Advanced Euchre QUIZ - PART 2

Ask questions, discuss and debate your strategies, euchre polls and more
Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Re: Advanced Euchre QUIZ - PART 2

Unread post by Wes (aka the legend) » Fri Jan 14, 2022 4:38 am

raydog wrote:
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.
Obviously I'm not surprised by this finding. Glad your data backs up my claim. I've tested hands like this vs Pass-Call Next (even tho that might sound crazy given we have no next cards), and calling won at a sample large enough to reach a 95% confidence interval. This hand type has never been tested against pass-pass until now. I know calling had a negative expected outcome, and I predicted pass-pass would have a higher negative expected outcome, but the numbers are closer than what I would've thought. Either way, I feel comfortable saying calling is best. Again even in the worst light--a statistical tie--ties go to the caller.

As far as some assumptions with this holding. When S1 calls spades we are always leading a spade. If we pass-pass, and the S2 calls red, I would always lead the red non-trump card (an argument could be made for leading the turned down suit through the maker tho given how dirty it is trying to force an overtrump). If S2 calls clubs, I would lead the KD (again an argument could be made for leading the turned down suit through the maker trying to create an overtrump situation). If it gets to my P in the 2nd rd, I would always lead trump if I have it. If he calls clubs I would lead the KD. If the action gets to S4 in the 2nd rd: when he calls red I'm leading the non-trump red card. If he calls clubs, I'm leading the KD.

Worthy of note, I've talked about this awhile back in other threads but my claims have never been verified.

Claim 1: Same situation BUT if we have:

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

Pass-pass is now better than calling spades in R1 becuz we have good defense in the 2nd round given that we block reverse next calls.

Claim 2: If we have:

(Card_Q-S) (Card_10-S) (Card_9-S) (Card_J-H) (Card_9-H)

Passing with the intention of calling hearts in the 2nd rd and probably leading the QS hoping the KS is buried is better than calling spades in the 1st rd.

The crux of this situation is S1 has no defense and nowhere to go in the 2nd rd. When that's not the case S1 should no longer make this very marginal call. Am I correct? Can you test claim 1 and 2?
raydog wrote:
Thu Jan 13, 2022 7:19 pm
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 think we've covered most of these possible hands from the dealer spot but if you discover new ones please let us know!

I'll throw one out there for you to test. It's not a good hand but I'm curious. Score is 0-0, you're the dealer.

Upcard: (Card_K-H)

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

Discard the 9S. If this hand is +EV--meaning it's better than passing--I'm gonna crap my pants. And just in case having that AH makes passing the better play test this hand too (same KH upcard):

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

BTW as far as other spots "where bidding, while generating a negative EV, is still favorable to passing" I'm predicting you'll find a lot of them in the 2 seat 1st rd.

I recently tested this 2 seat hand:

Upcard: (Card_Q-C)

We have: (Card_10-C) (Card_9-C) (Card_A-D) (Card_9-D) (Card_9-H)

Calling beat out passing at a 95% CI.

If S1 led a non-ace spade I always played off with the 9H. If S1 led the AS, I would trump in and then lead the AD on 2nd street unless my P threw off a diamond on the first lead suggesting they have another diamond. In that case I'm torn between leading the 9H or leading my last trump. With no more voids to work with, I think leading trump is best but don't know. If S1 led a diamond and I took that trick with my AD and my P threw off a spade, I would lead the 9h on 2nd street hoping my P is void. If on that diamond lead my P threw off a heart meaning he probably has another heart I honestly don't know the best lead on 2nd street, I would actually lead trump, but leading the 9H could be best. If my P threw off an Ace on my diamond trick I would always lead trump on 2nd street as my P is probably loaded.

And whenever my P takes the first trick they will always lead trump if they have it. If they only have 1 trump and use it on the first lead, then I would have them lead the AH if they have it or the cleanest fresh suit to lower the probability of me getting overtrumped; unless they had a signal to lead something else, E.G. if S1 leads a non-ace spade, and I throw off the 9H and my P takes the first trick with a trump and has no more trump and no off fresh AH to lead, then they should lead a non-ace heart if they have one as I am likely void in that suit now. If S4 has no hearts and no trump, then obviously leading diamonds is their only choice.

Jesus, that's a lot of assumptions behind a relatively simple marginal call. I'm exhausted after writing all that, probably worse for the reader lol. Just shows how complex a euchre hand can be.

As far as a more marginal 2S hand to test how bout this one:

Upcard (Card_Q-H)

We have (Card_A-H) (Card_K-H) (Card_K-C) (Card_Q-C) (Card_K-S)
raydog wrote:
Thu Jan 13, 2022 7:19 pm
I remember Eric Zalas could never really accept this, and was always focussed on +EV.
Eric Zalas was actually myopically focused on +EO (expected outcome). It somehow never occurred to him that a -EO call could be +EV overall if passing had a worse -EO--a critically important concept in euchre. It actually makes his entire work on euchre farcical yet I would still recommend his work to any expert becuz the hand samples and simulations Zalas did are interesting and important. You just have to ignore Zalas' flawed conclusions.
raydog wrote:
Thu Jan 13, 2022 7:19 pm
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.
I've thought of this problem before too. It's almost like we need two simulators. In one simulator everyone makes the correct play answering the question "what is the best play in a very tough game). In the other simulator you have 1 person making the correct play while everyone else plays like an amateur answering the question "what is the best play in a typical euchre game you'll be in on an app and usually in real life". But honestly I don't think this theoretical problem is that big of a deal and for the sake of parsimony I think we should ignore it. Plus it's not that hard to make adjustments to your findings. For example, say this 2 seat call was very slightly -EV overall:

Upcard (Card_Q-H)

We have (Card_A-H) (Card_K-H) (Card_K-C) (Card_Q-C) (Card_K-S)

We could still deduce that this hand is probably a call with an amateur P who is not calling with R+1+0, L+1+A, 2 non-bower trump + 2 aces, even 3 low trump- three suited-no off ace hands, and of course those marginal 2 non-bower trump + an off ace hands that are actually +EV calls.



raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Sat Jan 15, 2022 4:28 pm

"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."

Don't actually need to simulate this. Calling H alone and playing cautiously yields:
(2,127 / 6,106 / 1,767) sweeps / 1 pt / euchred
Imagine that, by taking partner along, you are able to reduce the euchres to just 200 and turn all the rest into sweeps (which is clearly never going to happen). You then get an outcome like (3,694 / 6,106 / 200),
which has an EV of +1.31. Which barely beats going alone in H, kamikaze! So it will never be better to take partner along.
______________________

Claim 1:
bid: EV = -0.67
pass/pass: EV = -0.61 less points by both sides, but favoring S1/S3

Claim 2:
bid: EV = -0.41
pass/bid R2: EV = -0.19

pass/bid R2: EV = -0.16
pass/pass: EV = -0.42
so pass/bid R2 best option

BTW, I found it slightly better, when bidding H R2 in claim 2, to lead the JH, followed by the QS:
tried 100,000 hands; 48,471 made it to R2
lead JH, then QS: EV = +0.04
lead QS: EV = -0.08
[these EVs can't be compared to those above because they only consider R2 bids]
____________________________
I'll throw one out there for you to test. It's not a good hand but I'm curious. Score is 0-0, you're the dealer.

Upcard: (Card_K-H)

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

Discard the 9S. If this hand is +EV--meaning it's better than passing--I'm gonna crap my pants.

bid: EV = -0.84
pass: EV = -0.75

And just in case having that AH makes passing the better play test this hand too (same KH upcard):

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

bid: EV = -1.07 (38 / 1,831 / 4,175)
pass: EV = -1.13 (S1R2 bids 77% of hands for an EV of +1.39!)
this is very close, but only slightly higher margin than the previous hand

I recently tested this 2 seat hand:

Upcard: (Card_Q-C)

We have: (Card_10-C) (Card_9-C) (Card_A-D) (Card_9-D) (Card_9-H)

Calling beat out passing at a 95% CI.

I get:
bid: EV = +0.09
pass: EV = +0.03

As far as a more marginal 2S hand to test how bout this one:

Upcard (Card_Q-H)

bid: EV = -0.37
pass = -0.14

We have (Card_A-H) (Card_K-H) (Card_K-C) (Card_Q-C) (Card_K-S)

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Sat Jan 15, 2022 7:31 pm

raydog wrote:
Sat Jan 15, 2022 4:28 pm
"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."

Don't actually need to simulate this. Calling H alone and playing cautiously yields:
(2,127 / 6,106 / 1,767) sweeps / 1 pt / euchred
Imagine that, by taking partner along, you are able to reduce the euchres to just 200 and turn all the rest into sweeps (which is clearly never going to happen). You then get an outcome like (3,694 / 6,106 / 200),
which has an EV of +1.31. Which barely beats going alone in H, kamikaze! So it will never be better to take partner along.
That's awesome man. Pretty cool finding there.
raydog wrote:
Sat Jan 15, 2022 4:28 pm
Claim 1:
bid: EV = -0.67
pass/pass: EV = -0.61 less points by both sides, but favoring S1/S3

Claim 2:
bid: EV = -0.41
pass/bid R2: EV = -0.19

pass/bid R2: EV = -0.16
pass/pass: EV = -0.42
so pass/bid R2 best option
Glad to see that both my claims are correct!! So much of my decision matrix revolves around what suits I block and what I dont block. It's uncanny how well this heuristic lines up with the math.
raydog wrote:
Sat Jan 15, 2022 4:28 pm
BTW, I found it slightly better, when bidding H R2 in claim 2, to lead the JH, followed by the QS:
tried 100,000 hands; 48,471 made it to R2
lead JH, then QS: EV = +0.04
lead QS: EV = -0.08
[these EVs can't be compared to those above because they only consider R2 bids]
Now that is a very interesting surprise but still one we can make sense of. This S1-R2 call basically has only two lead choices and they both suck. 1) lead the turned down suit or 2) lead the Right from just 2 trump with no aces to promote. I suppose leading the turned down suit is so bad that even a conventionally poor trump lead beats it out. Great stuff.
raydog wrote:
Sat Jan 15, 2022 4:28 pm
I'll throw one out there for you to test. It's not a good hand but I'm curious. Score is 0-0, you're the dealer.

