You are ignoring the author of this comment. Click to temporarily show the comment.

Thanks for the info. No problem with those (logical) rules. I wonder about the basis for the following: “Each player may bar one individual from kibitzing play at his table during a session without cause.”

You are ignoring the author of this comment. Click to temporarily show the comment.

In practice, it would be really difficult for a kibitzer to continue watching someone who asked politely (not to watch). At least, the player has a right not to show his/her hand.

You are ignoring the author of this comment. Click to temporarily show the comment.

Today, a friend and I were discussing the play in

Dummy: JT32 vs. Hand: K87 for two tricks.

It was not difficult to estimate that the best play was 7 to J and then 2 to K (about 80.5% chance).

Then, remembering your post, I wondered about the effect of 6-spot, and looked into

JT62 vs. K87 for two tricks.

Now, we again start with 7 to J, but then (unless J loses to A) no need to play 2 to K, but instead continue with 8 to T. This wins when LHO has Ax, loses when RHO has AQ. Ax is three 2-4 distributions (A3+A4+A5) while AQ is only one. So, it makes a (2 x 1.613 =) 3.2% difference.

6-spot seems more valuable than I've noticed in practice.

You are ignoring the author of this comment. Click to temporarily show the comment.

Thank you all for your comments and votes.

The original deal:

W: ♠Q4 ♥T83 ♦T863 ♣Q954

N: ♠T9832 ♥AK5 ♦- ♣AKJT7

E: ♠J765 ♥J4 ♦AQ954 ♣83

S: ♠AK ♥Q9762 ♦KJ72 ♣62

No doubt, 6♥ has a much greater chance than 7♥; however, 6♥ is problematic while 7♥ is easy.

7♥ is a make because there is only one normal line: ruff ♦, cash ♥AK, ♠A, ♥Q (discard ♠), ♠K, ♣ to J, ruff ♠… ♣ finesse is necessary even if ♠s are 3-3. Ruffing a ♠ is a precaution only against five ♣s to Q in W.

In 6♥, (as it seems) there are two reasonable lines, only one of which wins. On both lines, at tricks 1 and 2: discard ♣, win ♠ return. As of trick 3, on LINE-1: ruff ♦, cash ♥AK, ♠K, ♥Q, ♣A, ruff ♠… As of trick 3, on LINE-2: cash ♥AK, ♠K, ruff ♦, ruff ♠…

LINE1 wins.

NOTE: It came up in a 7-table event with the Lehman-scoring at the Jeofizik Bridge Sport Club (in Ankara) in January. The frequencies were: 4♥: 4 times, 3NT: once, 6♥: twice (made once).

You are ignoring the author of this comment. Click to temporarily show the comment.

I think you would also ruff a ♠ (after clearing trumps in 3 rounds) in the process.

As far as I know there is only one other line, which mainly risks a quick set when W has 2♠-3♥ (with ♥J or T), as you noted.

IMO, ruffing-♦-at-trick-3 line is as good as the other one. With Michael Rosenberg's subtle note above on the squeeze scenarios, it loses ONLY when W has ♦Q (longer than third), E has ♣Q (not stiff) and ♠s are 42+24; provided that ♥s are 23+32 and ♠s are not worse than 42+24. And so far, it seems to be the line selected by more people here and also among my bridge friends.

With respect to the odds on paper, when ♥s are not 41+14, it is little better. When considering the case where West has stiff ♥J or ♥T, the other line gains a bit to pass it just little. This is insignificant, and does not imply anything about which line wins on the original layout.

You are ignoring the author of this comment. Click to temporarily show the comment.

David B.: The other table was in 6♥ (result unknown), as noted in OP. Then you should vote for 7♥.

Anyway, I certainly thought that 7♥ would get more votes because the other table was in 6♥.

When ♥s are not 32+23, 6♥ has a very low chance (at most 2-3% and even that depends on the line of play) while 7♥ has no chance (seemingly).

When ♥s are 32+23 and ♠s are 33+42+24, 6♥ has a very good chance (say, near 90%); however, in that case, 7♥ will also succeed almost 50% of the time on the obligatory line. I thought more people would find 7♥ worth attempting a swing.

