,000 hands, 56-60% make it to S4 (% increases as the hands get better)

EVcall has the expected progression towards better numbers (less negative) as the trump cards get better.
EVpass doesn't change for the first two, then improves if the KC or AC is held as an off-suit stopper. And of course shifts towards notably worse if the JD is replaced with the QD.

So you unfortunately have two different non-linear functions affecting the EVcall and EVpass functions - not easy to come up with a simple heuristic.

irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Tue Jan 11, 2022 7:47 pm

Exactly my point, 7 to 7, with The Devil in the Details post:

"AND wins of 2 or 4 pts are irrelevant - they boost the calculated EV but do not make winning the game any more likely (S1/S3 just need 1 pt.)

In conclusion, EV may become an inaccurate metric when superfluous points are at stake. When either team has more than 6 points, and the closer to 10 points they are, the more care must be given to interpreting - or even caring about - EV. At a score of 6-0, EV is perfectly relevant."

IRISH

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Wed Jan 12, 2022 3:12 am

raydog wrote:
Tue Jan 11, 2022 3:24 pm
I think the best players keep winning consistently because they are best in all (other) areas, not because they are making this one, infrequent, marginally unfavorable call.
Perhaps. No doubt this could be true. I would only add that this type of call is not really infrequent. I'm making these type of dubious calls almost every game. And I know the other experts are too, probably not as often as me tho. There's a reason I have by far the highest euchre rate in my tournament and probably the 2nd highest euchre rate on my app. If my results weren't so favorable I'd worry bout this. I'm still concerned in the academic sense, like in the sense of a mystery I cannot solve, but whatever. As of now there's certain strategies I cannot change because they've just been too damn successful, even if there's some evidence against them.
raydog wrote:
Tue Jan 11, 2022 3:24 pm
More importantly, you cannot compare this to donating from S1 when you are winning 9-6 or 9-7, because the game ends at 10. Extra points don't count [looking at win/loss stats, putting aside for the moment the award in your tournament for total points scored].

The easiest way to think about this is to consider a player in S4 (dealer) whose team is tied 9-9. It is not hard to find hands where bidding presents a worse EV than passing, BUT where bidding is what must be done, because they win AT LEAST A POINT (usually 1 pt) more often than they lose AT LEAST A POINT (most often 2 pts, from being euchred).

bid: [0, 2, 38, 0, 60,0] EV = -0.78
pass: [1, 10, 15, 56, 14, 4] EV = -0.61

If you bid, you sweep 2% of the time, earn 1 pt 38% of the time and get euchred 60% of the time. This is the last hand of the game, so you win the game 40% of the time (you almost always win 10-9, but when you lose it's by a score of 9-11. Doesn't matter.)

If you pass, you win 4 pts 1% of the time, earn 2 pts 10% of the time, and earn 1 pt 15% of the time, mostly on partner's bids (euchring the opponents also gives you 2 pts). The opponents earn 1, 2 or 4 pts,
56%, 14% and 4% of the time, respectively. This is the better EV play, BUT you only win the game 26% of the time. The difference is that when you win, you often win 11-9, and when you lose, you often lose 9-10. But that does't matter!

Similarly, there are a small fraction of very bad hands (I haven't calculated this, but I would guess less than 5% of total hands) which S1 should donate (bid with the expectation of losing) if the score is 9-6 or 9-7 in their favor, because it actually gives them a BETTER chance of winning the game, even though it presents a WORSE EV than passing. Because losing 4 pts to the opponents ends the game there and then, AND wins of 2 or 4 pts are irrelevant - they boost the calculated EV but do not make winning the game any more likely (S1/S3 just need 1 pt.)

