01-24-22 OE Monday - got euchred 2

Ask questions, discuss and debate your strategies, euchre polls and more
Post Reply
User avatar
Dlan
Site Admin
Posts: 669
Joined: Thu Apr 19, 2018 10:08 pm
Location: Ohio

01-24-22 OE Monday - got euchred 2

Unread post by Dlan » Wed Jan 26, 2022 7:37 pm




Wes (aka the legend)
Posts: 1522
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Wed Jan 26, 2022 8:08 pm

Maybe I'm missing something but I would say hand was played well by everybody every street.

User avatar
Dlan
Site Admin
Posts: 669
Joined: Thu Apr 19, 2018 10:08 pm
Location: Ohio

Unread post by Dlan » Wed Jan 26, 2022 10:18 pm

I'm not a big fan of leading trump on a weak next call. While I can't back this up with numbers, I've had better luck leading offsuit junk.

irishwolf
Posts: 1237
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Thu Jan 27, 2022 10:47 am

Should have ordered spades.

Wes (aka the legend)
Posts: 1522
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Fri Jan 28, 2022 3:51 am

Dlan wrote:
Wed Jan 26, 2022 10:18 pm
I'm not a big fan of leading trump on a weak next call. While I can't back this up with numbers, I've had better luck leading offsuit junk.
We actually do have some numbers tho on this type of Next call (L+1+A) thanx to Ray, and the results were surprising:

Ray tested this hand:

Upcard (Dealer passes): (Card_9-H)

S1 calls Next with:

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

If I had called Next with that hand I would recommend leading the KD but it turns out that lead may be wrong:
raydog wrote:
Thu Dec 30, 2021 10:08 pm
Here's what I found, simulating the hand.

If S1 calls D, R2, better to lead the AC (EV = +0.31) than the JS (EV = +0.27). Leading the JH or KD are even worse. Why? I think the hand is only winnable if your partner has a couple of trump, or one trump and a couple are buried. You are looking for a cross ruff, maybe two even, but drawing trump will too often lose you control of the hand. (just guessing)
Quote is from this thread: viewtopic.php?f=8&t=491

So Rays numbers suggest that leading the off ace unpromoted is better than leading garbage and better than leading trump. That's a very interesting finding.

Here's more evidence from another L+1+A next call:

Upcard: (Card_K-S) (turned down)

Next call: (Card_J-S) (Card_9-C) (Card_A-D) (Card_9-D) (Card_9-S)
raydog wrote:
Thu Dec 30, 2021 6:01 pm
J-9S + A-9D + 9C (KS turned) 10,000 hands
S1, bid R1 (lead JS, then AD): EV = -0.19
S1, bid next R2 (lead AD): EV = -0.02
So better to bag and bed next, R2, in this case. Likely for the reason you mentioned.
As for leading the AD, I find that in general, suits are more evenly distributed when bidding gets to the 2nd round, probably just because the turned suit is more evenly distributed and this has an impact on how the other suits are distributed. Not a strong effect, but enough to make that Ace lead win a bit more and make it a sensible lead.
Quote is from this thread: viewtopic.php?f=8&t=780

Ray's work again suggests that leading the off ace is correct.

Why has leading the Ace unpromoted done slightly better than leading trump from this configuration? Rays guess was:
Why? I think the hand is only winnable if your partner has a couple of trump, or one trump and a couple are buried. You are looking for a cross ruff, maybe two even, but drawing trump will too often lose you control of the hand. (just guessing)
I think there's some truth to the above, there's obviously multiple factors at work here, but I do actually think there's a primary factor that hasnt been directly mentioned, and that has to do with the very nature of the hand configuration itself: L+1+A.

I think it is generally true that one should lead trump before their off aces but there's a countervailing problem with specifically L+1+A hands. 1) we want to lead trump to protect our off Ace but 2) When we do lead trump we are unprotecting our guarded left setting up the possibility that we get no tricks with both our trump when we have a guaranteed trick if we just don't lead trump. The reward of protecting our Ace may not be worth the risk. That's what Rays numbers are suggesting.

Here's the thing tho, the hand in the OP is not like the hands Ray tested in a very important way. With this holding...

