[billw55]

How many in the field bid 6?

[aguahombre]

 rhm, on 2013-September-13, 03:13, said:

A simulation(1000 deals) reveals that the North hand makes 6NT opposite 12-13 HCP, balanced with no 4 card major, 58.2% of the time.
Average number of tricks was 11.6

But we know that on such slam deals DD favors declarer. In practice 6NT will make roughly 50% of the time.

 Fluffy, on 2013-September-13, 05:25, said:

You should also add that best 13 balanced hand will accept.

And perhaps take note that a similation including 12 HCP opposite an invite which won't accept is mildly interesting ---the pertinent South hands are good 13-14 and the judgement by South of what is "good" on this Stayman auction might be beyond a machine or an Agua.

[Fluffy]

I think you have it wrong, good 13-14 are not interesting as we are debating jump to 6NT vs 4NT invitation, since opposite good 13-14 both actions end up in 6NT the decision is irrelevant, only matters the 12-13 hands.

[mikeh]

 cloa513, on 2013-September-12, 18:08, said:

If you are going to do quantitative raises- then South has to accept with a good 13 HCP- quant bids aren't science.

Your statement is correct, imo.

However, if you are implying that S has a 'good 13', we part company. The lack of Aces offsets the possession of the 10. The major suit holdings are bad, and, yes, I did see the stayman bid.

This is, imo, a mediocre 13, not a 'good' 13.

[Fluffy]

I think the ammount of aces is almost irrelevant when accepting a quantitative. Doubleton queen bothers me much more.

[VM1973]

 johnu, on 2013-September-13, 00:13, said:

Some players don't believe in simulations. I think you can gain a lot of insight into hand evaluation using simulations. Of course, the obvious problem is that every hand is different and you don't have a simulator available when playing a hand (except playing online )

The N hand is officially 19 HCP but with 3 aces, and concentrated honors, it's more like a 20 or 21. The S hand is 13 HCP, but with a super quacky hand, it's more like 11 or 12. Combined, it's about an average looking 32 combined HCP. From N's point of view, S is likely to have a bunch of quacks, and S should know that N has to have lots of aces and kings.

I first started with the actual north hand and simulated playing 6NT for 200 hands for south assuming no 4 card major.

12 HCP - 50%
13 HCP - 64%
14 HCP - 89%

From North's point of view, 6NT looks good opposite 13, excellent opposite 14, but not worth bidding opposite 12 (and with upgrading epidemic, there's probably some 11's to consider). With standard bidding, I don't see anything else to do than an invitational 4NT by north.

Then, I started with the actual south hand and simulated 200 hands for north with at least one 4 card major and 19 or 20 HCP.

19 HCP - 50%
20 HCP - 80%

Opposite an average 19 HCP hand, 6NT isn't worth bidding, but opposite an average 20 HCP hand, 6NT is excellent.

That seems to indicate that south should take another move to show that while not a maximum, they have 13 and not a bare minimum. If south makes an invitational move over 4NT, should north make the final mistake accept? With prime connected aces and kings, I think so even though on the lower end of the HCP range.

I rarely misbid a hand when partner's cards are face up.

Obviously there are problems with simulations, but assuming that the simulations are accurate I think that your interpretation of them is faulty.

If North were able to look at his hand and know that he has the percentages you stated above (although this is not possible at the table) then he could reason as follows:

If my partner has 12 (we assume 34% of the time) then the chance of making the bid is 50% so 17%
If my partner has 13 (we assume 33% of the time) then the chance of making the bid is 64% so 21.1%
If my partner has 14 (we assume 33% of the time) then the chance of making the bid is 89% so 29.4%

So then we have a 67.5 percent chance of making the slam opposite some sort of a 12-14 NT hand. This calculation seems to imply that if anyone is going to make a forward move it should be North, not South. Now at IMPs I think it's pretty clear that 67.5 percent isn't enough to risk a solid game in favor of a slam. At matchpoints, however, it's far more reasonable to assume that the entire field will be in 4NT making something so then you have a 67.5 percent chance of getting a top vs. a 32.5 percent chance of getting a bottom. Those are good odds.

That being said, I'm not aware of a special hand evaluation method that could be used to evaluate your NT shaped hand for slam purposes other than the standard methods already being employed.