Upcard: (Card_K-H)

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

Discard the 9S. If this hand is +EV--meaning it's better than passing--I'm gonna crap my pants.

bid: EV = -0.84
pass: EV = -0.75
Good work. That's still pretty damn close. This hand would be another excellent dealer donate candidate. And keep in mind the prospects of this hand are usually even better than one would think given that probably 95% of people pass biddable hands from seat 2, so Seat 2 will have a little help more often then he theoretically should.
raydog wrote:
Sat Jan 15, 2022 4:28 pm
And just in case having that AH makes passing the better play test this hand too (same KH upcard):

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

bid: EV = -1.07 (38 / 1,831 / 4,175)
pass: EV = -1.13 (S1R2 bids 77% of hands for an EV of +1.39!)
this is very close, but only slightly higher margin than the previous hand
Wow. It's crazy that one could actually argue that crap hand is a call and have the math backing him up. Just wow. The cost of passing is real folks.
raydog wrote:
Sat Jan 15, 2022 4:28 pm
I recently tested this 2 seat hand:

Upcard: (Card_Q-C)

We have: (Card_10-C) (Card_9-C) (Card_A-D) (Card_9-D) (Card_9-H)

Calling beat out passing at a 95% CI.

I get:
bid: EV = +0.09
pass: EV = +0.03
I would've predicted a wider margin but I'm still happy your results are consistent with mine. This hand being a +EV call or at least neutral is important because one can extrapolate a lot from this hand as this hand is the weakest hand from this configuration. People are playing too tight from the 2S-R1. They just don't realize it yet.
raydog wrote:
Sat Jan 15, 2022 4:28 pm
As far as a more marginal 2S hand to test how bout this one:

Upcard (Card_Q-H)

bid: EV = -0.37
pass = -0.14

We have (Card_A-H) (Card_K-H) (Card_K-C) (Card_Q-C) (Card_K-S)
So making this dubious call burns -.23 points. Not bad. This is the type of 2 seat call one needs to make when they have a big lead. Don't even let the 2nd rd happen. And what's more is it's still theoretically possible this call is actually +EV. Again, most of the time one plays euchre they will be teamed up with an amateur partner who passes biddables. And it's damn near perfectly predictable which biddable hands S4 is passing: R+1+0, L+1+A, 2 non-bower trump +2A, 3 low trump, 3 suited, no ace hands, and of course they're never making those really thin 2 low trump + Ace -EO calls that are actually +EV.

What's more, an amateur will not go alone enough, 2 trump loners are not even in their range. And many other 3 trump loners are just calls to them like say JhQh9h+Ac9c. So when S2 makes this dubious call they capture all those biddable hands S4 is passing and they don't even pay the full price of blocking S4 loners. This dynamic could possibly push this -EV to the plus category. And when one plays with randoms on an app or in real life, this is the situation they will be facing 95%+ of the time.

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Sun Jan 16, 2022 11:21 am

RAY,
I am curious about this hand where S1 pass with a result of 23% success rate of loners S2/S4. What card did your program lead to the first trick when S2 or S4 goes alone? If I were S1, I would lead the 10S or 9S to the first trick?

Irish

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.

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Sun Jan 16, 2022 12:00 pm

irishwolf wrote:
Sun Jan 16, 2022 11:21 am
RAY,
I am curious about this hand where S1 pass with a result of 23% success rate of loners S2/S4. What card did your program lead to the first trick when S2 or S4 goes alone? If I were S1, I would lead the 10S or 9S to the first trick?
For the record, I would also make the same spade lead against a loner for reasons Wolf already mentioned.

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Sun Jan 16, 2022 5:11 pm

So my program leads the KD from S1 when a loner is bid by S2 or S4. The logic behind this is that the KD and QH have the best chances of beating a low off-suit by the loner declarer, so best to lead one of them right away and avoid having to potentially decide which one to keep on the 4th trick.

I tested the lone hands from my simulation of problem #10, to see how the results would change if S1 instead led the 9S when one of the opponents bid alone.

If the lone bid was made R1 (S as trump), the results were disastrous, with S2's loner success rate almost doubling and S4's loner success rate more than tripling! If the lone was called holding just 2 trump, then S1's 3 trump will stop it, period - no need to lead trump. Just a miserable strategy.

If the lone was bid R2, however, there was some improvement for S1/S3 (more lones stopped and more euchres) if the 9S was led. The idea here is that, with only 3 spades remaining for the other 3 players + the kitty, I have a decent chance of hitting my partner's void in S when the declarer is NOT void in S. Here are the numbers:

KD led:
S2R2: (2,909 / 5,315 / 545) [4pts / 1 pt / euchred]
S4R2: (75 / 28 / 0)

9S led:
S2R2: (2,805 / 5,275 / 689)
S4R2: (71 / 32 / 0)

In aggregate, S1/S3 earn an additional 786 pts relative to S2/S4 by leading the 9S, so this is the better play. However, as these results were calculated based on 100,000 hands dealt, the net improvement in EV amounts to +0.0079, so not enough to change my initial conclusion.

Nonetheless, an astute strategy and a cool result!
Last edited by raydog on Mon Jan 17, 2022 1:34 pm, edited 1 time in total.

jblowery
Posts: 184
Joined: Tue Dec 11, 2018 10:43 am

Unread post by jblowery » Mon Jan 17, 2022 11:50 am

One concern about your simulator though. In reality, when your team passes clubs, your opponents, on average, are going to be a little stronger in spades. I doubt the simulator accounts for that. Does anybody think that changes anything?

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Jan 17, 2022 12:23 pm

I say NO to this if cards are randomly dealt: "Does anybody think that changes anything?"

But can impact this situation, IMO, is the tendency of Eldest to call next with a weaker hand than he might Cross the suit. The Jacks of the upcard, for sure are impacted as to now who hold them. S2 could have either Jack unprotected. If this is taken into consideration by the Simulator (programmer) to call thinner, then of course. I will let Ray answer but I have observed he has accounted for that situation vs crossing suit at S1.

IRISH

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Mon Jan 17, 2022 2:13 pm

When deciding which suit to bid, R2, my program gives a little boost to next calls by S1 and S3, and to green suit calls by S2 and S4. The suit of the potential call also affects the point threshold necessary to justify making a call (e.g., less pts needed for S1 to make a call in next than in a green suit). The EXTENT of this boost (or change in pt threshold) is determined by testing two different values over 1,000,000 random hands and adjusting until finding the value that optimizes the result. This is an iterative and recursive process which I spent hundreds of hours doing. And I still use it today to make adjustments when I discover a situation where my program bids incorrectly. So far I have not found it necessary to make any changes to the core premise (the idea of calculating pt values for hands), but I do occasionally introduce a "special situations" rule [like "only lead an ace against a lone call if you have 2 or more of them"].

I have simply NOTICED that S1, for example, can get away with a thinner call in next than in a green suit, R2; it's not something I explicitly program for, but it naturally gets taken into account during the process. When I try to explain the effect, I come to same conclusion as Irish. If a spade is turned down, it's not that my opponents in S2 and S4 have fewer clubs, they are simply less likely to have the JC and the JS, which happen to be the 2 most important cards for a clubs bid! It all revolves around those 2 bowers, nothing else. The effect is minimal in terms of length of suit, but because it pertains the two most highly valued cards, the effect is disproportionately important. If I am in S1, R2, and a spade is turned down, even if I don't have either black bower, just knowing that the chances that S2 or S4 have them is diminished allows me to confidently make a clubs bid with a weaker hand than I otherwise would.

True story: I once purposely cheated at my fraternity's annual euchre tournament because everyone kept saying "cheating is allowed". I was paired with one of the worst players in the field, so I decided to try an experiment. At the start of every hand, we did the following: if we didn't hold a bower in the turned suit we would simply say the word "pass" on our turn if we wanted to pass; but if we held at least one of the bowers we would add some other word ("I pass", "I'll pass", "Ummm...pass", etc.) That's it. With that tiny bit of added information we came in 3rd in the tournament (30 players). I later confessed to everyone what I had done, no one seemed to bat an eye. (we have since changed our ethos to overtly discourage cheating)

Just to say - bowers, and who has them, is WAY important!

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Jan 17, 2022 3:15 pm

But what upsets the apple cart is when an S2/S4 opponent Bags on Next . . .

True story to tell. I was once in a major tournament (Joe Andres World Series of Euchre..) ready to play. And a good player who had his girl friend as a partner. Apparently she was quite new to the game. I overheard him say to her. "You must call next in 1st seat. Don't worry about how strong your hand will be, I will have them." So me and my partner bagged big time. Nevertheless, the game was lopsided. So you have to beware!

And the thing about Next baggers . . . if don't have them you can't bag. So what is the probability that the Dealer can have next and Bag. 64 dollar question?

"I have simply NOTICED that S1, for example, can get away with a thinner call in next than in a green suit, R2; it's not something I explicitly program for, but it naturally gets taken into account during the process. When I try to explain the effect, I come to same conclusion as Irish. If a spade is turned down, it's not that my opponents in S2 and S4 have fewer clubs, they are simply less likely to have the JC and the JS "

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Mon Jan 17, 2022 4:21 pm

I don't believe a strategy of always bagging is going to make you a more winning player (though I can't PROVE that statement). Of course, if you know the proclivities of your opponents (tendency to bag or to call next) you can use that to your advantage. That's why computers are not going to take over the world (at least not computers programmed by me!)

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Jan 17, 2022 5:25 pm

raydog wrote:
Mon Jan 17, 2022 4:21 pm
I don't believe a strategy of always bagging is going to make you a more winning player (though I can't PROVE that statement). Of course, if you know the proclivities of your opponents (tendency to bag or to call next) you can use that to your advantage. That's why computers are not going to take over the world (at least not computers programmed by me!)
I am very skeptical of bagging next in general. I mean for one 95%+ of the euchre population does not call next. So the vast majority of time forgeddaboutit. In my weekly tournament of roughly 30 people, only 4 people call Next. Everyone else plays their hand. So around 87% of the euchre population in my tournament does not call Next and when I play on the app it's way worse than that (people who call Next on the app are true unicorns), so I'm still sticking with my overall 95% estimate.

**Total tangent here but worth mentioning: Going back at least 10 years (I don't know the players beyond that timeline) a player who doesn't call Next has NEVER won the yearly championship. That's pretty good evidence demonstrating the power of Next!**

Ok back to the topic. Let's forget about the 95% estimate of people who don't call Next. Let's just focus on the 5% that do. We have still NEVER PROVEN that bagging Next is a viable strategy. Like we haven't even established proof of concept, I.E. we don't even have one extreme example showing this strategy works that reaches a 95% confidence interval.

And BTW bagging Next from S4 or S2 only count's if someone is passing a biddable hand. If they're not passing a biddable, I.E. they would've passed anyway then that's not bagging Next, then they're just doing what everyone esle would do anyways. For example, Say the dealer has this hand:

Upcard: (Card_9-C)

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

or

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

Passing those hands do not count towards bagging Next as I would argue both hands should be passed anyways.

Again, a biddable hand must be passed or we are not really exploiting S1 for calling Next too much. Here's a true example of a hand that WOULD count towards bagging Next.

