Would you call this?

Ask questions, discuss and debate your strategies, euchre polls and more
Post Reply
sdu754
Posts: 42
Joined: Sat Jan 23, 2021 3:15 am

Would you call this?

Unread post by sdu754 » Sun Mar 05, 2023 1:56 pm

The score is 4-4, you are in the first seat and the dealer has turned down the 10H. You have JD, 9D, 9H, KC & 9C.

Would you pass this. I am on another site and people are arguing that this is a must call hand.



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

Unread post by Dlan » Sun Mar 05, 2023 4:11 pm

I would call diamonds and lead the 9c. Keep in mind you need your partner's help. Use your trump wisely. :)

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

Unread post by Wes (aka the legend) » Sun Mar 05, 2023 9:41 pm

sdu754 wrote:
Sun Mar 05, 2023 1:56 pm
The score is 4-4, you are in the first seat and the dealer has turned down the 10H. You have JD, 9D, 9H, KC & 9C.

Would you pass this. I am on another site and people are arguing that this is a must call hand.
This is absolutely a must call. Passing R+1 in Next when you don't block a reverse Next call would be terrible euchre.

sdu754
Posts: 42
Joined: Sat Jan 23, 2021 3:15 am

Unread post by sdu754 » Tue Mar 07, 2023 8:17 am

I'd like to see a simulation of this if possible.

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

Unread post by Wes (aka the legend) » Tue Mar 07, 2023 2:58 pm

sdu754 wrote:
Tue Mar 07, 2023 8:17 am
I'd like to see a simulation of this if possible.
A simulation is always nice. Who doesn't like direct evidence!? But in this case the indirect evidence we already have is strong enough to confidently draw the conclusion that passing R+1 in this spot is bad. Check out this exchange from another thread/hand:
irishwolf wrote:
Thu Dec 08, 2022 12:38 am
RAY,

One question I have, give S1: (Card_Q-H) (Card_10-H) (Card_9-H) (Card_Q-C) (Card_10-D) with (Card_J-H) upcard,

What are the EV for S2/S4 on R1?

Then ALL pass and S1 passes EV for S2/S4 vs S1 call next lead trump EV S1/S3?

Then what if S1 (basically a soft defensive donate) orders R1, the EV?

I actually suspect with nothing to call, still better to Pass for S1.



IRISH
Ray's response to Irishwolf"s query (check out what I bolded):
raydog wrote:
Thu Dec 08, 2022 1:58 pm
A lot to unwrap here, but I think I know what you want. This is why simulations are difficult - you have to zero in on the crux of the situation. This may take a few iterations.

To start, I ran a simulation of 100,000 hands, fixing S1's hand as Q-10-9H + 10D + QC (turn card JH), with S1 either passing or calling:

S1 passes:
S2,R1: (103 / 874 / 62 // 262 / 754 / 0)
S3, R1: (never calls with R turned)
S4, R1: 3,831 / 15,644 / 958 // 625 / 35,844 / 7,640)

S1, R2: (never calls)
S2, R2: (3,212 / 3,638 / 304 // 6,115 / 12,488 / 2,298)
S3, R2: (177 / 630 / 98 // 95 / 1,676 / 1,981)
S4, R2: (116 / 61 / 3 // 127 / 337 / 47)

S1 calls wp:
(0 / 0 / 0 // 0 / 28,830 / 71,170)

EV for S1 passing [from the perspective of S1]: -0.91
EV for S1 calling: -1.14

[scores are (sweep alone / 1pt alone / euchred alone // sweep wp / 1 pt wp / euchred wp)]

So better for S1 to pass (depending on score!)

Let's say it everyone passes, R1. Currently, my program has S1 passing, R2. But what if they call?

S1 passes, R2: EV = -1.13
S2, R2: (3,169 / 3,639 / 321// 6,056 / 12,430 / 2,346)
S3, R2: (2186 / 603 / 92 // 77 / 1,634 / 2,062)
S4, R2: (141 / 76 / 2 // 107 / 301 / 49)

S1 call, R2: EV = -1.01 [S1 calls D]
(0 / 0 / 0 // 958 / 9,638 / 22,695)

So actually better for S1 to call D, R2 - terrible EV, but better than if pass!

It looks like your "soft call" option is valid, but for R2, not R1.