[gnasher]

 VM1973, on 2013-September-13, 10:49, said:

Obviously there are problems with simulations, but assuming that the simulations are accurate I think that your interpretation of them is faulty.

If North were able to look at his hand and know that he has the percentages you stated above (although this is not possible at the table) then he could reason as follows:

If my partner has 12 (we assume 34% of the time) then the chance of making the bid is 50% so 17%
If my partner has 13 (we assume 33% of the time) then the chance of making the bid is 64% so 21.1%
If my partner has 14 (we assume 33% of the time) then the chance of making the bid is 89% so 29.4%

So then we have a 67.5 percent chance of making the slam opposite some sort of a 12-14 NT hand. This calculation seems to imply that if anyone is going to make a forward move it should be North, not South. Now at IMPs I think it's pretty clear that 67.5 percent isn't enough to risk a solid game in favor of a slam. At matchpoints, however, it's far more reasonable to assume that the entire field will be in 4NT making something so then you have a 67.5 percent chance of getting a top vs. a 32.5 percent chance of getting a bottom. Those are good odds.

At IMPs non-vulnerable, the break-even point for bidding a slam is 50%, as long as there's no risk of going -3. You gain 11 if you make, and lose 11 if you go down. Hence 67.5% would easily be enough to bid slam.

In fact, you're not doing the right calculation. If we're considering whether to invite with 4NT or just bid 6NT, the 14-counts and some of the 13-counts are irrelevant, because you'll get to 6NT regardless of what you bid. The hands where partner will pass 4NT are the 12-counts and the weaker 13-counts. On these hands, if we assume that the bad 13-counts are on average a bit better than the 12-counts, slam will be somwehere between 50% and 64%, and therefore good enough to bid.

Edit: All of this assumes, of course, that you regard the double-dummy figures as a close approximation to declarer's actual odds.

This post has been edited by gnasher: 2013-September-13, 11:34

[PhilKing]

 rhm, on 2013-September-13, 03:13, said:

A simulation(1000 deals) reveals that the North hand makes 6NT opposite 12-13 HCP, balanced with no 4 card major, 58.2% of the time.
Average number of tricks was 11.6

But we know that on such slam deals DD favors declarer. In practice 6NT will make roughly 50% of the time.

Rainer Herrmann

Exactly. This is the kind of deal where a DD sim will be bias toward declarer.

Is the 50% figure a best guess, or something more?

[johnu]

 VM1973, on 2013-September-13, 10:49, said:

Obviously there are problems with simulations, but assuming that the simulations are accurate I think that your interpretation of them is faulty.

If North were able to look at his hand and know that he has the percentages you stated above (although this is not possible at the table) then he could reason as follows:

If my partner has 12 (we assume 34% of the time) then the chance of making the bid is 50% so 17%
If my partner has 13 (we assume 33% of the time) then the chance of making the bid is 64% so 21.1%
If my partner has 14 (we assume 33% of the time) then the chance of making the bid is 89% so 29.4%

So then we have a 67.5 percent chance of making the slam opposite some sort of a 12-14 NT hand. This calculation seems to imply that if anyone is going to make a forward move it should be North, not South. Now at IMPs I think it's pretty clear that 67.5 percent isn't enough to risk a solid game in favor of a slam. At matchpoints, however, it's far more reasonable to assume that the entire field will be in 4NT making something so then you have a 67.5 percent chance of getting a top vs. a 32.5 percent chance of getting a bottom. Those are good odds.

That being said, I'm not aware of a special hand evaluation method that could be used to evaluate your NT shaped hand for slam purposes other than the standard methods already being employed.

First, your assumptions about high card distributions are wrong. You will have 12 HCP more than 41% more than 14 HCP and 16% more than 13 HCP. 2nd, North has probably the only available forward move for anybody not playing some kind of sophisticated relay system, showing a balanced 18+ to 20 by inviting with 4NT. South should pass with 12 HCP and that will be the right decision in the long run since slam is a tossup. Why would you want to be in a 50-50 slam? If South had the maximum 14 HCP, 6NT would be the next bid after 4NT. So it's up to South with 13 HCP, plus or minus any adjustment factors to either pass, jump to slam or transfer the final decision to partner.

EDIT: Simulations aside, this basically comes down to the historical guidelines for bidding 6NT, 33 HCP for 2 balanced hands. If I could add another requirement, it would be having 10 controls or more (ace = 2 controls, king = 1 control). If you are missing an ace and a king, slam is unlikely to be better than a finesse, and if they are in the same suit ...