You are ignoring the author of this comment. Click to temporarily show the comment.

It was constructed as a hypothetical setup; so, no bidding, no table presence factors, no defensive errors…

I've already made an analysis for the preconditions: ♥s are 32+23, ♠s are 33+42+24 (neglecting fall of ♠QJ), and neither defender has a singleton minor (using a suit-break-odds calculator with vacant-places input available on internet). With these conditions, the difference between the two lines in 6♥ is less than 1% (the success chances being 86-87%).

The case in which LHO has stiff ♥J or ♥T was not included. The total probability of stiff ♥J or ♥T in W is 5.64%. It is not easy to estimate how much of this can be converted to the overall success chance. However, it seems that this would change the result in favor of the second line by around 1%.

It's not only the books telling me about probabilities, there are also several programs (written by bridgeurs) that can be utilized (book knowledge rather serves the purpose of verification in some instances). For instance, a hand-generation program could be used to obtain the odds for this 6♥; but that would be burdensome and not meaningful, especially when the differences are small. Such results are pure probability (on paper); and as you imply, may not mean a thing in practice. Anyway, it would be irrational to claim that this or that line is better because its success chance is a few percent higher.

NOTE: “As an example, singletons tend to be less likely than probabilities suggest.” I need to check on this, to try to understand why (at least).

EDIT: Minor corrections/changes not to cause any misunderstanding.

You are ignoring the author of this comment. Click to temporarily show the comment.

Good and interesting point! It's really essential to cash ♥ first in the 4-card ending (as you noted “key play” above). If ♦K is played first, the triple squeeze on RHO is not exposed; and that's the reason why I couldn't see it in the beginning.

While we were discussing 6♥ at the club, 2 lines came forward. (1) At trick 3, ruff ♦, and continue as you said above (now, surely cash ♥ before ♦K). (2) At trick 3, cash ♥AK, and continue as I said above.

I asked three friends (also national players). Two of them chose Line 1, one saying: “I hate going down when trumps are 32 and I could have collected them.” The other chose Line 2.

The odds seem very close. I will try to put forward a rough estimation if I find time.

You are ignoring the author of this comment. Click to temporarily show the comment.

Michael R.: “If RHO started with 4-card ♠ and ♦Q, he is triple squeezed.”

Yes, if RHO had ♣Q in addition to 4-card ♠ plus ♦Q, he would be triple squeezed. Yet, how are we going to know that he is squeezed? What if LHO has ♣Q?

EDIT: Sorry, you are right, when RHO is triple squeezed we will automatically know that since he cannot hold the last 3 cards.

You also say: “if LHO had 4-card ♠, play ♣K. Otherwise finesse.” I believe this is the percentage play in your line (that is, not going for the triple squeeze when RHO has 4-card ♠. This is not correct, see EDIT above).

What do you think of the following alternative line? Trick 1: Pitch a ♣ 2: Win ♠ return 3: ♥A 4: ♥K 5: ♠ to hand 6: Ruff a ♦ 7: Ruff a ♠ If you don't get overruffed, you're home.

You are ignoring the author of this comment. Click to temporarily show the comment.

You select 6♥? If so, on which line? If you go for “one other possibility at trick 3”, why don't you want to be in 7♥? (Seems to me that such a line is better suited for 7♥, as Jeff L. and David W. have noted below.)

You are ignoring the author of this comment. Click to temporarily show the comment.

I'd double any suit, as long as I had a clear opener.

I much like the understanding that any take-out double at one level promises at least three cards in the three unbid suits. I don't think this is alertable.

You are ignoring the author of this comment. Click to temporarily show the comment.

David: “The more cards a side has in a suit the less likely uneven breaks are to occur,…”

I find this confusing. Do you mean that “uneven break” in one suit enhances “even break” in another suit?

Say we have 8 ♠s and 7 ♦s out (forget about the ♥s in this deal). In which case is 61+16 ♦ break more likely: when ♠s are 5-3 or 6-2? If you say “when ♠s are 5-3”, I will disagree.