Maybe you can glean the info you were looking for from the data I presented. If not, it would be too hard a slog to delve into all the considerations of when the other seats call or pass. I have written you a personal email to address this issue further.
So in this hand example, with S1 not blocking reverse next, it was correct for S1 to call Next with JUST the TD!!! It was correct for S1 to make this BS call becuz passing this BS hand actually performed worse. Now imagine if S1 actually had a real hand like R+1.

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

Unread post by jblowery » Tue Mar 07, 2023 3:26 pm

For that scenario above (with just one trump) it was actually a very close EV when the jack was turned down. May not be a good call if it was something like a king or queen.

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

Unread post by Wes (aka the legend) » Tue Mar 07, 2023 4:08 pm

jblowery wrote:
Tue Mar 07, 2023 3:26 pm
For that scenario above (with just one trump) it was actually a very close EV when the jack was turned down. May not be a good call if it was something like a king or queen.
Yes the analogy is not perfect (analogies never are) but the fact that it is correct to call Next--when one doesn't block reverse Next--with just one low trump in ANY SCENARIO is still strong evidence that calling Next with a real hand like R+1 in this spot is correct.

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

Unread post by Tbolt65 » Tue Mar 07, 2023 11:07 pm

Call Diamonds here like everyone else has said. I'd probably lead the 9 of clubs like dlan said myself as well.

Tbolt65
Edward

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

Unread post by jblowery » Wed Mar 08, 2023 9:39 am

Wes (aka the legend) wrote:
Tue Mar 07, 2023 4:08 pm
jblowery wrote:
Tue Mar 07, 2023 3:26 pm
For that scenario above (with just one trump) it was actually a very close EV when the jack was turned down. May not be a good call if it was something like a king or queen.
Yes the analogy is not perfect (analogies never are) but the fact that it is correct to call Next--when one doesn't block reverse Next--with just one low trump in ANY SCENARIO is still strong evidence that calling Next with a real hand like R+1 in this spot is correct.
I definitely agree that rt+1 in next is a no brainer call if you don't have reverse next blocked.

I just wanted to clarify that there is a difference in calling next when a jack is turned down vs. a lower card. I bet the EV comparison is generally not in your favor for calling next from S1 with just 1 low card in diamonds when a low heart is turned down. I thought Ray ran a simulation where even with 1 low card in next plus an off-suite ace it was negative (less negative EV if you just pass).

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

Unread post by irishwolf » Wed Mar 08, 2023 11:19 am

As to your the comment, "I bet the EV comparison is generally not in your favor for calling next from S1 with just 1 low card in diamonds when a low heart is turned down."

I will venture to say calling next leading the 9C with what holds that the EV will ALWAYS BE A + EV for any card turned down by R1.

S1 per much holds 2.25 probable tricks if played correctly so needs help from S3. Only about 20 - 25% of the time S3 will not be able to assist. So all those might be Euchres. Sometimes S3 will be loaded and they get some Sweep=s to negate those euchres. I say easy to estimate a positive EV for any turned down card R1, all pass. Sometimes you get bagged on. So what you have to make the call! Bring it!

If I am wrong I will eat my words. So perhaps Ray can run a Simulation for 10H or the JH turned down tro demonstrate the call by S1 R2.

IRISH

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

Unread post by jblowery » Wed Mar 08, 2023 1:39 pm

OK. Just to clarify, I DO agree to ALWAYS call next with rt + 1 and not having reverse next blocked, regardless of what the dealer turns down.

I'm talkinga about when S1 ONLY has one low trump or one low trump + an ace. I was talking specifially about Raydog's simulation from 12/8/22 quoted above.

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

Unread post by Wes (aka the legend) » Wed Mar 08, 2023 2:34 pm

jblowery wrote:
Wed Mar 08, 2023 9:39 am

I definitely agree that rt+1 in next is a no brainer call if you don't have reverse next blocked.

I just wanted to clarify that there is a difference in calling next when a jack is turned down vs. a lower card. I bet the EV comparison is generally not in your favor for calling next from S1 with just 1 low card in diamonds when a low heart is turned down. I thought Ray ran a simulation where even with 1 low card in next plus an off-suite ace it was negative (less negative EV if you just pass).
I can't remember if Ray did a simulation on that or not. That said, I don't think the upcard moves the needle that much, or at least not enough where I would care to change my strategy. As you probably recall, I am calling Next in that spot no matter what the turned down card is. I'm simply not passing when I don't block reverse Next at the score designated by the OP (4-4).