[VM1973]

 johnu, on 2013-September-13, 11:37, said:

First, your assumptions about high card distributions are wrong. You will have 12 HCP more than 41% more than 14 HCP and 16% more than 13 HCP. 2nd, North has probably the only available forward move for anybody not playing some kind of sophisticated relay system, showing a balanced 18+ to 20 by inviting with 4NT. South should pass with 12 HCP and that will be the right decision in the long run since slam is a tossup. Why would you want to be in a 50-50 slam? If South had the maximum 14 HCP, 6NT would be the next bid after 4NT. So it's up to South with 13 HCP, plus or minus any adjustment factors to either pass, jump to slam or transfer the final decision to partner.

EDIT: Simulations aside, this basically comes down to the historical guidelines for bidding 6NT, 33 HCP for 2 balanced hands. If I could add another requirement, it would be having 10 controls or more (ace = 2 controls, king = 1 control). If you are missing an ace and a king, slam is unlikely to be better than a finesse, and if they are in the same suit ...

These numbers that you so cavalierly toss about are only true in a vacuum. Yes, if you pick up any random hand your chances of having 12 HCP are 41% more than having 14 HCPs. However, once you know that your partner has 19 HCPs these numbers are no longer valid. They would need to be recalculated at the table and this would be difficult under the time constraints available.

[gwnn]

Yes if you have 19, the odds favour the 12 counts even more. So assuming 34-33-33 is even wronger than before.

[rhm]

 PhilKing, on 2013-September-13, 11:36, said:

Exactly. This is the kind of deal where a DD sim will be bias toward declarer.

Is the 50% figure a best guess, or something more?

No its my best guess

Rainer Herrmann

[EricK]

 billw55, on 2013-September-13, 06:55, said:

How many in the field bid 6?

It was about 50/50

[VM1973]

 gwnn, on 2013-September-13, 13:26, said:

Yes if you have 19, the odds favour the 12 counts even more. So assuming 34-33-33 is even wronger than before.

The main point of my post was simple: That if any person should have taken further action, it was North.

Now perhaps you disagree with my reasons for thinking so. Perhaps you think my calculations are completely off. Perhaps you think that I should have analyzed it in a different way.

Nevertheless, do you disagree with the main point, namely that knowing that opposite a 12-count he has a 50-50 shot of making 6NT and opposite a 13+ count his odds improve, that the bidder should have bid 6NT had he been able to work this out at the table? If not, then why are you so eager to argue about inconsequentials?

Does it really matter if the probability is 67.5 percent, 66 percent, 63 percent, or even 53 percent as long as it is over 50 percent? I swear if you met someone with a gun and he said, "I was thinking about flipping a coin to see whether or not I should kill you, but since the odds of you winning are 80 percent, I figured I'd just let you go" then I'm sure you would argue the point that it's really 50-50 and convince him to flip the coin to see if he should shoot you or not.

So since you enjoy arguing about inconsequentials, why don't we start by arguing about your use of the word wronger, which is incorrect. You should have said more wrong. In fact a simple Google search showed that wronger occurs only 269,000 times worldwide whereas more wrong occurs more than 1.27 million times. Perhaps we can even have 20 messages back and forth and start insulting each other's mothers.

It will provide a good distraction from the apparent failure of the Wonk point count and the general lack of solutions offered on this forum.

[PhilKing]

 VM1973, on 2013-September-14, 09:06, said:

Nevertheless, do you disagree with the main point, namely that knowing that opposite a 12-count he has a 50-50 shot of making 6NT and opposite a 13+ count his odds improve, that the bidder should have bid 6NT had he been able to work this out at the table? If not, then why are you so eager to argue about inconsequentials?

But the odds are not nearly that good.

The DD sim telling us we have about a 50% chance opposite a 12 count is simply not real bridge. Even RHM, who is one of the biggest proponents of DD sims, said that you need about a 58% result to equate to a 50% at the table success rate for 6NT hands based on high card power.

Note his results gave 58% for 12 and 13 counts (and we can assume some, and in my view most, of these would accept an invite. Even good 12s with a five-card minor should bid 5m over 4NT).

