[rhm]

 MrAce, on 2011-January-02, 18:21, said:

-How many of them actually goes down without double dummy defense ? did u check each of them ?

No, I can not with 1000 deals. But double dummy works both ways. Declarer also does not play double dummy. The end result is that these errors tend to cancel each others. Tests have shown that double dummy result come close to average results at the table. At 3NT declarer makes slightly more often than double dummy suggests and in grand slams the defense does slightly better.

Quote

-How do u know if we will stop in 3♥ with 500 of them, did u check 500 hands 1 by 1 and decided those hands would pass 3♥ or did programme do it for u by bean counts ?

No, of course not. I prefer simulation with typical 1000 deals to get statistical valid results. (variation is reasonably small) The drawback is that you can only look at samples but not at every deal and analyze it. The double dummy analyzer(deep finesse) just tells me how many tricks in hearts I get on each deal.
For my analysis I simply assumed that West would accept in 50% of the deals and exhibit excellent judgment and would not accept, unless there were at least 9 tricks available to declarer in ♥ and would always accept, when there were more. My experience is that most player accept invitations on many more hands.

Quote

-Did u make a restriction for hands, especially 10-11 hcps where ♥ suit doesnt worth to overcall at 2 level ? Since we have ♥ K, can pd have anything less than Qxxxxx in your deals ?

No I kept the simulation simple and I can refine it. But if you have 6 of your 13 cards in ♥ and 9-10 HCP only in very few deals will your suit be anything less than Qxxxxx. I can rerun my simulation, but I doubt it makes a difference

Quote

-Did u check what are the results for 3NT in some of those hands where pd can bid 3NT ?

No, but will you and partner know, when to prefer to stay in 3NT, when West holds a six or seven card ♥ suit?

Quote

-Did u check what happens to 4♠ when pd has 4♠ and an accepting hand ?

No. I kept it simple, there will be few deals were partner will have 4 cards in ♠ besides 6 cards in hearts.

Quote

-What kind of hands is pd raising to 4, assuming my hand didn't cue but just raised to 3♥ ? You need to have a decent hand to bid 4 when pd only raises.

My analysis assumed simply that partner would bid 4 ♥ on 50% of the random deals and that he would always accept when there are 10 tricks or more available and never accept, if 3♥ would already go down. In other words my analysis assumed partner would exhibit perfect judgment.
The results in practice will be less good for the inviter.

Quote

-Did u restrict, at least a very good portion of hand to only ♣ lead ?

Double dummy results work both ways. Perfect defense and perfect declarer play. For example declarer always finds a trump queen. Checks have shown that these assumption are wrong of course, but tend to cancel each others and double dummy results come very close to what will happen on average in practice, when a board is played a number of times (except for grand slams).

I am open, I did not try to make assumptions so that the simulation results would support my assessment. I was interested in an unbiased outcome. I can change or refine any of my assumptions and rerun the simulation.

Rainer Herrmann

[cherdano]

 rhm, on 2011-January-03, 03:21, said:

Tell me what you consider unrealistic. No big deal.
The assumption that a typical Bridge player would accept in about 50% seems realistic if not conservative to me.
Many accept, unless dead minimum for their previous call, that is in many more cases than 50%. In fact I myself belong to the invite heavy, accept often school

Why do you call a 3H raise an invite? I would call it a raise, promising less than invitational values, since with invitational values advancer would bid 3C.
But 655321 already told you that and you ignored it, so I am not sure there is much point in saying it again.

[rhm]

 cherdano, on 2011-January-03, 05:15, said:

Why do you call a 3H raise an invite? I would call it a raise, promising less than invitational values, since with invitational values advancer would bid 3C.
But 655321 already told you that and you ignored it, so I am not sure there is much point in saying it again.

I am not sure what you are aiming at. Are you claiming that a ♥ raise is preemptive and non-invitational or do you like splitting hairs?
There may be a difference in strength between a cuebid of 3♣ and a ♥ raise, but there is no doubt that absent of some very special agreements anyone at the table would consider a raise of an over-call of 2♥ to 3♥ as an invitation to 4♥.
Call it what you like. This is a discussion group on Bridge not on semantics.
If you and 655321 claim that you would rarely accept a 3 ♥ "raise", fine. But this means, you will still turn a lot of pluses in minus scores or small minus scores in larger ones.