EVEN IF changing the upcard swings this Next Call to a -EV call, at worst it's gonna be a very small loser. I'll take that small loss to control some variance. And keep in mind with a strong partner who knows how to hand read, that hypothetical small -EV call may be flipped into a small +EV call. Remember, If your P knows that you always have reversed Next blocked when you pass in that spot that allows him to play his hand significantly better on defense and offense. Too many examples abound, but on defense, say against a 2S call, this will allow S3 to lead the Right in spots he otherwise wouldn't knowing S1 has the guarded Left. On offense S3 can go alone more in the opposite suit. For example: A black card was turned down, I pass, S2 passes and you have:

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

Since I passed, you know I MUST have the (Card_J-D) and with that knowledge you now know it's best to go alone.

Here's another one. Black card turned down, I pass, 2S passes, you have (Card_A-H) (Card_J-H) (Card_9-H) (Card_K-C) (Card_Q-C)

Since you know I MUST have the guarded Left you go alone in hearts knowing you effectively have JJX. After you go alone S2 leads the AS and you trump in with the 9H. Then you lead the Right but the Left doesn't come out. In normal scenarios you'd have to play it safe and lead the KC on 3rd street, but now you can lead the AH on 3rd street with impunity knowing I have the Left.

So EVEN IF Ray's simulation shows this weak Next call to be a small loser when other upcards are turned down this can still be a +EV call with a very strong partner. No simulation can account for this kind of synergy. Admittingly what I'm talking about is highly contrived. I'm basically referring to highly advanced play that maybe a handful of people in the world know about. But in my personal life it's extremely relevant. Whenever I play with Edward everything I've talked about is in play. I also have a 15 year old second cousin who's my roommate, Jacob, who I play euchre with a lot--Don and Wolf met him and Wolf played lots of euchre with him. In case you forgot Wolf he's the one who farts alot :P--I've trained him over the last so many months. He's an expert hand reader too now.
Last edited by Wes (aka the legend) on Wed Mar 08, 2023 9:57 pm, edited 1 time in total.

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

Unread post by jblowery » Wed Mar 08, 2023 3:18 pm

I agree Wes. I think it's going to be close either way. I generally only play with random partners, except my U. Curd girlfriend (but those are just casual games in Euchre 3D) so I'm interested in even a small EV difference.

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

Unread post by raydog » Sun Mar 12, 2023 6:43 pm

I had a look at this hand.

My simulator would bid this hand, and lead the 9C. But I had a look a few different options:
bid D, R2, lead 9C: (3,877 / 23,821 / 12,501) [2 pts. / 1pt / euchred] EV = +0.29
bid D, R2, lead 9H: EV = +0.17
bid D, R2, lead JD then 9H: EV = +0.16
pass: EV = -0.70
total games that make it to R2: 40,199 of 100,000

So this is definitely a hand to bid, and leading the 9C seems to be the way to go.

I also had a look at the same hand but with the JH turned:
bid D, R2, lead 9C: (1,585 / 11,665 / 7,184) EV = +0.02
bid D, R2, lead 9H: EV = -0.31
bid D, R2, lead JD then 9H: EV = 0.00
bid D, R2, lead JD then 9C: EV = 0.00
pass: EV = -0.91
total games that make it to R2: 20,434 of 100,000

So again, this is definitely a hand to bid, and once again leading the 9C seems to be the way to go.

What I found curious here was that S1 was much less successful in their bid when the R was turned. Here is my explanation. My simulator has S4 ALWAYS calling trump, R1, if they have R+1 trump. So in this instance, if they have ANY other H in their hand and the JH is turned they will pick it up. Which is why only 1/2 as many hands make it to R2. (I have simply found that this is the best way to maximize EV). Thus, the hands that make it to R2 are all hands in which S4 has NO hearts - which means they have a bit more cards in the other suits. Hence, better defense when another suit is called by S1, and less success by S1. I find this to be a satisfactory explanation.

A word about my simulator: it bids and plays every hand to maximize EV. It has no concept of a player's tendencies - to bag, to bid aggressively or lightly - or of the score. These are of course factors which arise in games among humans, and it can be useful if not critical to take them into account. But WAY too difficult to program. So they are factors which you should overlay on the results my simulator spews out. When there is a close call, those factors may well make a difference.

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

Unread post by irishwolf » Sun Mar 12, 2023 8:24 pm

MAKES SENSE TO ME RAY!