Upcard: (Card_9-S) (turned down)

Next call: (Card_J-S) (Card_A-C) (Card_A-S) (Card_9-H) (Card_9-D)

...We never have to worry about unprotecting our Left and getting stripped. That obviously can't happen when we have L+A in trump. Also, leading this off ace unpromoted is less favorable becuz it's the turned down suit. And we already know leading our other junk offsuit is not correct based on Rays other work. So in this hand we actually get the benefits of leading trump without the added risk of unguarding our Left. And as mentioned leading a turned down Ace is not gonna get the same favorable results as in Rays other examples. Therefore I would hypothesize that the best way to play this hand is the way S1 played it, Lead the Left to help promote that off ace.

Another hypothesis worth mentioning:

1) Whenever we call Next with 2 trump + an Ace we should always lead trump, the only exception may be when we hold L+1 in trump for the above reasons as long as that L+1 is not L+A in trump.

So for example if the 9D is turned down and S1 calls next with:

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

S1 should always lead trump.

Finally as far as this idea:
irishwolf wrote:
Thu Jan 27, 2022 10:47 am
Should have ordered spades.
I doubt that will be true. I think S1 played this hand the best way possible. But I'd be very interested in seeing that spade call tested. It wouldn't shock me if I was wrong.

irishwolf
Posts: 1237
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Fri Jan 28, 2022 11:20 am

I was being facetious but it would have worked calling spades. I would call next but I would have led 9D or 9H. Only leading 9D would have worked in this particular hand when calling Next.

Wes (aka the legend)
Posts: 1522
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Fri Jan 28, 2022 2:31 pm

irishwolf wrote:
Fri Jan 28, 2022 11:20 am
I was being facetious but it would have worked calling spades. I would call next but I would have led 9D or 9H. Only leading 9D would have worked in this particular hand when calling Next.
Just woke up and realized I was wrong to rule out leading junk because the examples are not analogous to the actual hand. All we can say right now from the data we have is it looks like leading a non-turned down Ace is better than leading junk. But in the actual hand it's a turned down ace. There's a chance leading junk is better than leading a turned down ace. Who knows. I'm still betting S1 played it the best way possible.

User avatar
LeftyK
Posts: 260
Joined: Wed Jun 10, 2020 9:45 am
Location: North Carolina

Unread post by LeftyK » Fri Jan 28, 2022 6:17 pm

Monday was a chittshow as I went 3/10 but I'm telling you this hand here DOWN 8-9 I'm calling with RAce and leading the right ! this play cannot be overlooked.

raydog
Posts: 205
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Sun Feb 06, 2022 1:47 pm

I simulated this hand, and couldn't find a distinction between leading trump and leading 9H, but either is better than leading the AS: (R2 C call by S1; 100,000 hands)
lead 9H: EV = +0.02
lead AS: EV = -0.14
lead JS (then AS): EV = +0.02

I also looked at bidding R1, bidding R2 and passing both rounds (from S1 perspective):
bid S, R1: EV = -0.02
bid C, R2: EV = +0.30
pass both rounds: EV = +0.19
[this scenario includes the results of R1 bids, so not comparable to first set of results]

Wes (aka the legend)
Posts: 1522
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Sun Feb 06, 2022 4:56 pm

raydog wrote:
Sun Feb 06, 2022 1:47 pm
I simulated this hand, and couldn't find a distinction between leading trump and leading 9H, but either is better than leading the AS: (R2 C call by S1; 100,000 hands)
lead 9H: EV = +0.02
lead AS: EV = -0.14
lead JS (then AS): EV = +0.02

I also looked at bidding R1, bidding R2 and passing both rounds (from S1 perspective):
bid S, R1: EV = -0.02
bid C, R2: EV = +0.30
pass both rounds: EV = +0.19
[this scenario includes the results of R1 bids, so not comparable to first set of results]
Once again awesome work Ray. Now we can put to bed forever the idea that calling spades R1 is better than calling Next R2 from this configuration. This will help anyone who wondered about that.

I'm glad to see that we now have good evidence--most likely the best evidence we'll ever have--backing up the claim that S1 played this hand optimally. The surprise of course is there appears to be two optimal paths here:

1) Call Next, Lead Left

&

2) Call Next lead garbage.