You are ignoring the author of this comment. Click to temporarily show the comment.

It seems that E=3424 with ♦A cannot be handled after cashing ♥A. Then; cashing ♥A first (Tom's line) wins when E=2434, cashing ♥K (your line) wins when E=3424 or 3415. I don't know which one offers a higher chance.

Besides these 3, I could not find any other winning case (provided that we naturally cash one ♥ first).

EDIT: It also boils down to how many ♠s West (or E) has. Giving West 5 ♠s, cash ♥K; 6 ♠s, cash ♥A.

Okan Zabunoglu

Okan Zabunoglu

Okan Zabunoglu

NOTE: I personally don't object to your statement although I think that the inner dynamics have played the major role.

Okan Zabunoglu

No problem with those (logical) rules. I wonder about the basis for the following:

“Each player may bar one individual from kibitzing play at his table during a session without cause.”

Okan Zabunoglu

I couldn't help wondering about the basis.

Okan Zabunoglu

Okan Zabunoglu

Dummy: JT32 vs. Hand: K87 for two tricks.

It was not difficult to estimate that the best play was 7 to J and then 2 to K (about 80.5% chance).

Then, remembering your post, I wondered about the effect of 6-spot, and looked into

JT62 vs. K87 for two tricks.

Now, we again start with 7 to J, but then (unless J loses to A) no need to play 2 to K, but instead continue with 8 to T. This wins when LHO has Ax, loses when RHO has AQ. Ax is three 2-4 distributions (A3+A4+A5) while AQ is only one. So, it makes a (2 x 1.613 =) 3.2% difference.

6-spot seems more valuable than I've noticed in practice.

Okan Zabunoglu

The original deal:

W: ♠Q4 ♥T83 ♦T863 ♣Q954

N: ♠T9832 ♥AK5 ♦- ♣AKJT7

E: ♠J765 ♥J4 ♦AQ954 ♣83

S: ♠AK ♥Q9762 ♦KJ72 ♣62

No doubt, 6♥ has a much greater chance than 7♥; however, 6♥ is problematic while 7♥ is easy.

7♥ is a make because there is only one normal line:

ruff ♦, cash ♥AK, ♠A, ♥Q (discard ♠), ♠K, ♣ to J, ruff ♠…

♣ finesse is necessary even if ♠s are 3-3. Ruffing a ♠ is a precaution only against five ♣s to Q in W.

In 6♥, (as it seems) there are two reasonable lines, only one of which wins.

On both lines, at tricks 1 and 2: discard ♣, win ♠ return.

As of trick 3, on LINE-1: ruff ♦, cash ♥AK, ♠K, ♥Q, ♣A, ruff ♠…

As of trick 3, on LINE-2: cash ♥AK, ♠K, ruff ♦, ruff ♠…

LINE1 wins.

NOTE: It came up in a 7-table event with the Lehman-scoring at the Jeofizik Bridge Sport Club (in Ankara) in January. The frequencies were: 4♥: 4 times, 3NT: once, 6♥: twice (made once).

Okan Zabunoglu

As far as I know there is only one other line, which mainly risks a quick set when W has 2♠-3♥ (with ♥J or T), as you noted.

IMO, ruffing-♦-at-trick-3 line is as good as the other one. With Michael Rosenberg's subtle note above on the squeeze scenarios, it loses ONLY when W has ♦Q (longer than third), E has ♣Q (not stiff) and ♠s are 42+24; provided that ♥s are 23+32 and ♠s are not worse than 42+24. And so far, it seems to be the line selected by more people here and also among my bridge friends.

With respect to the odds on paper, when ♥s are not 41+14, it is little better. When considering the case where West has stiff ♥J or ♥T, the other line gains a bit to pass it just little. This is insignificant, and does not imply anything about which line wins on the original layout.

Okan Zabunoglu

Anyway, I certainly thought that 7♥ would get more votes because the other table was in 6♥.

When ♥s are not 32+23, 6♥ has a very low chance (at most 2-3% and even that depends on the line of play) while 7♥ has no chance (seemingly).