Upcard (Card_K-C)

(Card_J-C) (Card_K-S) (Card_Q-S) (Card_J-D) (Card_K-D)

Ok, R+1+0 is unequivocally a biddable hand when one doesn't block all suits, unless Rays simulator says otherwise. So now we are truly passing a biddable hand to exploit the fact that S1 is supposedly calling Next too much.

Ok so there's your testable hypothesis. Prove this biddable hand is actually a pass vs someone like me in S1 who aggressively calls Next:

Upcard (Card_K-C)

(Card_J-C) (Card_K-S) (Card_Q-S) (Card_J-D) (Card_K-D)

If that's the case, then we have proof of concept that bagging next is a viable strategy. Until then, I'm basically in this camp:
I don't believe a strategy of always bagging is going to make you a more winning player (though I can't PROVE that statement).
More specifically, I wouldn't say so much that "I don't believe", it's actually more like this: even if one can prove a few cases where bagging Next is the correct play, this spot will be so rare it will have minuscule effect on one's win rate. IOW as fun as this topic sounds, I think it's mostly a waste of our time.

And BTW I don't think this is something Ray should waste his time testing. It would probably be too much trouble with little payoff to create a S1 who calls Next at say my frequency and there's a lot of embedded assumptions in there (hands I'll pass, and hands I'll jump the fence with). For example I don't call Next in this spot:

Dealer turns down (Card_9-S)

I have: (Card_K-C) (Card_10-C) (Card_A-D) (Card_J-D) (Card_10-S)

I call diamonds cuz I believe its the slightly better play or close enough to a tie that I'd rather jump the fence to keep S2-S4 off my scent.

The truth is the best way to test this hypothesis would be for me to run the test on my kitchen table becuz I probably call Next more than anyone on this planet making me the ideal candidate to test plus I know exactly how I play all configurations. But even If I ran that test and found a few spots where I'm exploitable what's that information really worth? How many people out there actually play like me? The only real value I see from running this test is we could actually put to bed the idea of bagging next altogether. What I mean by that is, IF I test this hand vs me and it turns out that bagging Next is STILL -EV--that one is still better off calling with that biddable hand--then that's it fellas. If bagging Next doesn't work vs me then it's game over for bagging Next.

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Jan 17, 2022 5:36 pm

I agree with, ..." don't believe a strategy of always bagging is going to make you a more winning player (though I can't PROVE that statement) "

Yep, Bagging is a short-range strategy. For winning in euchre in the long run, you have to bid on biddable hands that have a probably + EV. And bid depending on score & situation to prevent a more negative EVs which most players seldom do (meaning to bid even though you will be euchred. But bagging can be used to your advantage as well, but you have to be right when used. That is what makes Euchre fun and interesting. No game like it! IMO

What is also interesting is, everyone who has played much euchre has a particulate base style of play. Even if they try to disguise and change up at times. They always return to that base style. In other words, it is their personality that is hard wired.

Overall, EUCHRE IS A GAME OF AGGRESSION!, that is If you want a great winning recording. Conservatism doesn't work as you are passing on "biddable hands" and points.
That's my story and I am sticking to it! :lol:

IRISHWOLF

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Jan 17, 2022 6:12 pm

I don't know about you guys but I can't wait til Ray tests the next quiz hand. The next hand being:

11) The score is 0-0. You're in the 3rd seat. The dealer upcard is the (Card_10-H)

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

I think most people said call. I said pass. I think there's a decent chance I'm wrong on this one. I hope I am, as that would be an exciting result to me. Of course the problem with S3 hands tho is the correct answer could easily depend on who your partner is. I'm not trying to escape being wrong tho. My claim at the time or at least implied claim is that this hand is a pass even with an amateur P who invariably doesn't play the 2nd rd that well. But it could easily be the case that with an amateur P this hand is actually a call, but when I have a partner like say Edward then it's a pass. I think its probably impossible for Ray's simulator to break it down like that but it doesn't have to be that precise. If Ray's simulator shows that calling is +EV vs passing that is all the information we need to know that passing with an amateur P is incorrect and since that is the spot we'll invariably be in like 95% of the time that information is very valuable! If I wanna test this hand to see if passing is correct with Edward as my P or say my clone as my P that's really something I should test myself kitchen table style. Of course if Ray's test shows that passing is +EV vs calling then I wouldn't even need to do that kitchen table sample. Now I would know that passing is best with Edward as my P, but unfortunately we'd all still be left to guess if passing is correct with an amateur P.

So we should all be hoping Rays work shows this call to be +EV becuz that's way more valuable information for us. If Ray's work shows passing to be +EV that doesn't really prove my initial claim that passing is correct becuz we still don't know if passing is best with an amateur P as I don't think Ray's simulator can capture that. Bottom line let's all hope I'm wrong on this one becuz If I'm right I'm not sure Rays simulator can really prove it.

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Jan 17, 2022 7:54 pm

Oh ya! I SAID ORDER AND I WILL SAY IT AGAIN! --- IRISH

11) The score is 0-0. You're in the 3rd seat. The dealer upcard is the (Card_10-H)

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

I think most people said call. I said pass.

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Mon Jan 17, 2022 8:04 pm

I'm not sure what it means to say "95% of players don't call next." My stance is that, if I'm in S1R2, I understand I can call next with a weaker hand than I can call green, and expect the same results, that's all. I won't always call next - sometimes I'll pass, sometimes I'll call the other color suit. But despite next being only 1/3 of the suits I could possibly bid, I will bid it more than 1/2 the time, so disproportionately often.

I just set my program to run 10,000 random hands, and tallied S1R2 bids:

decisions to make: 2,539
# of bids made: 1,778 (70% of hands)
# of next bids made: 1,239 (70% of hands bid)

What I should do now is look at the average EV of hands bid in next vs. hands bid in a green suit. And also compare those EVs to what would have transpired if I had passed in S1R2.

But I think I'll go watch football instead :)

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Jan 17, 2022 8:31 pm

raydog wrote:
Mon Jan 17, 2022 8:04 pm
I'm not sure what it means to say "95% of players don't call next." My stance is that, if I'm in S1R2, I understand I can call next with a weaker hand than I can call green, and expect the same results, that's all. I won't always call next - sometimes I'll pass, sometimes I'll call the other color suit. But despite next being only 1/3 of the suits I could possibly bid, I will bid it more than 1/2 the time, so disproportionately often.

I just set my program to run 10,000 random hands, and tallied S1R2 bids:

decisions to make: 2,539
# of bids made: 1,778 (70% of hands)
# of next bids made: 1,239 (70% of hands bid)

What I should do now is look at the average EV of hands bid in next vs. hands bid in a green suit. And also compare those EVs to what would have transpired if I had passed in S1R2.

But I think I'll go watch football instead :)
I'm just saying that for 95%+ of players they are simply playing their hand. They are not calling next unless they have a good hand in Next, which in my book doesn't really count for calling Next. A true Next call to me is a call S1 doesn't really wanna make but he makes it anyways for defensive purposes as that is the essence of calling Next. You make marginal Next calls when you don't block reverse Next.

For example, score is 0-0 and the dealer turned down the (Card_9-S)

S1 has: (Card_K-C) (Card_Q-C) (Card_A-D) (Card_9-D) (Card_10-S)

An expert will be calling Next (clubs) every time in that spot. An amateur is passing every time in that spot given that they don't have a conventional calling hand. The idea of calling Next for defensive purposes doesn't register for an amateur. That concept will even seem almost ridiculous to most people as their top priority is to not get euchred and beyond that they hope to get lucky and win.

You're welcome to test that hand if you want. :)

PS: In case it's relevant, if I had this hand I would pass. Same card turned down:

(Card_K-C) (Card_Q-C) (Card_A-D) (Card_J-H) (Card_10-S)

Now that I have reverse Next blocked there is no need for me to call marginal for defensive purposes.

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Mon Jan 24, 2022 11:00 am

Question #11:
The score is 0-0. You're in the 3rd seat. The dealer upcard is the (Card_10-H)

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

Here's what I found (100,000 hands; about 86,000 make it to S3 for a decision):
bid w/ partner: (10,358 / 55,184 / 20,386) EV = +0.41
pass: EV = +0.54 [33,310 bids by S4 = 39%; 24,068 euchres of S4 = 72%]

I found that passing was better, simply because S4 will call and get euchred so often [this may not be true in an actual game! You need to know your opponent] AND partner in S1 has free rein on R2 if S4 passes. It's not bad to bid - the EV is positive - but it could potentially be higher if you pass and the dealer + partner play as expected (perhaps a big "if"?)

I played around a bit with the scenario. I tried AH as the turn card; I tried A-QH in S3 hand; I tried A-9C in S3 hand; I tried A-9H in S3 hand and KH turned; I looked at bidding alone in every scenario. In each case, better to pass.

The huge caveat here of course is the expectation of how S4 and S1 will play if you pass. In each of the scenarios listed above, there is a positive EV when bidding - it just gets significantly larger if you pass.

It probably helps a lot that S3 holds the JD, which will be a big boon if partner in S1 bids next aggressively, R2.

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Jan 24, 2022 11:05 pm

raydog wrote:
Mon Jan 24, 2022 11:00 am
Question #11:
The score is 0-0. You're in the 3rd seat. The dealer upcard is the (Card_10-H)

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

Here's what I found (100,000 hands; about 86,000 make it to S3 for a decision):
bid w/ partner: (10,358 / 55,184 / 20,386) EV = +0.41
pass: EV = +0.54 [33,310 bids by S4 = 39%; 24,068 euchres of S4 = 72%]

I found that passing was better, simply because S4 will call and get euchred so often [this may not be true in an actual game! You need to know your opponent] AND partner in S1 has free rein on R2 if S4 passes. It's not bad to bid - the EV is positive - but it could potentially be higher if you pass and the dealer + partner play as expected (perhaps a big "if"?)
Very good stuff Ray. I think we can now say unequivocally that if one has a good partner they should pass this hand from R1-S3. That debate is over to me. The only question left is: is this still a pass with an amateur P. I think that's a problem me or Irishwolf would have to solve at the kitchen table. There's two main variations we'd have to probably test to get a conclusive result. 1) The dealer and S1 are both amateurs and 2) The dealer is a good player but S1 is an amateur. By amateur I simply mean the typical random player you play with online and most people you play with in real life. For example, say the dealer passes the TH, the typical amateur is passing all these Next calls from S1-R2:

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

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

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

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

(Card_K-D) (Card_Q-D) (Card_A-H) (Card_9-H) (Card_9-C)
raydog wrote:
Mon Jan 24, 2022 11:00 am
I tried A-QH in S3 hand;
Just to make sure I understand what you're saying. You tried this hand correct? (assume same TH upcard and same 0-0 score for all hands):

(Card_J-D) (Card_A-H) (Card_Q-H) (Card_K-C) (Card_Q-C)

Not so subtle brag. Me and Edward (Tbolt65) finished first place in the Las Vegas partnership tournament on Saturday. I'll tell you what I would do with all these hands with Edward as my P, I.E. someone I can trust:

I would pass this hand. Having approx 2 tricks in Next, no reason to call marginal. This is an easy sandbag situation. With an amateur P I'm not sure.
raydog wrote:
Mon Jan 24, 2022 11:00 am
I tried A-9C in S3 hand;
This is the hand you tried correct?:

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

Again, easy pass. I have 2 tricks in Next. Bag the dealer all day with Edward as my P. However, this hand is now strong enough in the absolute sense that I would hypothesize that we should call this with an amateur P.
raydog wrote:
Mon Jan 24, 2022 11:00 am
I tried A-9H in S3 hand and KH turned; I looked at bidding alone in every scenario. In each case, better to pass.
So you tried this hand:

Upcard: (Card_K-H)

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

Again, I would pass. 2 tricks in Next. Easy bag. No need to call marginal with a P I can trust. With an amateur P I'm not sure.
raydog wrote:
Mon Jan 24, 2022 11:00 am
The huge caveat here of course is the expectation of how S4 and S1 will play if you pass. In each of the scenarios listed above, there is a positive EV when bidding - it just gets significantly larger if you pass.