Points to Wolf and Don for suggesting the latter line. My mind completely dismissed it obviously probably due to confirmation bias or something. And this leads to where your findings get really interesting Ray, not from your actual numbers but from the inference we may be able to draw from them.

If Wolf and Don's claim--that leading garbage is best for this exact hand--is a statistical tie with leading trump that actually suggests that leading garbage is the best line overall whenever we have L+1+A when that ace is from the turned down suit. Why? This goes back to the problem of leading trump from L+1.

There's a cost we incur of potentially unguarding our Left and getting no tricks from our 2 trump. The only time we don't incur this cost is when we have precisely L+A. So if leading garbage is a statistical tie with leading trump from L+A I think we can infer that leading garbage will be best from L+K or worse due to the unguarding cost of leading trump. The only countervailing problem we have to worry about are those times our garbage consists of an offsuit King or two as there is some marginal value gained by leading trump and protecting that off King those times it is boss.

That's why the following hand would be the best hand possible to test the hypothesis that leading garbage is best overall from this configuration (assume same spade upcard turned down):

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

If the above hand were tested and leading garbage was best that would definitively prove to me that Dons/Wolf's claim that leading garbage is the best line overall is correct. As of now I still strongly suspect they are right.

raydog
Posts: 205
Joined: Thu Sep 16, 2021 6:56 pm

Unread post by raydog » Sun Feb 06, 2022 9:52 pm

I simulated Wes's hand, it was again quite hard to make a statistically significant distinction. With S1 bidding C, R2, and testing 100,000 hands:
lead KH: EV = 0.00
lead KC, then AS (if KC wins): EV = 0.00
lead JS, then AS (if JS wins): EV = -0.03
lead AS: EV = -0.14

But overall, still best for S1 to bid C, R2:
bid R1: EV = -0.14
bid C, R2: EV = +0.12
pass both rounds: EV = -0.16

Wes (aka the legend)
Posts: 1522
Joined: Wed Jun 13, 2018 3:03 pm

Unread post by Wes (aka the legend) » Mon Feb 07, 2022 1:33 am

raydog wrote:
Sun Feb 06, 2022 9:52 pm
I simulated Wes's hand, it was again quite hard to make a statistically significant distinction. With S1 bidding C, R2, and testing 100,000 hands:
lead KH: EV = 0.00
lead KC, then AS (if KC wins): EV = 0.00
lead JS, then AS (if JS wins): EV = -0.03
lead AS: EV = -0.14

But overall, still best for S1 to bid C, R2:
bid R1: EV = -0.14
bid C, R2: EV = +0.12
pass both rounds: EV = -0.16
Good stuff Ray. Two things of note:

1) The (L+A)+turned down suit Ace is the best hand type out of this configuration for leading trump (in theory) and yet leading garbage was a statistical tie vs leading trump (the Left)

2) I handpicked the very best (L+K)+turned down suit Ace for leading trump. I basically tried to rig it against leading garbage and yet we had another statistical tie between leading garbage vs leading trump (KC).

I think that's very good evidence validating Don and Wolf's idea that leading garbage is the best lead overall from this configuration, IE L+1+A when that Ace is from the turned down suit.

And to recap on the other L+1+A configuration, when our off ace is NOT from the turned down suit, then leading the off Ace is best as long as it's not a tripleton off ace.

Good stuff.

A potential confounding factor to the above is if S1's partner bags heavy from S3. This is something both me and Edward do. That dynamic could swing things back to always leading trump from these Next call configurations. The simulator results--specifically the small ev difference between competing lines--are certainly close enough where 3rd Seat play style could make a difference. Just food for thought.

irishwolf
Posts: 1237
Joined: Tue Apr 24, 2018 9:33 pm

Unread post by irishwolf » Mon Feb 07, 2022 1:50 am

I am repeating myself but no mystery to me, with that hand calling next leading junk in opposite color is best. You have two junk suits and if S4 wins and leads back you can duck and this gets S2 in play.You need a little help from your partner and get him into play. He could have a void or an Ace to your junk lead. Just so many more options leading so you can SNEAK IN THE BACK DOOR FOR A POINT. You can't pass there at most any score. Nothing works 100%.

"I was being facetious but it would have worked calling spades. I would call next but I would have led 9D or 9H." No need to order there, R1.

Glad Ray work provide evidence.

IRISH

Post Reply