In conclusion, EV may become an inaccurate metric when superfluous points are at stake. When either team has more than 6 points, and the closer to 10 points they are, the more care must be given to interpreting - or even caring about - EV. At a score of 6-0, EV is perfectly relevant.
Yea that's exactly what I would predict, that most donates up 9-6/9-7 are not worth it. I know the two situations are not analogous but to me it's still fairly easy to infer that if it's incorrect to make a -.36 ev call up 6-0 to control some variance, reflexively donating up 9-6/9-7 every time we're unguarded is probably not worth it for many/most hand configurations. Your guess of less than 5% hands are worth donating with lines up with my thinking here.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Wed Jan 12, 2022 3:39 am

raydog wrote:
Tue Jan 11, 2022 4:26 pm
Wes:

(Card_10-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D): EVcall = =0.99; EVpass == -0.84
Even this result blows my mind. The idea that this crappy call only has an overall -EV of -.15. This is definitely a good dealer donating candidate at the right scores.
raydog wrote:
Tue Jan 11, 2022 4:26 pm
(Card_Q-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D): EVcall = -0.87; EVpass = -0.86
There's the point of indifference. Nice!
raydog wrote:
Tue Jan 11, 2022 4:26 pm
(Card_K-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D): EVcall = -0.65; EVpass = -0.75
Awesome stuff man. Another hand to add to the calling side of the ledger.
raydog wrote:
Tue Jan 11, 2022 4:26 pm
(Card_A-C) (Card_9-C) (Card_J-D) (Card_10-D) (Card_9-D): EVcall = -0.48; EVpass = -0.48
Another point of indifference reached. Sweet. As I always say, ties go to the caller.
raydog wrote:
Tue Jan 11, 2022 4:26 pm
and I tried:
(Card_Q-C) (Card_9-C) (Card_Q-D) (Card_10-D) (Card_9-D) : EVcall = -0.86; EVpass = -1.12

All tested for 10,000 hands, 56-60% make it to S4 (% increases as the hands get better)
Wow. From the above I would predict that calling is better than passing with even this hand:

(Card_10-C) (Card_9-C) (Card_Q-D) (Card_10-D) (Card_9-D)

Am I right?
raydog wrote:
Tue Jan 11, 2022 4:26 pm
EVcall has the expected progression towards better numbers (less negative) as the trump cards get better.
EVpass doesn't change for the first two, then improves if the KC or AC is held as an off-suit stopper. And of course shifts towards notably worse if the JD is replaced with the QD.

So you unfortunately have two different non-linear functions affecting the EVcall and EVpass functions - not easy to come up with a simple heuristic.
Yea I see the "problem" but I think we can still create some kind of decent heuristic. It's looking something like this so far: Whenever we're the dealer and we have a 2 trump, 2 suited hand with no offsuit aces and no jacks plus nothing in Next, we should call.

When our outside offsuit includes a Jack--thus lowering the cost of passing a bit--then we need Kx or better in trump to call.

The only thing left to analyze are those times our offsuit is 3 Next cards instead of 3 red cards. Like are these hands still worth a call:

(Card_10-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)

(Card_Q-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)

(Card_K-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)

(Card_A-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)

(Card_K-C) (Card_9-C) (Card_J-S) (Card_10-S) (Card_9-S)

(Card_A-C) (Card_9-C) (Card_J-S) (Card_10-S) (Card_9-S)

irishwolf
Posts: 1042
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Thu Jan 13, 2022 12:36 am

The results of this would be of interest.

But I know what I would do here. And I will climb out on that limb. Calling clubs with any three, pass the first one. If S1 had tenancies to calling Next frequently, I would pass that one. The other four, I am calling clubs.

I am assuming here, Club the up card, 9C (I think) and I had to discard a low red cad. I am calling Clubs as indicated. Even with the two with JS, call clubs, because seldom will it come back to the dealer to call spades. That is the only one I might question. Main reason is, Red will be called much more than I desire it to be. Especially with a Expert at S1 who crosses suit a lot. I hate a partner who bags from the Dealer seat when he has a biddable hand. Screws S2 on calling thin!

So if Ray is doing these, see what he comes up with.

Irish

When our outside offsuit includes a Jack--thus lowering the cost of passing a bit--then we need Kx or better in trump to call.

