Anomoly or some valuation principle

[hanp]

I just reran the simulation without all the 10's and 9's. Now 6NT makes only 23% of the time. Bridge is a wonderful game.

[gwnn]

what the

*****

.

[pooltuna]

[hv=d=e&v=b&s=skt53hjt973dcak93]133|100|Scoring: MP

uncontested

 

2NT 3♦

3♥* 3♠

3NT 4♣

4NT ?

* usually only two hearts three hearts with no club and diamond control is possible.[/hv]

 

Amazingly after the slam killer opening you got to bid all three of your suits and partner chose not to raise.

 

Are you worth another bid? Perhaps based on the ♥ T9 etc

my gut feeling is that to make another call you need partner so hold specifically 2 of the top 3 ♥ and since he did not make a 4♥ call over 4♣ I am even more inclined to believe he doesn't have it.

[mikeh]

I would pass. All this discussion of simulations made me think a bit more, but I still pass. I really think that double-dummy analyzes stink when deciding what to do here.

 

Partner has heard us announce a diamond void and wants to play 4N.

 

Double-dummy scenarios never lose to the spade Q if he holds the J. They never lose to Jxxx onside in spades when he has Axx. They never lose to the short club Q offside when he has J10x(x). They never lose to Jxxx on his right when he holds Q10x, and so on. All of these extra chances, that no human would ever get right (that may be an overbid....on some defences, the defence removes the guess or makes the play more plausible), will add non-trivial percentages to the likelihood of 6N making. And when the defence assets are so limited, the defence gains very little from defending double-dummy....heck...they can false-card and make devious plays all they like....dd analyze sees right through them, unlike most human declarers.

 

BTW, the differences between the simulations with and without the 9's and 10's will only partly be because those cards increase the at the table playability of the hands. A lot will be because of the double dummy nature of the play...look at the examples I posited above to see how the club 10 would influence the play of that suit...as does the spade 10 when declarer holds AQx or AJx. In fact, if we attribute one half of the difference to the double dummy advantage, this suggests that bidding slam will be significantly worse than break-even.

[bluecalm]

Yeah, double dummy analysis is not too reliable when it comes to slam decisions.

I saw at least one case when double dummy analysis suggested a slam at ~60% but after reviewing the hands manually I didn't want to be there.

 

I think the only way to learn more about this situation is to generate hands meeting requirements for partner's bidding and then go one by one manually and see if you want to be in slam. Some work but after say 30 hands one should have quite a good idea.

 

I just reran the simulation without all the 10's and 9's. Now 6NT makes only 23% of the time. Bridge is a wonderful game.

 

I saw this scenario in many game decisions. Oftern 4hcp with well placed T's/9's is worth almost as much as 6hcp with low spots. From my experience with simulations it's clear that for example having 5th card in long suit is worth much much less than having good spots when contemplating NT game. For example:

 

Tx KQxx xx K8xxx makes 3NT in 44% of cases opposite 16-17balanced while for:

Tx KQTx T9x KT8x it's 57% and for:

73 KQ53 763 KJ43 it's 54%

[mikeh]

My first simulation was as follows: I gave partner 3253, 3262 or 2263 shape and 20-21 HCP. 6NT made 71% of the time.

 

In the new simulation I gave partner 19-20 HCP if he has 6 diamonds. Also, since I want to know how often 6C will be better when partner has 3 clubs I'm not giving him a 3-2-6-2 distribution.

 

New results: 6NT makes 59% of the time, 6C makes 53% of the time and at least one of them makes 74% of the time.

 

However, I don't think partner will be able to judge which slam is best. So it seems that if we want to move to slam, then we should just bid 6NT.

 

The choice is therefore pass or 6NT, I think it is a close call.

why can't opener be 3=2=4=4 with weak clubs? Is he supposed to NOT bid 4N with AKQx Jxxx in the minors? Or even AKQx Qxxx?

 

I also think that most 21 counts with 5 good diamonds upgrade to 22. AKQ10x is not a holding that I value as 9 hcp :)

 

These comments add to the problem with double-dummy simulations: not only does the programme play infallibly, but different people will have different constraints.

 

Finally, are we agreed that 4♣ was a slam try? Would we be introducing a 4 card club suit in an effort to avoid 3N when we have only game ambitions and 4=5=0=4?

 

If 4♣ carries with it a suggestion of slam interest, there may be some hands within your simulation on which some partners would choose a call other than 4N even with good diamonds.

[Siegmund]

Double-dummy analysis routinely says 31HCP are enough for 6NT to be a good contract. No surprise that it does the same here. There is some truth in that, even allowing for a few too many two-way finesses being picked up; we've all had a LOT of 490s in our lives on hands with considerably fewer points. But it's quite a leap to deliberately bid contracts we know will be very shaky and the rest of the field likely will not find.

 

I am "blessed" with one frequent partner who almost invariably bounces me to a slam if I open a 15-17 1NT and she has a flat 15. I do make between 1/3 and 1/2 of these slams she puts me in; but I would get just as many matchpoints by getting a 490 against the field's 460, without having to struggle to avoid the -50 (or -300).

[Cascade]

I really think that double-dummy analyzes stink when deciding what to do here.

Only if you follow them blindly.

 

Hence the thread.

 

We had the hand last weekend. Partner passed and we made an easy 12 tricks for a poor match point score.

 

I wondered how often 12 tricks would be available so did some simulations some with double dummy analysis. I wasn't completely convinced by the results so asked here.

 

After all everyone knows that forums is much better than double dummy.

[cherdanno]

Actually, did we show a diamond void? Might we not bid the same way with 4=5=1=3 to find a possible 5-3 club fit (and help partner evaluate)?

[Pict]

Presumably 4NT showed 3244. It's time to make your play.

 

I'd pass.

[hanp]

Presumably 4NT showed 3244.

No.

[dake50]

We are getting 4xS + 4XC +3xD +1xH or some other combination??

[hanp]

why can't opener be 3=2=4=4 with weak clubs?

I agree that partner can be 3-2-4-4 with weak clubs.

 

Is he supposed to NOT bid 4N with AKQx Jxxx in the minors? Or even AKQx Qxxx?

 

I don't think that he is supposed to not biid 4NT with AKQx Jxxx in the minors.

 

I also think that most 21 counts with 5 good diamonds upgrade to 22.

 

I agree that most 21 counts with 5 good diamonds upgrade to 22.

 

AKQ10x is not a holding that I value as 9 hcp :P

 

Me neither.

 

If 4♣ carries with it a suggestion of slam interest, there may be some hands within your simulation on which some partners would choose a call other than 4N even with good diamonds.

 

I think we agreed 5 times in a row. A bonding experience.