It probably helps a lot that S3 holds the JD, which will be a big boon if partner in S1 bids next aggressively, R2.
Yep, S3 holding that JD is critical. With a good P I have always suspected it's correct to bag most L-X-X hand combos from S3-R1.

What I love about your work is it supports how me and Edward play. IMO me and Edward have probably established the best S1-S3 chemistry in the euchre universe. That doesn't mean we play perfect tho. In fact there are plays we make I'm very skeptical of. Just that we're ahead of the curve becuz most people including good players call from S3 too much. That's one of the weird ironies of this game. The typical amateur passes way more often than the expert but from S3-R1 the amateur probably calls more often than the expert.

A fun game for you Ray would be to try to figure out the strongest S3 hand where it's still correct to pass. Assume S4 and S1 plays the way they play in your simulator. No adjustments needed. For example, I've always wondered if this hand is bag-worthy:

Dealer upcard: (Card_10-S)

Action gets to S3 who holds:

(Card_J-S) (Card_J-C) (Card_9-S) (Card_9-C) (Card_9-D)

Edward will pass that hand from S3 when I'm his P becuz he knows in the 2nd rd I will always call something veering towards Next if I don't block reverse next. I know your simulator can't capture that dynamic. Don't worry about that. I'm just curious what results it would spit out with your S1 simulator P.

If that hand ends up being a call, would things change if we gave S3 a red bower?

(Card_J-S) (Card_J-C) (Card_9-S) (Card_9-C) (Card_J-H)

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Jan 24, 2022 11:53 pm

LOL Getting into the Details, so many ifs & butts here: Here my answer.

The original answer to # 11 was: THIS HAND IS TOO WEAK TO ORDER.
I answered otherwise, it is a strong hand. And Ray's simulations Proves that with EV = +0.41 23.7% Euchre rate, minus the 12.0% SWEEPS NET = 11.7 negative points. That is guaranteed. I say, Bird in the Hand is Worth 2 in the Bush! I knew there was no guessing that this was a positive EV. Seldom will I pass on a strong positive EV. So it all Depends for me but at 0 to 0?

Now to the Passing with the Ifs & Butts. I don't know what the Programming is to get the 39% Euchre rate?
Does the Dealer always order with just JH + the upcard? Statistically this is 28%, or 24,080 of the 86,000 hands. How many of those are in that 33,310 hands that your simulator had S4 making trump? I wish I knew!

There are many a player, all the euchre players out there, what number of those will pass that 10H upcard? Is it 50%, 70%? I believe I had a post that it was wise to order with JH + 1, when many doubted it on OE.

Of course S4 will order with JH/9H, Ah/9H, and JH/AH/9H but what does that amount to? Perhaps, 16% + 16% + 1% = 34%. How many of those hands equate to S4 getting euchred 39%? Switching seats, if I am the dealer with JH 10H 9H + AC or other doubleton, it would be difficult for S3 to euchre me. If he trumps the 1st trick and leads the Left, I would play low and not over trump. Would the simulator play this way?

As to S3 scoring a point, consider also, any lead of a Trump (65%) or an Ace of 3 (65%) puts S3 in a good situation.

There are hands I would Pass (like 8 to 8), so it does all depend on score and situation. What is the caliber of who is at S4 & S1? S4 would have to be an aggressive player for me to Pass. And in my thinking, S1 has to be a good player, and aggressive at that! I like things I can control. If I pass, I control little to nothing!

If I knew more about how the Simulator played, I might agree with Passing at this score!

There is more but I will leave it at that!

IRISH

Ray said, Here's what I found (100,000 hands; about 86,000 make it to S3 for a decision):
bid w/ partner: (10,358 / 55,184 / 20,386) EV = +0.41
pass: EV = +0.54 [33,310 bids by S4 = 39%; 24,068 euchres of S4 = 72%]

I found that passing was better, simply because S4 will call and get euchred so often [this may not be true in an actual game! You need to know your opponent] AND partner in S1 has free rein on R2 if S4 passes. It's not bad to bid - the EV is positive - but it could potentially be higher if you pass and the dealer + partner play as expected (perhaps a big "if"?)

The huge caveat here of course is the expectation of how S4 and S1 will play if you pass. In each of the scenarios listed above, there is a positive EV when bidding - it just gets significantly larger if you pass.
Last edited by irishwolf on Wed Jan 26, 2022 12:14 pm, edited 1 time in total.

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Tue Jan 25, 2022 1:06 am

irishwolf wrote:
Mon Jan 24, 2022 11:53 pm
LOL Getting into the Details, so many ifs & butts here: Here my answer.

The original answer to # 11 was: THIS HAND IS TOO WEAK TO ORDER.
I answered otherwise, it is a strong hand. And Ray's simulations Proves that with EV = +0.41
Just to be clear, Rays simulations shows this call has a positive expected outcome, I.E. in the absolute sense it generates points. But this call is -EV according to Rays work becuz passing beats out calling. We don't need to get bogged down in semantics. My claim that "this hand is too weak to order" is correct when we have a good P we can trust. That concept is really what I was shooting for too, the idea that apparently strong hands in S3 are actually "too weak" to order. I bask in the counterintuitive.

Unfortunately my claim is incomplete in a big way becuz we simply don't know if this call is also -EV when we have an amateur as our P. That may remain an open question forever. Ray's simulator can't give us that answer. The onus is on us to find it, and frankly the onus is really on you cuz as long as I have a girlfriend I know I won't make the time to test this spot with an amateur P and I know you just might be crazy enough to do it! If you ever do test this spot under that assumption and it shows that with an amateur P we should make this call at 0-0, I'm pretty sure I would easily accept your results as your test would be the best evidence I have. BTW I'm not saying the burden of proof is on you. I made the claim. The burden of proof is clearly on me. I'm just saying in my current situation I don't care enough to make the necessary time to test this out. Whether you like it or not, you are our only hope Obi Wan Kenobi :) The claim I made at the time was too ambitious and I regret that now.

Overall I really don't see us in much disagreement here. After Rays work, I think we both can see that this hand is a pass with a good P we trust. And you've already seen me walk back my claim with an amateur P.

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Tue Jan 25, 2022 1:22 am

raydog wrote:
Mon Jan 24, 2022 11:00 am
It probably helps a lot that S3 holds the JD, which will be a big boon if partner in S1 bids next aggressively, R2.
Here's another indirect way showing how important holding that JD is. Awhile back Irishwolf demonstrated that it is better to call with this hand from 3S-R1 than pass:

Upcard: (Card_K-H)

(Card_Q-H) (Card_10-H) (Card_9-H) (Card_A-S) (Card_K-S)

I was so skeptical of his finding I basically made a fool out of myself. Then I tested it myself and it was very clear at a greater than 95% confidence interval that this hand was indeed a +EV call, I.E. calling beat out passing. I had to eat a lot of internet crow after that. I'm now so confident that Irishwolf is right about this hand that if your simulator showed this hand to be a pass I would assume there's something wrong with your simulator.

One of the main reasons calling beats out passing with this holding even with an expert P in S1 is becuz this hand hits a Next call so poorly. Not having that JD can indeed change things!

Sometimes an apparent strong holding is "too weak" to call and sometimes an apparent weak holding is "too strong" to pass!!! Gotta love this game.

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Tue Jan 25, 2022 1:51 am

Yes, "Overall I really don't see us in much disagreement here." AGREED!

I would go on to say, that you and I probably are in agreement 95+ percent of the time. I think I could also say, I know of no other player that I agree with more when it comes to euchre . . .
Not an argument, or to be disagreeable. It's just a discussion to get me to the nth degree. It is a part of the game that is as much a Hobby to me, like a Rubik's Cube to solve.

That is what makes me a better player.

Lol, and sometimes I even disagree and question myself.

IRISH

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Thu Jan 27, 2022 3:12 pm

I missed your comment until today. CONGRATULATIONS to you and Ed. But no surprise that you two would do well. And it says a lot, about all those hands and discussions on OE.

COOL, but tell us more!
How many players?
How many games and the format?
What were your game scores?

Me and Edward (Tbolt65) finished first place in the Las Vegas partnership tournament on Saturday.[/i]

Tbolt65
Posts: 820
Joined: Sun Jun 16, 2019 9:14 pm
Location: Las Vegas

Unread post by Tbolt65 » Thu Jan 27, 2022 4:50 pm

irishwolf wrote:
Thu Jan 27, 2022 3:12 pm
I missed your comment until today. CONGRATULATIONS to you and Ed. But no surprise that you two would do well. And it says a lot, about all those hands and discussions on OE.

COOL, but tell us more!
How many players?
How many games and the format?
What were your game scores?

Me and Edward (Tbolt65) finished first place in the Las Vegas partnership tournament on Saturday.[/i]



Thanks Irishwolf.


11 teams
A best 2 of 3 double elimination bracket tournament.
We won all 4 series and all games with-in those series. 8-0 undefeated.
A few close games. No real blow outs. Maybe 1.

Tbolt65
Edward

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Thu Jan 27, 2022 5:05 pm

irishwolf wrote:
Thu Jan 27, 2022 3:12 pm
I missed your comment until today. CONGRATULATIONS to you and Ed. But no surprise that you two would do well. And it says a lot, about all those hands and discussions on OE.