The only thing left to analyze are those times our offsuit is 3 Next cards instead of 3 red cards. Like are these hands still worth a call:

(Card_10-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)

(Card_Q-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)

(Card_K-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)

(Card_A-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)

(Card_K-C) (Card_9-C) (Card_J-S) (Card_10-S) (Card_9-S)

(Card_A-C) (Card_9-C) (Card_J-S) (Card_10-S) (Card_9-S)
Top

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Thu Jan 13, 2022 3:55 pm

Question #9:
Your team is down 9-6. You're in the 1 seat. The dealer just turned down the (Card_9-H)

You have (Card_A-H), (Card_J-H), (Card_A-C), (Card_Q-C), (Card_A-S)

I stuck this hand in my program and it bid H alone, R1. Which is the answer Wes now agrees with, and the one I found to be the best overall.

S1R1, H alone (lead JH then AH, always): (3342 / 4313 / 2345) [+4 pts / +1 pt / euchred] EV = +1.30
S1R1, H alone (lead JH then AS if JD doesn't appear): (2127 / 6106 / 1767) EV = +1.11
The second way of playing yields fewer 4 pts and fewer euchres, but not in a 1:1 ratio, so the EV is lower. More importantly, it leads to winning the GAME 33.1% of the time as opposed to 41.8% of the time if the first way is used.

But let's say you "misclick", and the hand ends up going to R2; what do you do then?
Only 62-63% of hands will make it that far, and in those other bid hands you will lose the game about 1250 times (opponents earn at least a point). You'll euchre the opponents a fair bit, R1, but not enough to justify passing.

If you bid alone in D and lead the JH, the outcome is (332 / 3000 / 2912) EV = -0.22 [14.6% chance of winning the game]

I also compared bidding D w/partner (JH lead) and bidding C w/partner (QC lead):

bid D: EV = +0.14 (10,736 / 30,329 / 21,390) win game 15.8%
bid C: EV = +0.22 (9,208 / 34,129 / 19,118) win game16.0%

Bidding D yields more sweeps but also more euchres. In the end, this tightens the gap between odds of winning the game under the 2 scenarios (because those euchres include superfluous points for the opponents), but it is slightly better to bid clubs. Or call it a tie, as the numbers are so close.

Of course, you'll never have to make this decision because you'll bid alone, R1!

So C is actually a better bid than D!

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Thu Jan 13, 2022 5:31 pm

I assumed 9C as turn card, 9H discarded by dealer (or left in dealer's hand if passes!)

Ran 10,000 for each scenario (except one, noted below).

(Card_10-C) (Card_9-C) (Card_Q-D) (Card_10-D) (Card_9-D) [from previous series]
bid: EV = -1.01
pass: EV = -1.15
5,755 hands make it to dealer
If dealer passes, S1 calls alone more than 1/3 of the time in R2, and makes almost 1/2 of them!
Best to bid; essentially donating.

(Card_10-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)
bid: EV = -0.87
pass = -0.88
5,473 hands make it to dealer
Pretty much a toss-up here. 40% less successful loners by S1R2

(Card_Q-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)
bid: EV = -0.71
pass: EV = -0.79
5,615 hands make it to dealer
Better to bid.

(Card_K-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)
bid: EV = -0.53
pass: EV = -0.70
5,640 hands make it to dealer
Better to bid.

(Card_A-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)
bid: EV = -0.35
pass = EV = -0.32
Given the sudden improvement in the passing score, and the closeness of the EVs, I ran this again with 100,000 hands:
bid: EV = -0.36
pass EV = -0.35
56,924 hands make to dealer
Another toss-up. It seems holding a singleton Ace (likely in a green suit, albeit a short suit) helps A LOT on defense against opponents' bids.

(Card_K-C) (Card_9-C) (Card_J-S) (Card_10-S) (Card_9-S)
bid: EV = +0.47
pass: EV = -0.56
7,408 hands make it to dealer
Was there ever any doubt about this result?

(Card_A-C) (Card_9-C) (Card_J-S) (Card_10-S) (Card_9-S)
bid: EV = +0.58
pass: EV = -0.19
7,430 hands make it to dealer
Once again, no surprise here, and the AC once again helps a lot if passing.