Mxx
Posts: 23
Joined: Wed Mar 27, 2019 1:20 am

Unread post by Mxx » Mon Mar 13, 2023 7:57 pm

I can't prove it like a simulator can, but my experience is lead the Jd then the 9c.
1. Your partner will likely have at least two trump, you're not going to strip them (which in my mind is the main reason to lead the 9c first)
2. Opponents will likely now have zero trump
3. Aces are made good
4. You can lead the 9c second and set up your Kc as a winner (possibly after you play 9d on a spades lead)
5. You still leave open the option that your partner will trump in on an opponent's Ace

Prove me wrong! I play on the Karman app against random opponents in the 90s.

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

Unread post by raydog » Mon Mar 13, 2023 10:53 pm

Mxx, you raise some interesting points. Which I'd like to address.

First, my simulator program is not the final word. It is not completely optimized, and does make some poor bids / plays which I continually strive to correct. But it does employ a coherent system of rules and is rarely far from the mark (from my experience testing scores of scenarios proposed on this forum). Nonetheless, I do always attempt to understand WHY it arrives at the results it does, because that is how we will ultimately improve our play, personally.

Simple random card distribution is no mystery: it's not too hard (with a little knowledge of probabilities) to figure how likely an opponent is to hold an ace, or 2 trump, or that crucial bower. What if far more difficult to calculate is CONDITIONAL probability. So in this particular case, given that only a minority of hands make it to R2 for bidding [about 40%, by my simulator's analysis], what is NOW the probability that S3 holds the JH, or that S4 holds the AC, or every player other player holds a S? It all depends on the subset of hands which is likely to be passed by everyone, R1. [I recently calculated that, given the 5 cards in S1's hand and the turn card, there are about 600 million possible distributions of the remaining cards among the other 3 players and the kitty]. And it is crucial to understand the likely composition of that card distribution, if we are to draw reasonable conclusions. And, for me, that is where a simulator comes in handy, because it automatically culls the lot. BUT, you need to have confidence in the simulator!

Returning to the hand we are analyzing here, I tested the odds of S3 holding various # of trump, GIVEN that everyone passed R1.
0 trump: 13.5%
1 trump: 39.4%
2 trump: 35.5%
3 trump: 10.6%
So your assertion that partner (S3) probably has 2 or more trump almost plays out (47.1%: 100-13.5-39.4).

I also reran my simulation, using 3 choices of play:
lead 9C: (4,126 / 24,787 / 10,974) [2 pt sweep / 1 pt / euchred] EV = +0.28
lead JD, then 9C: (4,203 / 23,23,937 / 11,747) EV = +0.22 [an improvement over JD followed by 9H!]
lead KC: (2,927 / 25,399 / 11,561) EV = +0.20

This is a tighter range than I found with my original scenarios, but still favors leading the 9C - with a margin I deem to be statistically significant. So what is going on?

The key factors to look at are sweeps and euchres. And avoiding a euchre is 3 times as important as getting a sweep. (Winning 3 or 4 tricks is identical, point-wise). Leading the JD DOES lead to more sweeps, but the cost in additional euchres is too much to bear. I looked at the odds of that initial 9C lead being trumped: 31.8%. Would you have guessed it was so low? To me that shows that S3 is very likely to have a C [indeed, with a R2 bid, card distribution tends to be much more even - fewer voids], so if an opponent has the A, that is a sweep out the window no matter what (but you can set up that KC!) And maybe by leading that JD you are ruining the possibility of a later overtrump, which would scupper a euchre or lead to a sweep. There are a myriad of possible scenarios, it's too complex to analyze. Which is where I think a RELIABLE simulator can do the analysis for you.

I think your analysis of the scenario is correct, but incomplete. You state situations where leading the R is favorable, but don't mention the cases where it could be detrimental. And, crucially, you can't predict the LIKELIHOOD of each scenario occurring. And therein lies the usefulness of a simulator. IF it does its job well (and I understand your skepticism!)

I can't PROVE you wrong, I can only say you are quite right to challenge my results and make me think about if my simulator's results are reasonable. In this case, I think they are. But you are of course free to play as you like.

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

Unread post by irishwolf » Mon Mar 13, 2023 11:10 pm

Prove me wrong! I think it is up to you to prove the Simulator wrong? Do you have hands played or just your GUT feel? But how would you offer that other proof be provided, another 40,000 hands by the simulator?

Humans are terrible at gut probabilities according to the expert statisticians.