COOL, but tell us more!
How many players?
There were 11 teams, so 22 players.

irishwolf wrote:
Thu Jan 27, 2022 3:12 pm
How many games and the format?
Edward ran the tournament so I don't know the exact details. My job was to show up and not make any mistakes. What I do know is we had to face 4 teams. Each matchup was best 2 out of 3. If anyone lost a matchup they were not out of it. You had to lose 2 matchups to be eliminated. Me and Edward ended up going 8-0. 4 sweeps.
irishwolf wrote:
Thu Jan 27, 2022 3:12 pm
What were your game scores?
We really only had one close matchup between this married couple. Our first game we were up 9-7 on their deal. I donated and stopped a game winning loner and the husband wouldn't stop whining about how that's not right. We shouldn't be allowed to do that, etc. He then asked me sarcastically if I was paid to play euchre implying I'm taking it too seriously. I told him that I was indeed a professional euchre player (obviously I was trolling him as there is no such thing!) We ended up winning that game 10-9 on my deal.

Then next game vs the same couple, we were down 6-4 on my deal, I turned down some garbage red hand, the wife passed on S1, Edward goes alone in black from S2 and gets 4 points. Total game changer. Now we're up 8-6 and the game felt over at that point. I can't remember what that lady had in S1-R2. What I do know is given Edward's hand it's easy to deduce that whatever she had she cannot pass up 6-4. Worst case scenario she gets euchred and it's 6-6 on her deal with 53% equity. She has to gamble in that spot, and based on my hand and Edwards hand, a Next call had to have been pure gold. So most likely a -5 to -6 pt swing for her team becuz she won't call for defensive purposes (I'm sure she doesn't even understand the concept).

By the way two teams whined and bitched about our donates. The above team and my girlfriend's team--it was actually my gf who was doing all the whining for her team saying me and Edward suck the fun out of the game. I wanted to tell her there aint no shame in losing to a legend but I'm sure that wouldve ruined the rest of my night.
irishwolf wrote:
Thu Jan 27, 2022 3:12 pm
Me and Edward (Tbolt65) finished first place in the Las Vegas partnership tournament on Saturday.[/i]

Our big break came when the other expert team, Pat and "Uncle Phil", Lost in the final 4.

(Scroll down if you want to see what they look like: https://www.vegaseuchre.com/hall-of-fame)

I figured we were destined to play them for the championship but they got upset by the team I call the bridge ladies. The bridge ladies are both over 80, they are really good bridge players, and they are not bad euchre players either but they are certainly not close to expert. At least one of them will call as the dealer with R+1+0 which instantly gets some respect from me as so many people still pass that hand. I would say in any game vs Pat and Phil, me and Edward only have around 55% equity. Our edge vs them is very thin. Not having to play them was huge.

As far as any mistakes:

1) Edward ordered me up from S2, I had 3 trump and played my hand in a way to show him I had 3 trump. Later in the hand Edward gambles on a King lead throwing off trying to manufacture 2 pts. At that point tho I still showed him I had a trump left. One should not gamble on a King when their P still has trump, better to trump that king and play the see-saw game. No biggie, just cost us 1 pt.

2) Early in the game, dealer turned down a (Card_Q-H)

I went alone in Next S1-R2 with:

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

And got euchred. I wonder if Ray's simulator would agree with that lone call.

3) Controversial hand: Mid-game, Edward just calls hearts from the dealer spot with this holding:

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

I think that's a loner, but it's never been tested so who knows. The call worked out great tho as I had the Right and the Left was in enemy hands.

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Thu Jan 27, 2022 5:48 pm

Thanks for sharing the tournament. Very interesting. Pictures interesting too. I haven't played a tournament quite like your format. With luck in the game of ~35% the best does not always win!

The whiners on the Donate must have been from Western N.Y., lol.

Not sure about that loner try with the 9H. Opponents had to 2 & 2 or one had 3 as covering your 9H. (Probably bagging on Next.)

I tested similar hand before and loners made was 6 - 8% due to that 9H, any 9X. That's a swing for the fence down by a bunch, IMO.

2) Early in the game, dealer turned down a (Card_Q-H)

I went alone in Next S1-R2 with:

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

And got euchred. I wonder if Ray's simulator would agree with that lone call.

IRISH

User avatar
Dlan
Site Admin
Posts: 672
Joined: Thu Apr 19, 2018 10:08 pm
Location: Ohio

Unread post by Dlan » Thu Jan 27, 2022 6:09 pm

irishwolf wrote:
Thu Jan 27, 2022 3:12 pm

Me and Edward (Tbolt65) finished first place in the Las Vegas partnership tournament on Saturday.[/i]


Congratulations to Wes And ED on their win at Las Vegas

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Thu Jan 27, 2022 7:38 pm

Back to the Loner. With this hand, if both opponents have 2 & 2, you have to lead trump twice to knock out the trumps. But if you lead twice and one opponent has 3, you are Dead on arrival unless they lead to your AC.
About 50% of the time, opponents will have trumps 1 & 1, 0 & 1 or 1 & 0 which you deplete their trump(s). But any two trumps will stop your loner, also about 50%. Even then you now have to contend with getting by with that 9H. If I were one of the opponent, I know you are block Hearts, thus, I am saving a Heart regardless of that AC lead.
Not sure how that simulator would deal with that situation? Would it play optimally in a Squeeze? What makes it tough are those border line H's calls which the opponent now bags on Next. Who will go there?
However, to continue, I see it less than 8% success rate with a Euchre rate of about 8%. Equal to your success rate, but what about those sweeps? One can only estimate at about 20% - 25%.

So run your Simulator Ray!

IRISH


2) Early in the game, dealer turned down a (Card_Q-H)

I went alone in Next S1-R2 with:

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

And got euchred. I wonder if Ray's simulator would agree with that lone call.

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Thu Jan 27, 2022 7:52 pm

Dlan wrote:
Thu Jan 27, 2022 6:09 pm
irishwolf wrote:
Thu Jan 27, 2022 3:12 pm

Me and Edward (Tbolt65) finished first place in the Las Vegas partnership tournament on Saturday.[/i]


Congratulations to Wes And ED on their win at Las Vegas


Thanks!! Would've been more fun if somehow you and Wolf came as partners. One can dream!

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Thu Jan 27, 2022 8:04 pm

irishwolf wrote:
Thu Jan 27, 2022 7:38 pm
Back to the Loner. With this hand, if both opponents have 2 & 2, you have to lead trump twice to knock out the trumps. But if you lead twice and one opponent has 3, you are Dead on arrival unless they lead to your AC.
About 50% of the time, opponents will have trumps 1 & 1, 0 & 1 or 1 & 0 which you deplete their trump(s). But any two trumps will stop your loner, also about 50%. Even then you now have to contend with getting by with that 9H. If I were one of the opponent, I know you are block Hearts, thus, I am saving a Heart regardless of that AC lead.
Not sure how that simulator would deal with that situation? Would it play optimally in a Squeeze? What makes it tough are those border line H's calls which the opponent now bags on Next. Who will go there?
However, to continue, I see it less than 8% success rate with a Euchre rate of about 8%. Equal to your success rate, but what about those sweeps? One can only estimate at about 20% - 25%.

So run your Simulator Ray!

IRISH


2) Early in the game, dealer turned down a (Card_Q-H)

I went alone in Next S1-R2 with:

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

And got euchred. I wonder if Ray's simulator would agree with that lone call.
As far as how I played the hand: Obviously I led the Right on first street and the 2 seat showed void in trump and S4 played the 9D. So right away I'm like "oh god". I led my ace on 2nd street and it walked. Then I bit the bullet and led trump again hoping S4 did not start with 3 (I'm not gonna play my 9H there cuz that's just totally giving up, I'd rather hide my loser and take the risk). Well S4 started with AdKd9d, so she cleaned me out after I led trump on 3rd street. S4 also had the Kh, so there was no way I could escape.

Part of my reasoning for "going for it" was my loser was the turned down suit so the enemy is less likely to have it plus I was also counting on the fact that even if they had it they might incorrectly get rid of that suit before 5th street. You are right Wolf that people should tend to save a heart in that spot but I was counting on them not knowing that. In this case it didn't matter as S4 was loaded.

But yea regardless it probably wasn't a +EV loner.

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Thu Jan 27, 2022 9:23 pm

Wes (aka the legend) wrote:
Thu Jan 27, 2022 7:52 pm
Dlan wrote:
Thu Jan 27, 2022 6:09 pm
irishwolf wrote:
Thu Jan 27, 2022 3:12 pm

Me and Edward (Tbolt65) finished first place in the Las Vegas partnership tournament on Saturday.[/i]


Congratulations to Wes And ED on their win at Las Vegas


Thanks!! Would've been more fun if somehow you and Wolf came as partners. One can dream!


BTW Ray is coming to Vegas in March. He should feel morally obligated to come to our Thursday tournament so me and Edward can meet him. I'm sure he'll be there. Can't wait to meet him!

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Thu Jan 27, 2022 10:16 pm

To Ed & Wes,

If you hold a tournament sometime in the months of June to end of September (4 month window), I could make plans to attend. Not sure who my partner would be, but can work on that in the interim. Need some advance notice to plan as this is about visiting some relatives out West too.

Who knows maybe get some of these other OE participates to attend as well!

And just maybe have some additional Euchre sessions as well! Now that would be fun. Play all day, Play all Night!

"Thanks!! Would've been more fun if somehow you and Wolf came as partners. One can dream!"

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Tue Apr 19, 2022 10:51 am

Long period of silence on the OE forum, not sure why. Personally, I have a lot of fodder to tweak and improve my program, and have been doing so. Am just now looking at this thread (I didn't realize it had been updated with additional posts), I'd like to belatedly congratulate Wes and Ed on their tournament victory!

I see lots of hands to test in this thread, and please understand that this is as much an opportunity for me to improve my simulator as to provide insight into the correct play. So I will continue work on my program, with your valuable feedback, and report back if I find significant results. It's a two-way street, and I appreciate you helping me provide a tool which may eventually guide or support your current strategy of play.

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Wed Apr 20, 2022 9:58 pm

I'd like to respond to a few questions / comments from this thread (with much delay), regarding Wes' quiz question #11.

Irish: I can give you an idea when my program will have S4 call trump (considering S4's hand after picking up the turn and discarding).
* always with 3+ trump
* never with 1 trump (even R + 3 off-suit aces - I found it better to pass in that situation)
* with 2 trump: - will pass w/ R+1 IF all other suits are blocked, else will call;
- with A+1, will pass with 2 off-suit aces OR if 4-suited (which pretty much means 2 off-suit aces),
else will call;
- with K+1, will call if 2-suited AND have an off-suit A, else will pass;
- with x-x trump, will call with an off-suit green A OR w/ 3 off-suit Aces, else will pass

All this is internally consistent with how my program currently has other seats bidding and how the cards are played (generates the best EV). If other parts of the program are not playing optimally, these criteria could change slightly as I tweak other parts of the program. But those are the criteria currently.