When ♥s are 32+23 and ♠s are 33+42+24, 6♥ has a very good chance (say, near 90%); however, in that case, 7♥ will also succeed almost 50% of the time on the obligatory line. I thought more people would find 7♥ worth attempting a swing.

Okan Zabunoglu

I've already made an analysis for the preconditions: ♥s are 32+23, ♠s are 33+42+24 (neglecting fall of ♠QJ), and neither defender has a singleton minor (using a suit-break-odds calculator with vacant-places input available on internet). With these conditions, the difference between the two lines in 6♥ is less than 1% (the success chances being 86-87%).

The case in which LHO has stiff ♥J or ♥T was not included. The total probability of stiff ♥J or ♥T in W is 5.64%. It is not easy to estimate how much of this can be converted to the overall success chance. However, it seems that this would change the result in favor of the second line by around 1%.

It's not only the books telling me about probabilities, there are also several programs (written by bridgeurs) that can be utilized (book knowledge rather serves the purpose of verification in some instances). For instance, a hand-generation program could be used to obtain the odds for this 6♥; but that would be burdensome and not meaningful, especially when the differences are small. Such results are pure probability (on paper); and as you imply, may not mean a thing in practice. Anyway, it would be irrational to claim that this or that line is better because its success chance is a few percent higher.

NOTE: “As an example, singletons tend to be less likely than probabilities suggest.” I need to check on this, to try to understand why (at least).

EDIT: Minor corrections/changes not to cause any misunderstanding.

Okan Zabunoglu

While we were discussing 6♥ at the club, 2 lines came forward.

(1) At trick 3, ruff ♦, and continue as you said above (now, surely cash ♥ before ♦K).

(2) At trick 3, cash ♥AK, and continue as I said above.

I asked three friends (also national players). Two of them chose Line 1, one saying: “I hate going down when trumps are 32 and I could have collected them.” The other chose Line 2.

The odds seem very close. I will try to put forward a rough estimation if I find time.

Okan Zabunoglu

Okan Zabunoglu

Yes, if RHO had ♣Q in addition to 4-card ♠ plus ♦Q, he would be triple squeezed. Yet, how are we going to know that he is squeezed? What if LHO has ♣Q?

EDIT: Sorry, you are right, when RHO is triple squeezed we will automatically know that since he cannot hold the last 3 cards.

You also say: “if LHO had 4-card ♠, play ♣K. Otherwise finesse.” I believe this is the percentage play in your line (that is, not going for the triple squeeze when RHO has 4-card ♠. This is not correct, see EDIT above).

What do you think of the following alternative line?

Trick 1: Pitch a ♣

2: Win ♠ return

3: ♥A

4: ♥K

5: ♠ to hand

6: Ruff a ♦

7: Ruff a ♠

If you don't get overruffed, you're home.

NOTE: You didn't ruff ♦J at trick 3!

Okan Zabunoglu

If you go for “one other possibility at trick 3”, why don't you want to be in 7♥?

(Seems to me that such a line is better suited for 7♥, as Jeff L. and David W. have noted below.)

Okan Zabunoglu

Okan Zabunoglu

I much like the understanding that any take-out double at one level promises at least three cards in the three unbid suits. I don't think this is alertable.

Okan Zabunoglu

Okan Zabunoglu

I find this confusing. Do you mean that “uneven break” in one suit enhances “even break” in another suit?

Say we have 8 ♠s and 7 ♦s out (forget about the ♥s in this deal). In which case is 61+16 ♦ break more likely: when ♠s are 5-3 or 6-2?

If you say “when ♠s are 5-3”, I will disagree.

Okan Zabunoglu

Then; cashing ♥A first (Tom's line) wins when E=2434, cashing ♥K (your line) wins when E=3424 or 3415.

I don't know which one offers a higher chance.

Besides these 3, I could not find any other winning case (provided that we naturally cash one ♥ first).

EDIT: It also boils down to how many ♠s West (or E) has. Giving West 5 ♠s, cash ♥K; 6 ♠s, cash ♥A.