I also checked this hand again, assuming S4 holding the KH rather than the 9H, and still discarding it if bidding:
(Card_10-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)
bid: EV = -0.87
pass: EV = -0.77
5,523 hands make it to dealer.
Here, that bit of extra strength in the hand tilts the balance towards passing, as I'm sure it would as well with the other toss-up hand (with the A-9C).

raydog
Posts: 78
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Thu Jan 13, 2022 7:19 pm

Question #10:
Your team is up 4-2. You're in the 1 seat. The dealer upcard is the (Card_A-S)

You have (Card_Q-S), (Card_10-S), (Card_9-S), (Card_K-D), (Card_Q-H)

Tested on 10,000 hands,
bid: EV = -0.66 (564 / 3,720 / 5,716) 564 sweeps; 57% euchre rate
pass: EV = -0.74 (2,700 loner calls by S2/S4, 23% success rate; 750 sweeps; 1,180 total calls by S3)

While I think there is a reasonable argument to be made, based on these results, that bidding is the better option, there are a slew of assumptions that need to be made about how well my program mimics the reality of how all these hands would be bid and played. So I'd be more confident concluding that bidding is not a bad decision - it's not going to cause any significant EV hit - but passing is not necessarily a terrible option.

I am personally going to revisit a whole subsets of hands which my program currently passes and try to identify very clear cases where bidding, while generating a negative EV, is still favorable to passing. And this will be a hand I bid. My program already does this in many cases, but I'm finding more and more "holes" in my algorithm. I remember Eric Zalas could never really accept this, and was always focussed on +EV.

The downside is that, if most players don't bid this way then my program will veer farther and farther from the actual subset of hands left in play late in R2 (and by "actual" I mean what one will encounter when playing with live players). Hopefully that won't undermine the results that come from making the "correct" bid.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Thu Jan 13, 2022 7:47 pm

raydog wrote:
Thu Jan 13, 2022 5:31 pm
(Card_K-C) (Card_9-C) (Card_J-S) (Card_10-S) (Card_9-S)
bid: EV = +0.47
pass: EV = -0.56
7,408 hands make it to dealer
Was there ever any doubt about this result?

(Card_A-C) (Card_9-C) (Card_J-S) (Card_10-S) (Card_9-S)
bid: EV = +0.58
pass: EV = -0.19
7,430 hands make it to dealer
Once again, no surprise here, and the AC once again helps a lot if passing.
Lol had a total brainfart and forgot we had 3 trump from these configurations. Sorry to waste your time on these two.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Fri Jan 14, 2022 3:00 am

raydog wrote:
Thu Jan 13, 2022 3:55 pm
Question #9:
Your team is down 9-6. You're in the 1 seat. The dealer just turned down the (Card_9-H)

You have (Card_A-H), (Card_J-H), (Card_A-C), (Card_Q-C), (Card_A-S)

I stuck this hand in my program and it bid H alone, R1. Which is the answer Wes now agrees with, and the one I found to be the best overall.

S1R1, H alone (lead JH then AH, always): (3342 / 4313 / 2345) [+4 pts / +1 pt / euchred] EV = +1.30
S1R1, H alone (lead JH then AS if JD doesn't appear): (2127 / 6106 / 1767) EV = +1.11
The second way of playing yields fewer 4 pts and fewer euchres, but not in a 1:1 ratio, so the EV is lower. More importantly, it leads to winning the GAME 33.1% of the time as opposed to 41.8% of the time if the first way is used.

But let's say you "misclick", and the hand ends up going to R2; what do you do then?
Only 62-63% of hands will make it that far, and in those other bid hands you will lose the game about 1250 times (opponents earn at least a point). You'll euchre the opponents a fair bit, R1, but not enough to justify passing.

If you bid alone in D and lead the JH, the outcome is (332 / 3000 / 2912) EV = -0.22 [14.6% chance of winning the game]

I also compared bidding D w/partner (JH lead) and bidding C w/partner (QC lead):

bid D: EV = +0.14 (10,736 / 30,329 / 21,390) win game 15.8%
bid C: EV = +0.22 (9,208 / 34,129 / 19,118) win game16.0%

Bidding D yields more sweeps but also more euchres. In the end, this tightens the gap between odds of winning the game under the 2 scenarios (because those euchres include superfluous points for the opponents), but it is slightly better to bid clubs. Or call it a tie, as the numbers are so close.

Of course, you'll never have to make this decision because you'll bid alone, R1!