For the 10H down, the Simulator just proved it 39,887 hands that leading the 9C is statistically a better lead. I add to that, based on my experience, it gives the marker, S1 a lot of options after leading the 9C.

My simulator would bid this hand, and lead the 9C. But I had a look a few different options:
bid D, R2, lead 9C: (3,877 / 23,821 / 12,501) [2 pts. / 1pt / euchred] EV = +0.29
bid D, R2, lead 9H: EV = +0.17
bid D, R2, lead JD then 9H: EV = +0.16
pass: EV = -0.70
total games that make it to R2: 40,199 of 100,000

Ridiculous, what more do you need?

Irish

Mxx
Posts: 23
Joined: Wed Mar 27, 2019 1:20 am

Unread post by Mxx » Tue Mar 14, 2023 8:42 pm

Hi Raydog, thanks for taking the time to investigate further and explain the results of your simulator.

It looks like I am proved wrong in the world of the simulator! I can only justify my stance based on your comment below:
...factors which arise in games among humans, and it can be useful if not critical to take them into account.
For example: you mention the simulator always picks up at S4 when holding the Right (which I wholeheartedly agree with), but this is a play I see missed all the time on the app I play, even among highly ranked players.

I was surprised to see S3 has 2+ trump in only 47% of hands. Maybe I should keep a running tally of my games to see if my glasses are too rosy.

Would you be willing to indulge me and see what the EVs are for a 9d lead? On this site's Euchre lessons (which are excellent), there is a suggestion that when bidding next for S1 holding Right +1 and an offsuit ace, you should lead the smaller next card to hit S3's likely Left. I have tried it (including for next calls with Right + 1 and no ace, like the OP's question) and found it doesn't work enough. But it looks cool when it does!

Mxx

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

Unread post by raydog » Wed Mar 15, 2023 3:39 pm

I have indulged you and had a look at the scenario of leading the 9D. I compared that to leading the 9C or the JD, over 100,000 hands. I also had a look at which seat is likely to hold the JH and the AD, if bidding reaches R2.

Of 100,000 tested, 40,094 make it to R2, where S1 bids D (next):
- lead 9C: EV = +0.29
- lead JD: EV = +0.23
- lead 9D: EV = -0.14

who has the JH [which was the R in the first round, and is now the L]?
- S2: 18.9%
- S3: 45.6%
- S5: 5.3%
- buried: 30.2%

who has the AD?
- S2: 27.9%
- S3: 25.5%
- S4: 31.8%
- buried: 14.8%

Note that we would expect any one of the players to hold a given card 5/18 = 27.8% of the time (knowing S1 doesn't hold it and it isn't turned), and would expect it to be buried 3/18 = 16.7% of the time. But since we are only looking at the 40% of hands that make it R2, these percentages are skewed, sometimes drastically.

So what does all this mean? First, we can see that S1 fares much more poorly when leading the 9D. Sure, S3 holds the L a little less than half the time, and will take this trick in those cases. But when S3 DOESN'T have the L, S2 or S4 is likely to have it, or perhaps the AD (if the L is buried), which results in giving a free trick to the opponents if they held a single trump. I haven't (and won't) delve down each branch of the probability tree. But I do want to highlight that the L is buried almost 1/3 of the time, that S4 has the best chance of having the AD [having not bid H, R1, S4 is proportionately short on H, so proportionately longer in the other suits, so has an increased likelihood of holding the particular card, the AD], and so it is not implausible that S1 hurts himself more than he helps himself by leading a low D.

Consider this one particular case. I find that when S1 calls next, R2, and holds the L + a low trump, it IS best to lead that low trump. Once again, partner is very likely to have the R and win the trick, BUT if one of the opponents holds the R, there is a good chance they will have to burn it on this first trick, making the L good. If S1 LEADS the L, it is wasted in almost every case (except when the R is buried). The only difference between the two scenarios is WHICH bower S1 holds (with S3 always being the odds-on favorite to hold the other one), but the best strategy is notably different.

So in my mind, while a case can be made for leading the 9D (as you said), a case can also be made for leading the JD, and it is not always possible (and certainly not in this case) to adequately assess the pros and cons and the true likelihood of every favorable or unfavorable scenario arising. A good simulator can do the analysis in 3 minutes, and has.

I'm sticking with leading the JD.

Mxx
Posts: 23
Joined: Wed Mar 27, 2019 1:20 am

Unread post by Mxx » Thu Mar 16, 2023 12:18 am

Thanks raydog!

Post Reply