Also, if S3 calls trump, S1 will lead trump (best trump if they hold a bower OR 3+ trump, else worst trump), else will generally lead an A, else will lead their worst card (trying to avoid the turned suit). General guideline.

I agree that the decision to bid or pass [S3, with the hand given in quiz question #11] is hugely dependent on how S4 and S1 play. I can only assume, for the purposes of my program, that every player is trying to optimize EV, but that is not the case in real life, either intentionally or out of lack of experience. So there's always that big caveat to my results.

Wes: you cited the following hand: S3 holds Q-10-9H + A-KS (KH turned), and said that it is better to call than to pass. My program currently passes the hand, but when I test it I found it is better to call. Here are some more details of what I found.

tested 100,000 hands, fixed S3 hand + turn, other 18 cards random. About 86% of hands make it to S3 for a decision (of the other 14%, most are calls by S2).

if S3 calls: (16,609 / 44,899 / 25,251) [sweep / 1 pt / euchred] EV = +0.32
if S3 passes: S4 calls 37,325 hands for an EV of -0.24
S1,R2 calls 41,231 hands (65% of those in next) for an EV of +0.32
S2,R2 calls 6,823 hands for an EV of +0.79
S3,R2 calls 1,380 hands for an EV of -0.07
overall EV = +0.19 (for S1/S3)

So if the bogey is to get that EV of +0.32 (achieved by bidding), S1/S3 fall short when S4 bids (only +0.24), match it when S1 bids (S1 actually holds the L 58% of the time and the R 49% of the time they call next, so not such a disadvantaged situation), and fall well short the minority of time S2 calls, R2 (-0.79) or S3 calls, R2 (-0.07). Thus, better to call, R1.

Now I need to generalize this situation in order to describe in what situations S3 should call; my program is clearly not using the right criteria.

You also suggested a couple of other hands from your tournament with Ed, and wondered if calling alone was warranted.

S1 holds Q-J-10D + AC + 9H (QH), looking at R2 next call (40% of hands make it to R2).
alone: EV = 1.38 [loner success rate = 15%]
wp: EV = +1.23
The problem with this result is that I need to optimize my program for which cards the opponents hold to the end when bidding alone. My intention is simply to keep track of how many cards can be beaten, but there is a deeper problem with that.
Say S1 bids alone (above hand), and S4 has both the AS and the QH, both singletons. The AS beats 5 other cards, while the QH beats only 2 other cards. S4 will keep track of which cards are played leading up their turn on the 4th trick and reevaluate how many remaining cards can be beaten by the AS and the QH. If, say, the K and 9S are played some point, but no hearts, that would still leave 3 cards which the AS beats and only 2 cards that the QH beats, so it should be statistically better to hang on to the AS. One could argue that S1 is more likely to go alone with a low H (next) card than a low S (green) card, but that is something I would need to program in, so I would be intentionally biasing the results! If I find that a lone call with a singleton 9 in next suit gives better relative results than a lone call with a singleton J in a green suit, then perhaps I do indeed to program that in. But then the opponents will become aware of that bias [not through the program "cheating" and letting them see the opponents cards, but through statistical outcomes over thousands of games], and modify the thinking that "my AS beats 5 other cards", because S1 will not go alone holding a J, 10 or 9S, so it really only beats 2 cards. But if the opponents start holding the QH rather than the AS as the last card, that will make going alone with the JS or 10S more attractive, and change how S1 bids alone.....
It's a bit of a pickle.

At any rate, I'm going to have a fresh look at which cards the defensive holds to the end when defending against a lone (in my program). And I would say that it was not crazy for Wes to bid this hand alone.

The other hand to examine was S4 holding A-K-QH + AC + AS [I assumed QH turned, and S4 discarded the 9D to end up with that hand]. Testing 100,000 hands, I found:
bid alone: EV = +1.18
bid wp: EV = = +.99
In this instance I had a look at how some of the hands were played and found some choices which I think were poor, but I need to investigate further. My best conclusion here is once again that calling alone wasn't so crazy, but I can't say definitively that it was better - my results may be skewed by incorrect play on a minority of crucial hands.

Tbolt65
Posts: 820
Joined: Sun Jun 16, 2019 9:14 pm
Location: Las Vegas

Unread post by Tbolt65 » Wed Apr 20, 2022 11:16 pm

Dlan wrote:
Thu Jan 27, 2022 6:09 pm
irishwolf wrote:
Thu Jan 27, 2022 3:12 pm

Me and Edward (Tbolt65) finished first place in the Las Vegas partnership tournament on Saturday.[/i]


Congratulations to Wes And ED on their win at Las Vegas


Thanks Dlan.

Tbolt65
Edward

Tbolt65
Posts: 820
Joined: Sun Jun 16, 2019 9:14 pm
Location: Las Vegas

Unread post by Tbolt65 » Wed Apr 20, 2022 11:21 pm

irishwolf wrote:
Thu Jan 27, 2022 10:16 pm
To Ed & Wes,

If you hold a tournament sometime in the months of June to end of September (4 month window), I could make plans to attend. Not sure who my partner would be, but can work on that in the interim. Need some advance notice to plan as this is about visiting some relatives out West too.

Who knows maybe get some of these other OE participates to attend as well!

And just maybe have some additional Euchre sessions as well! Now that would be fun. Play all day, Play all Night!

"Thanks!! Would've been more fun if somehow you and Wolf came as partners. One can dream!"

Yeah I definitely want to get some more tournament going for teams. I think I'm going to try for either June or July. I have to come up with a format first before I set anything in stone with store. I gathered there was some unhappy campers at my best 2 of 3 double elimination tournament style bracket. So I'll have to come up with something that people will come to again. Don't worry I will NEVER make it 8hands X-rounds points only euchre. That's straight up garbage. Oh I'll play it if someone runs o ne like it because I love euchre that much but you'll never see me run one of those. It will always be up to 10points full game and tie breakers if any will always be played out, It will never go to points.


Tbolt65
Edward

Tbolt65
Posts: 820
Joined: Sun Jun 16, 2019 9:14 pm
Location: Las Vegas

Unread post by Tbolt65 » Wed Apr 20, 2022 11:38 pm

Bidding alone in Seat 1, you can be more liberal in doing so. You have the first lead which is so very important. Plus if you have that Js or 10s either turn down suit or not. You put the squeeze on your opponents to potentially throw the wrong card away and that's what you want. All that aside. Going alone with say j-10-9 for loser in the first seat Has a better shot of making it especially if it's the turned down suit. However say spade was turned down and you have x-x-x trump Ac 9h. I will not being going alone. Ever. Playing for two no matter how far your down is just the better overall strategy.

Also, From 3rd, 2nd or 4th 9.9 times out of 10. I'm taking my partner to ensure two points with those losers suits. I have gone alone but it is important to maximize consistent points instead of just settling for 1point the majority of the time while only get a loner some of the time.


Tbolt65
Edward

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Fri Apr 22, 2022 12:20 am

RAY,
RE: Loner or Not? (Card_A-H) (Card_K-H) (Card_Q-H) (Card_A-S) (Card_A-C)

To your Simulator results for ..."The other hand to examine was S4 holding A-K-QH + AC + AS [I assumed QH turned, and S4 discarded the 9D to end up with that hand]. Testing 100,000 hands, I found:
bid alone: EV = +1.18
bid wp: EV = = +.99
In this instance I had a look at how some of the hands were played and found some choices which I think were poor, but I need to investigate further. My best conclusion here is once again that calling alone wasn't so crazy, but I can't say definitively that it was better - my results may be skewed by incorrect play on a minority of crucial hands."

After reading it I was not going to respond as I also had worked on this hand. However, be that as it may, I said what the h e l l, post my results because it's fun to do. My calculations are for going alone (Statistically - law of large numbers):

Loner Points: 104
Euchres: 18 to 22 (taking the higher number)
1 pointers: 52
EV = + 116 (but the range is 116 to 120)

How did I arrive at that number, and I don't need a Simulator. Both Jacks void simultaneously for S1 & S3 is 26%. So this is the Loner rate 26 x 4 = 104. It could be slightly lower if S1 leads to S3 void in AC or AS but consider 3 suits to choose from, etc. etc. And the Euchre rate is when both Jacks at one Opponent (13%), Plus one opponent has Bower + 2 other trumps or all four trumps 9 to 10% in total. However, must lead to the void Diamonds or if Clubs is led - lead clubs. Makes me wonder if the simulator considers the suit to lead to? Anyway, it is highly variable and I estimated it at 9 to 11% so a total of 23% but about half will result in Euchres. Euchre rate 9 to 11%. Euchre is always variable as to what card to lead.
----------------

Now for taking my partner along for the ride:

Marches: 40 to 44 (I used 42) +84
Euchres: 12 to 14 (I used 13) -26
1 pointers: 45 (+45)
EV = + 103 (BUT IT COULD BE LOWER)

How did I arrive at a lower EV? S2 has to have JH or JD and JH is buried. That number is 28% + 16.7% = 44.7% for sweeps. But you have to adjust for S1 leading to S3 void in that suit. There are 3 suits so its a guess. So it could be slightly lower (3% or so). Then using the same logic as above for both bowers or Bower + 2 factoring in that S2 can help with a trump or the Diamond void suit. So the Euchre rate will be lower by around 8 to 10% taking your partner along. However, the EV will never exceed going alone.

So my results is quite similar to your Simulator for what it is worth. And doing hands does not satisfy this type of euchre problem. But that is not end of story as to go alone or not. I suggest to go alone rests on the score & situation. Do you need a loner? Are you behind in score by 3 or more points? Maybe the score is 7 to 7 and you want 2 points and not risk the higher euchre rate? If the score is 0 to 0, maybe so! But the bottom line in the Law of Large Numbers, taking your partner along will Never exceed going alone.

IRISH

Tbolt65
Posts: 820
Joined: Sun Jun 16, 2019 9:14 pm
Location: Las Vegas

Unread post by Tbolt65 » Fri Apr 22, 2022 2:03 pm

Tbolt65 wrote:
Wed Apr 20, 2022 11:21 pm
irishwolf wrote:
Thu Jan 27, 2022 10:16 pm
To Ed & Wes,

If you hold a tournament sometime in the months of June to end of September (4 month window), I could make plans to attend. Not sure who my partner would be, but can work on that in the interim. Need some advance notice to plan as this is about visiting some relatives out West too.

Who knows maybe get some of these other OE participates to attend as well!

And just maybe have some additional Euchre sessions as well! Now that would be fun. Play all day, Play all Night!

"Thanks!! Would've been more fun if somehow you and Wolf came as partners. One can dream!"