Also, wrong use of "wronger" is generally ironic, or a reference to this

[ggwhiz]

 EricK, on 2013-September-13, 00:53, said:

It wasn't the only one of the evening - 1nt + 4 on 13 opposite a prime 11 springs to mind. However we came top almost 10 % above pair in second, so agreeing to play together wasn't a complete error!

I can't bid to slam with any confidence but the timing of these hands may be key. At MP's and looking at a great scorecard North may have been astute to just protect an obvious lead late in the session.

[gwnn]

Welcome back VM! I was replying to your whole (second) post, and see PhilKing's post for why the percentages are actually crucial. Shrugging and saying "33-33-33 is OK I guess" is silly but when someone gives a significantly better estimate and you just point out (unwittingly) that your estimate is even worse than he pointed out so we should just use your estimate anyway, how do you call that? Maybe instead of criticising the usage of an expression coined by Asimov, you could help me with this one? I'm thinking of ignorant or infantile but I'm sure there are better words out there...

edit: But I will oblige and argue about inconsequentials, too, that is indeed one of my guily pleasures. Merriam-Webster gives "wronger" as the only comparative of "wrong," and the HCP evaluation of 4-3-2-1 is actually called Milton-Work. Is this latter mistake just a sloppy misspelling or some juvenile sex joke?

[TWO4BRIDGE]

 EricK, on 2013-September-12, 15:13, said:

First time partnership, so no real system discussion:

---

1NT was 12-14, 4NT was quantitative.

Clearly 12 tricks are easy on any lead.

Just a perfect fit, or did either partner misjudge the hand?

Slam wouldn't look so hot if North's minor suits were reversed ....

[Mbodell]

I really think both players bid fine and that blaming or pushing either is resulting.

N hand is not strong enough to force, invite is right (4333, no T, having staymaned for the majors where most of the strength is).

S hand is, I'd guess, around 40% of the nt range (I.e., better than about 40% of the 1nt openers, worse than 60%) given the stayman and invite. There's no good spots outside ♦, no 5 or 6 card suit, no A. I think that's a not accept. You can picture partner with 19 or 20 points. If he has 2 A, you are down. If he has 4 A, that is 16 hcp, and he'll only have 3 or 4 more and you'll only have 8 tricks which becomes 10 when you give him a major K but that is 19 hcp in AAAAK and you still only have 10 tricks. With 3 if he is missing a major A he must have the K or you are down. If he does have AAAK with the major missing that only gives you 8 tricks (with 1 loser) and he's used 15 points. The other major K gives you 10 tricks and 18 points, and a major J or the club Q would be the 11th trick but likely use all his points. If he's missing the club A, then AK,AK,A is 18 HCP and 10 tricks. A major J would set up an 11th trick with the ♣ hook for 12. A ♣Q would set up an 11th trick in clubs, but use 20 hcp and still be a trick light. Three A with the diamond A missing is pretty great for you as AK,AK,A gives you 11 tricks and him just 18 hcp and any major J is 12 (unless he has AKJ tight of a 3 card major) and the club J gives a finesse for 12. Replacing a major K in any of these with QJ of clubs is usually worse but could be about the same. Overall, that isn't a great situation.

[johnu]

 VM1973, on 2013-September-14, 09:06, said:

The main point of my post was simple: That if any person should have taken further action, it was North.

Nevertheless, do you disagree with the main point, namely that knowing that opposite a 12-count he has a 50-50 shot of making 6NT and opposite a 13+ count his odds improve, that the bidder should have bid 6NT had he been able to work this out at the table? If not, then why are you so eager to argue about inconsequentials?

I do disagree with your main point. Several posters think single dummy results will be several percentage points less successful than double dummy. If it's true, nobody has a real handle on what the difference really is for a particular hand. IMHO, the difference might be smaller than the variance due to your choice of assumptions, and your skill as a declarer (and maybe the defenders skill) will also have a larger impact on how successful the contract will be.

For sake of argument, let's say that based on the skill level of opener and the opponents, the single dummy expectation from North's point of view is exactly 50% if opener has 12 HCP. Even with the context of 12 HCP hands, there are "good" 12 pointers and "cr*p" 12 pointers. Suppose they are equally likely and "cr*p" 12 pointers have a success rate of 40%, while "good" 12 pointers have a success rate of 60%. Do you think it makes no difference if opener can reject the slam try with the cr*p 12 pointer?