So C is actually a better bid than D!


Good stuff man. Not sure how I initially came to the conclusion that going alone in Next was best. Glad my instinct is right that going alone in Hearts and kamikaze leading the AH is best.

The only question I have left with this hand is this: Is going alone and kamikaze leading the AH better than just calling hearts and playing cautiously, I.E. leading the Right but then only leading the AH if the Left comes out, leading the AS followed by the AC otherwise. We know kamikaze leading the AH is better loner vs loner, but does it beat out just calling with hearts.

Wes (aka the legend)
Posts: 1418
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Fri Jan 14, 2022 3:18 am

raydog wrote:
Thu Jan 13, 2022 5:31 pm
I assumed 9C as turn card, 9H discarded by dealer (or left in dealer's hand if passes!)

Ran 10,000 for each scenario (except one, noted below).

(Card_10-C) (Card_9-C) (Card_Q-D) (Card_10-D) (Card_9-D) [from previous series]
bid: EV = -1.01
pass: EV = -1.15
5,755 hands make it to dealer
If dealer passes, S1 calls alone more than 1/3 of the time in R2, and makes almost 1/2 of them!
Best to bid; essentially donating.
My god this is amazing stuff man. Who would've thought this call would be +EV vs passing. Awesome.
raydog wrote:
Thu Jan 13, 2022 5:31 pm
(Card_10-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)
bid: EV = -0.87
pass = -0.88
5,473 hands make it to dealer
Pretty much a toss-up here. 40% less successful loners by S1R2
I think it's pretty amazing that this crappy hand reaches an indifference point. Wow. I mean it makes sense as we all know why--the cost of passing is brutal with these holdings--but there's no way I ever would've hypothesized it.
raydog wrote:
Thu Jan 13, 2022 5:31 pm
(Card_Q-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)
bid: EV = -0.71
pass: EV = -0.79
5,615 hands make it to dealer
Better to bid.
So awesome man. Mindblowing stuff.
raydog wrote:
Thu Jan 13, 2022 5:31 pm
(Card_K-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)
bid: EV = -0.53
pass: EV = -0.70
5,640 hands make it to dealer
Better to bid.
Wow that's a significant gap. Nice.
raydog wrote:
Thu Jan 13, 2022 5:31 pm
(Card_A-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)
bid: EV = -0.35
pass = EV = -0.32
Given the sudden improvement in the passing score, and the closeness of the EVs, I ran this again with 100,000 hands:
bid: EV = -0.36
pass EV = -0.35
56,924 hands make to dealer
Another toss-up. It seems holding a singleton Ace (likely in a green suit, albeit a short suit) helps A LOT on defense against opponents' bids.
Yep, seeing these results it totally makes sense that Kc9c would actually be a better call that Ac9c. No way my brain catches this without your simulator tho. For the sake of simplicity I would treat this hand--Ac9c--as an indifferent point and put it in the call bucket. Like I said, tie goes to the caller. Also, given that like 95% of the time you'll be playing with an amateur partner who will pass biddables from the 2 seat, in real life this hand is probably slightly +EV. Conversely with an expert partner who knows that calling from the 2S-R1 with say QcTcAd9d9h is +EV, making this thin dealer call may be slightly negative but practically no players in the world even know that.
raydog wrote:
Thu Jan 13, 2022 5:31 pm
I also checked this hand again, assuming S4 holding the KH rather than the 9H, and still discarding it if bidding:
(Card_10-C) (Card_9-C) (Card_Q-S) (Card_10-S) (Card_9-S)
bid: EV = -0.87
pass: EV = -0.77
5,523 hands make it to dealer.
Here, that bit of extra strength in the hand tilts the balance towards passing, as I'm sure it would as well with the other toss-up hand (with the A-9C).
Interesting stuff man. At the very worst what we have here is a very good dealer donate candidate. Up 3 points+ I would always make this call. And with an amateur P who passes biddables from the 2 seat, this hand may go into the call bucket. From the 2 seat the typical amateur is passing R+1+0, L+1+A, 2 non-bower trump +2 aces, and certainly other marginal +EV calls like 2 non-bower trump + an off ace when they don't block much.

Post Reply