Yeah I definitely want to get some more tournament going for teams. I think I'm going to try for either June or July. I have to come up with a format first before I set anything in stone with store. I gathered there was some unhappy campers at my best 2 of 3 double elimination tournament style bracket. So I'll have to come up with something that people will come to again. Don't worry I will NEVER make it 8hands X-rounds points only euchre. That's straight up garbage. Oh I'll play it if someone runs o ne like it because I love euchre that much but you'll never see me run one of those. It will always be up to 10points full game and tie breakers if any will always be played out, It will never go to points.


Tbolt65
Edward


Ok here is the link, Irishwolf. It's official too.

viewtopic.php?f=17&t=862


Tbolt65
Edward

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Fri Apr 22, 2022 6:29 pm

Irish, I'm glad to see you arrived at nearly the same results simply by examining the probabilities. I consider that some validation of my simulator. I actually found a slightly higher euchre rate when taking partner along, which doesn't make sense, but also found a couple of scenarios where the wrong play of the cards was made. Correcting those will bring my EV more in line with what you predicted.

Returning to the question of S3 bidding R1, I tested a dozen hands and found it quite rare that S3 should call with a not obvious strong hand, like the one posted [Q-10-9H + A-KS, KH turned]. Even when bidding is EV+, there are so many ways to earn even MORE points by passing (euchring S4 when they call, R1; supporting S1 when they call next, R2; euchring S2 (and supporting S1) when they call anything, R2, by virtue of having 2 or 3 off-suit aces) that even some fairly strong hands are better passed.

One example is A-Q-10H + AS + AC (9H turned). I find an EV of +0.66 when bidding but and EV of +0.84 when passing: lots of euchres when S4 bids; lots of sweeps and few euchres when partner bids, R2.

The scenario I narrowed down to was:
- 3 trump
- 2-suited
- an off-suit green ace
- NOT holding R or A trump (so basically, any 3 of K-Q-10-9 trump)
(and by the way, always pass when R is turned; otherwise the turn card isn't a factor)

Bidding with these hands resulted in an EV boost of +0.14 on average.

As always, this result presumes that S4 and S1 will bid rationally (optimizing EV), and that the score isn't a factor.

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Fri Apr 22, 2022 8:32 pm

RAY,
1) I feel the need to respond. Not so much on the first except about the euchre rate in that it requires S2 to be of no help (in Diamonds), and that it an opponent has to be loaded with Diamonds or doubleton Clubs or Spades. Particularly both bowers waiting on Next. Still a ~13% euchre rate and about 8% better than going alone. That trivial tho and glad we are in close agreement. If we differ by 10% then something is up.

"I consider that some validation of my simulator. I actually found a slightly higher euchre rate when taking partner along, which doesn't make sense, but also found a couple of scenarios where the wrong play of the cards was made. Correcting those will bring my EV more in line with what you predicted."

2) This one is one was my hand that Wes & I battled over with many exchanges. I did many hands and analysis before posting. I could not get a better EV by S3 passing. I think I recall about 58% that R2S1 calling trump. That is worrisome for me and so much out of my control if at S3. S3 passes, he can only help in Spades. In addition, if S1 calls he basically is on his own and has to have 2.75 tricks. Euchre rate will be high for him. I guess we will just have to agree to disagree because that is a 'biddable hand' . So I disagree that Passing is Better than S3 ordering. Especially, if you have a 'passive' partner. I would have to go back to my data for specifics
. "Returning to the question of S3 bidding R1, I tested a dozen hands and found it quite rare that S3 should call with a not obvious strong hand, like the one posted [Q-10-9H + A-KS, KH turned]. Even when bidding is EV+, there are so many ways to earn even MORE points by passing (euchring S4 when they call, R1; supporting S1 when they call next, R2; euchring S2 (and supporting S1) when they call anything, R2, by virtue of having 2 or 3 off-suit aces) Moving the goal posts now. that even some fairly strong hands are better passed."

3) You jump to this hand. 1. & 2. are hard wired for me but this example I am more fixated on your comments. For me, we now have apples and oranges comparing to (2) above.
Why? Because now you have TWO singleton aces instead of an Ace dbltn. Even replace one with Next Ace. S1 can call anything and you will score points. If the Dealer orders he in big trouble. And you also have S2 blocked with a potential euchre. Of course here the decision is PASS with two Aces. So we are in agreement about IF TWO ACES, but that is not the Hand: QH 10H 9H AS KS.
Pick another combination and I might agree.


"One example is A-Q-10H + AS + AC (9H turned). I find an EV of +0.66 when bidding but and EV of +0.84 when passing: lots of euchres when S4 bids; lots of sweeps and few euchres when partner bids, R2."

IRISH
Last edited by irishwolf on Fri Apr 22, 2022 11:54 pm, edited 1 time in total.

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Fri Apr 22, 2022 10:24 pm

Sorry, Irish, regarding point #2 my comment wasn't clear, as I can now tell rereading it. I DO agree with you that Q-10-9H + A-KS (KH turned) should be bid by S3, as I initially said when I did the simulation. I went further to try and discover how many similar hands might exist, and I simply meant to say that there aren't many of them. Only a small pool with very particular characteristics, which I then delineated. Yes, I DO then move the goal posts to note that seemingly stronger hands should actually be passed, because the R2 points + defensive points are so much higher. So I think we are in agreement on those points.

Wes (aka the legend)
Posts: 1541
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Apr 25, 2022 2:40 am

raydog wrote:
Wed Apr 20, 2022 9:58 pm
I'd like to respond to a few questions / comments from this thread (with much delay), regarding Wes' quiz question #11.

Irish: I can give you an idea when my program will have S4 call trump (considering S4's hand after picking up the turn and discarding).
* always with 3+ trump
* never with 1 trump (even R + 3 off-suit aces - I found it better to pass in that situation)
That's interesting. Whether I order with R+0+3A depends on what I have blocked or not. If I have a hand like this (assume score is 0-0):

Upcard: (Card_J-S)

Dealer hand: (Card_A-C) (Card_A-D) (Card_K-D) (Card_A-H) (Card_10-H)

I would pick up given that I block no suits. I think it's plausible that passing could beat out calling given that having 3 off aces is still pretty good defense. If your simulator shows it's better to pass with that specific hand then it's pretty much game over for all R+0+3A configurations as far as calling. Altho even if that were the case I'd still make that care in situations where my team has a nice lead.

For the record I would also pick up with this type of R+0+2A hand:

Upcard: (Card_J-S)

Dealer hand: (Card_A-D) (Card_K-D) (Card_A-H) (Card_10-H) (Card_9-C)

Same story, if the simulator shows that passing beats out calling with that specific hand it's hard to imagine a +EV call from the R+0+2A configuration.

I also call with R+0+A if I block no suits. Example hand:

Upcard: (Card_J-S)

Dealer hand: (Card_A-D) (Card_Q-H) (Card_10-H) (Card_9-H) (Card_9-C)
raydog wrote:
Wed Apr 20, 2022 9:58 pm
* with 2 trump: - will pass w/ R+1 IF all other suits are blocked, else will call;
I agree with that approach, but that has to be proven. I.E. take the best R+1+0 hand you can think of with all suits blocked. If passing beats out calling then it's a pretty safe bet that's always gonna be the case. For example, maybe a hand like this (spades is trump):

Upcard: (Card_A-S)

Dealer hand: (Card_J-S) (Card_J-D) (Card_K-H) (Card_10-C) (Card_9-C)

If passing beats out calling with that hand, then it's pretty much game over for calling with R+1+0 when we have all suits blocked.

raydog wrote:
Wed Apr 20, 2022 9:58 pm
- with A+1, will pass with 2 off-suit aces OR if 4-suited (which pretty much means 2 off-suit aces),
else will call;
- with K+1, will call if 2-suited AND have an off-suit A, else will pass;
- with x-x trump, will call with an off-suit green A OR w/ 3 off-suit Aces, else will pass
I think we've already established that 3 suited marginal 2 trump + an Ace hands are +EV calls when we block no suits for the 2nd round. See hand 2 in this very quiz.
raydog wrote:
Wed Apr 20, 2022 9:58 pm
All this is internally consistent with how my program currently has other seats bidding and how the cards are played (generates the best EV). If other parts of the program are not playing optimally, these criteria could change slightly as I tweak other parts of the program. But those are the criteria currently.
Overall looks pretty good to me. I don't expect perfection.
raydog wrote:
Wed Apr 20, 2022 9:58 pm
Also, if S3 calls trump, S1 will lead trump (best trump if they hold a bower OR 3+ trump, else worst trump), else will generally lead an A, else will lead their worst card (trying to avoid the turned suit). General guideline.
If S2 has A-X in trump, they should always lead the Ace in this spot. And yeah if S1 has no trump then off aces are the next best lead, and if S1 has no trump no aces, I would recommend S1 lead his cleanest suit. For example if S3 calls trump (spades) and S1 has this hand:

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

S1 should lead the 9D. The idea for S1 here is to try to minimize the probability of S3 getting overtrumped on 1st street.
raydog wrote:
Wed Apr 20, 2022 9:58 pm
I agree that the decision to bid or pass [S3, with the hand given in quiz question #11] is hugely dependent on how S4 and S1 play. I can only assume, for the purposes of my program, that every player is trying to optimize EV, but that is not the case in real life, either intentionally or out of lack of experience. So there's always that big caveat to my results.
True. We always have to adjust for an amateur in S1. How much to adjust is always the question and no simulator can answer that definitively. Your results on these S3 calls/passes are still vitally important tho as it will aid us in this judgement process.
raydog wrote:
Wed Apr 20, 2022 9:58 pm
Wes: you cited the following hand: S3 holds Q-10-9H + A-KS (KH turned), and said that it is better to call than to pass. My program currently passes the hand, but when I test it I found it is better to call. Here are some more details of what I found.
I would call that the Irishwolf hand since he's the first to bring this spot up :)
raydog wrote:
Wed Apr 20, 2022 9:58 pm
tested 100,000 hands, fixed S3 hand + turn, other 18 cards random. About 86% of hands make it to S3 for a decision (of the other 14%, most are calls by S2).

if S3 calls: (16,609 / 44,899 / 25,251) [sweep / 1 pt / euchred] EV = +0.32
if S3 passes: S4 calls 37,325 hands for an EV of -0.24
S1,R2 calls 41,231 hands (65% of those in next) for an EV of +0.32
S2,R2 calls 6,823 hands for an EV of +0.79
S3,R2 calls 1,380 hands for an EV of -0.07
overall EV = +0.19 (for S1/S3)

So if the bogey is to get that EV of +0.32 (achieved by bidding), S1/S3 fall short when S4 bids (only +0.24), match it when S1 bids (S1 actually holds the L 58% of the time and the R 49% of the time they call next, so not such a disadvantaged situation), and fall well short the minority of time S2 calls, R2 (-0.79) or S3 calls, R2 (-0.07). Thus, better to call, R1.
Good stuff. This backs up Wolf's results and my results. So now we've tested this spot in 3 independent ways and it's a call. That's as close to a "proof" as we'll ever get. And YET what makes this game so fun and interesting is there will still be times me and Wolf will pass this very hand!! Imagine our team is down 9-6. I'd rather take the small EV hit and pass hoping my team gets lucky with either euchring the dealer or winning the game on a S1-R2 loner. I'm pretty sure Wolf would pass this hand in this spot too. And I don't think there's a way for a simulator to prove this pass is right or wrong. The idea that making a -EV play will increase our team's chances of winning the game is a controversial one but I think it's valid.
raydog wrote:
Wed Apr 20, 2022 9:58 pm
Now I need to generalize this situation in order to describe in what situations S3 should call; my program is clearly not using the right criteria.
My take would be this:

Assuming S3 has no bowers in their hand they should always call with 3 trump + an outside green suited ace except vs a Jack upcard, then bag. If S3 has an outside suited Next ace then S3 should pass and bag. For example:

Upcard: (Card_9-S)

S3 has: (Card_K-S) (Card_Q-S) (Card_10-S) (Card_A-C) (Card_K-C)

S3 should pass. Becuz S3 now hits a Next call decently I believe passing will beat out calling BUT this is ONLY the case when S1 is an expert. If S1 is an amateur then I think we should call this.

What about a hand like this:

Upcard: (Card_9-S)

S3: (Card_K-S) (Card_Q-S) (Card_10-S) (Card_A-D) (Card_K-H)

I think this hand is now a pass. Without that outside suited ace, I believe passing wins out now.
raydog wrote:
Wed Apr 20, 2022 9:58 pm
You also suggested a couple of other hands from your tournament with Ed, and wondered if calling alone was warranted.

S1 holds Q-J-10D + AC + 9H (QH), looking at R2 next call (40% of hands make it to R2).
alone: EV = 1.38 [loner success rate = 15%]
wp: EV = +1.23
The problem with this result is that I need to optimize my program for which cards the opponents hold to the end when bidding alone. My intention is simply to keep track of how many cards can be beaten, but there is a deeper problem with that.
Say S1 bids alone (above hand), and S4 has both the AS and the QH, both singletons. The AS beats 5 other cards, while the QH beats only 2 other cards. S4 will keep track of which cards are played leading up their turn on the 4th trick and reevaluate how many remaining cards can be beaten by the AS and the QH. If, say, the K and 9S are played some point, but no hearts, that would still leave 3 cards which the AS beats and only 2 cards that the QH beats, so it should be statistically better to hang on to the AS. One could argue that S1 is more likely to go alone with a low H (next) card than a low S (green) card, but that is something I would need to program in, so I would be intentionally biasing the results! If I find that a lone call with a singleton 9 in next suit gives better relative results than a lone call with a singleton J in a green suit, then perhaps I do indeed to program that in. But then the opponents will become aware of that bias [not through the program "cheating" and letting them see the opponents cards, but through statistical outcomes over thousands of games], and modify the thinking that "my AS beats 5 other cards", because S1 will not go alone holding a J, 10 or 9S, so it really only beats 2 cards. But if the opponents start holding the QH rather than the AS as the last card, that will make going alone with the JS or 10S more attractive, and change how S1 bids alone.....
It's a bit of a pickle.
I don't think it's a pickle worth worrying about. I bet your simulator plays this spot close enough to optimally. It doesn't really have to be perfect to capture reality well becuz humans--even expert ones--aren't perfect either. Given the decent gap in EVs I feel comfortable now with the idea that going alone is best with this hand. Thanx for doing this Ray!
raydog wrote:
Wed Apr 20, 2022 9:58 pm
The other hand to examine was S4 holding A-K-QH + AC + AS [I assumed QH turned, and S4 discarded the 9D to end up with that hand]. Testing 100,000 hands, I found:
bid alone: EV = +1.18
bid wp: EV = = +.99
In this instance I had a look at how some of the hands were played and found some choices which I think were poor, but I need to investigate further. My best conclusion here is once again that calling alone wasn't so crazy, but I can't say definitively that it was better - my results may be skewed by incorrect play on a minority of crucial hands.
Not surprised at this result. I strongly suspected going alone was better and that's a pretty big gap so I think we should all feel confident that is the case. We don't need ironclad proof, just solid evidence and the above meets that criterion. Now the question is what about this hand:

S4: (Card_A-H) (Card_K-H) (Card_Q-H) (Card_A-C) (Card_K-C)

Is it better to go alone with that hand? and if the answer is yes then we need to see if that's the case for all outside suited aces. I.E. is this hand also a loner:

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

My prediction: AhKhQh-AcKc will be a loner but AhKhQh-AcQc will not be a loner.

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Mon Apr 25, 2022 11:57 am

Irish, I am only replying because I don't want to leave you hanging. Regarding point #3, 3 posts ago, you claim I am comparing apples and oranges, but I see us as being in perfect agreement, and am unsure why you bothered posting that comment.

Like Wes, I look at these specific hands and am naturally drawn to similar hands (sometimes with "critical" - i.e., apples and oranges - differences), because I want to know how to program my simulator to play better (Wes just wants to play better himself). My fundamental tenet was "some S3 hands don't look strong enough to bid but actually are". I am simply contending that the hand I cited looks strong enough to bid, but should actually be passed - and you agree with that. It supports my tenet, and we arrive at the same conclusion - why do you insinuate that "you don't agree"? [you say "if you pick another combination I might agree" insinuating that you don't agree with this example - but we DO agree!]

I am confused.

I am certain this entirely a semantic argument, but just want to highlight it so we understand that there is sometimes a communication gap between us. I value your insight and would love to communicate better! I take absolutely no offense at your comment, I am just trying to parse what I need to respond to, and this one seemed like you wanted a response (from the wording) but also like no response was necessary (since we were in complete agreement).

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Mon Apr 25, 2022 12:02 pm

On another topic, I revisited the hand where S4 holds AS + A-KH + 9D + AC (QH turned) - final holding for S4 is AS + A-K-QH + AC - after having tweaked my program for some poor play of the hand.

I now find an EV of +1.16 when bidding alone, and and EV of +1.03 when bidding with partner. Which is exactly the results Irish predicted. Uncanny... [or is it?]

Nice work, Irish!

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Apr 25, 2022 12:45 pm

Ray,

RE: About "I am confused"

You said, " Returning to the question of S3 bidding R1, I tested a dozen hands and found it quite rare that S3 should call with a not obvious strong hand, like the one posted [Q-10-9H + A-KS, KH turned]. Even when bidding is EV+, there are so many ways to earn even MORE points by passing (euchring S4 when they call, R1; supporting S1 when they call next, R2[/b] " I thought you were saying that Passing was just as good or better - which I would disagree.

Thus, I said what I did because I thought you were putting the hand with QH 10H 9H AS/KS in the same category or even suggesting one 'might' pass with this hand. I clearly thought no way pass on the hand but indeed pass with two singleton Aces. I focused on the statement above. So who knows maybe you had other hands that I might agree with your analysis.

However, we in agreement. Let's don't over read into this.
it's all good, move on!


IRISH

irishwolf
Posts: 1319
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Apr 25, 2022 1:21 pm

RE: I now find an EV of +1.16 when bidding alone, and and EV of +1.03 when bidding with partner. Which is exactly the results Irish predicted. Uncanny... [or is it?]

I don't want to misread, but if any way that suggests just Luck in analysis, NO Way! i spent a fair amount of time capturing all the Variables in that hand in my Statistical Calculations. There are a couple variables that might alter the results +/- 3% over the course of the LAW OF LARGE NUMBERS (your 100,000 or 1,000,000) to infinity. It just confirms that I can do this with accuracy. But it also confirms you Simulator. There are some hands that requires confirmation (for me anyway) as calculations of critical play cannot be captured. I think that for your Simulator as well. For me I have to do hands to see the correct play. I know, I am OCD about this. However, for example the hand 10S up, S2 has KS 9S KH KD 10d. S4 has AS 10S AC QH 9H or AS 10s AD QH 9H - or even JS 10S AC QH 9H weak hands. Your analysis with trump high lead low is straight forward. But leading to S2 void when sloughing KH I have no confidence in your EV posted. In fact, KH slough is better!

Sure stronger hands with 3 trumps, both bowers, about the same or better. But what is critical is winning a point with Marginal hands. But that is just me. It matters little I differ with anyone on this.

IRISH

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Mon Apr 25, 2022 3:40 pm

Irish, all good on the first point. I misled YOU (unwittingly) with the way I stated my premise. We are in agreement, move on.

On the second point, the uncanny point was not that our numbers were essentially the same (you are calculating the mathematical probabilities, and they don't lie), but the fact that they were exactly the same, to 2 decimal points! All sorts of minor adjustments to make for rare cases, so a real coincidence to get perfectly the same answer. Just musing that inconsequential fact. No surprise that your analysis was correct!

raydog
Posts: 260
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Mon Apr 25, 2022 4:29 pm

Ah, Wes, for every response you always have 3 new questions!

Regarding the hands you proposed in you latest post: S = simulator; W = Wes

S4, A-10H + A-KD + AC (JS)
pass: EV = +0.40 S
bid: EV = +0.35 W

S4, A-10H + A-KD + 9C (JS)
pass: EV = +0.06 S
bid: EV = -0.06 W

S4, Q-10-9H + AD + 9C (JS)
pass: EV = -0.44 S
bid: EV = -0.38 W

S4, JS + KH + JD + 9-10C (AS)
pass: EV = +0.24
bid: EV = +0.33 S
[I should correct a previous error by saying here that my program will pass with R+1 trump and all suits blocked IF the 2nd trump is K or lower, otherwise bid; so here bids because the 2nd trump is an A]

S4, JS + KH + JD + 9-10C (KS)
pass: EV = +0.26 S
bid: EV = +0.23
_______________________________________

S3, K-Q-10S + A-KC (9S)
pass: EV = +0.49 S, W
bid: EV = +0.40

S3, K-Q-10S + AD + KH (9S)
pass: EV = +0.21 S, W
bid: EV = +0.12
________________________________________

S4, A-KH + A-KC + 9D (QH)
alone: EV = +0.74 W
wp: EV = +0.81 S

S4, A-KH + A-9C + 9D (QH)
alone: EV = +0.49
wp: EV = +0.81 S, W

All this analysis basically shows that my program is generally internally consistent - it mostly bids correctly according to how it sees the results turning out. But the assumptions made as to how those bids are made by each seat - and whether they reflect how human players bid in real games - is, as always, debatable. Though I contend they SHOULD usually be bidding the way I have programmed (score and specific knowledge of other players' tendencies notwithstanding). And, my program still has its flaws.

Post Reply