Let's do the mathematics based on the simulation.
The best you can theoretically do is

You accept in exactly 29.5% of all cases, where you make game.
You would still have 382 minus scores instead of 116 and the minus score on the 116 deals would be larger.

But in practice even if you accept in the correct frequency here, you will do much worse and miss many games and reach many, which will go down and on some of these you will get doubled.
Is there really a net gain?

Rainer Herrmann

[whereagles]

 rhm, on 2011-January-03, 03:58, said:

double dummy works both ways. Declarer also does not play double dummy. The end result is that these errors tend to cancel each others. Tests have shown that double dummy result come close to average results at the table. At 3NT declarer makes slightly more often than double dummy suggests and in grand slams the defense does slightly better.

Where did you get those tests results? Common sense says double dummy gives the defenders more chances.

[rhm]

 whereagles, on 2011-January-03, 06:04, said:

Where did you get those tests results? Common sense says double dummy gives the defenders more chances.

Have a a look at

http://crystalwebsit...my_accurate.htm

To cite from this site:

"The most important general finding is that double dummy analysis is very accurate as compared to actual play from OKBridge."

For 4 level contracts in trump suits the actual declarer made 9.84 tricks and the double dummy result was 9.85 tricks (based overall on 30 million plays). So there was no declarer advantage in actual play for 4♥ contracts.

Rainer Herrmann

[whereagles]

Thanks. I'll have a look at it asap.

[hotShot]

 whereagles, on 2011-January-03, 06:04, said:

Where did you get those tests results? Common sense says double dummy gives the defenders more chances.

The numbers indicate that the defense loses tricks to double dummy at low level contracts,
while the declarer loses playing slams.

This correlates with the amount of information the defense and the declarer possesses.
In a low level contract the defense has almost no information, while it is often unavoidable for the declarer to lose a specific trick playing a slam.

[MrAce]

Rainer; first of all ty for the answers.

-I think someone with 6 cards suit and stoppers in opponent suit, trying 3NT is not uncommon. But i see your point.

-2 level overcalls are prefered to be made by 6 cards, i agree, but also almost all unbalanced hands, especially after precision 2♣ (like 5♥+4♠ hands, 5♥+4♦ hands,5♥+5♦ hands and some rare 6-5 with ♦.

-Also some hands where overcaller has 7♥ but less than 10 hcp (8-9 hcp)and choosed to not preempt due to hand being too good for it, or suit being poor and hcps are on side (side ace, or void or well placed honors etc..)

-Opener's pd not raising ♣ does not mean too much, and a deceptive factor,since it is common to hide the fit with very weak hands when one believes showing it will help opponents to judge their hand better. And trust me it works more often than showing and finding a good save. I witness a lot of hands everyday, where one side would never reach to the game or slam if they were let alone, but preempt and raises may make it atractive to bid it.

-In some of the hands, where opener has 11-12 hcp and overcaller has 10-11 hcp, opener's pd will then have 8-9 hcp and probably would take some action instead of pass, that also eliminates some hands where pd is as weak as 9-11 hcp.

Now, on the other hand, if you read my post again where i reacted you, you will see that i was against being picky from BOTH overcaller and responder side as a response to your previous post. If you are as picky on the overcaller side, i believe a good portion of these hands that fails at 3 level, will be eliminated at the first place. Because as oppose to computer, humans (at least good players) make their calls by looking at the hand rather than just looking at hcps.

And i also think accepting with % 50 of the hands is wrong assumption. 3♥ is not a cue. Where a cue can still be just an invite and willing to pass if pd bids 3♥ with a dead minimum, a 3♥ asks for more than a normal or ave+ overcall to raise to 4.

I believe if test is made more accurate, the chances of making game (IF accepted by overcaller) will be above % 35 and thats what required in IMPS vulnerable if i am not wrong.

[gwnn]

rhm do you also call

1♥-p-2♥

an invitation to 4♥? or just a raise?

people basically bid

2♣-2♥-p-3♥
or
1♠-2♥-p-3♥

on the same set of hands as 1♥-p-2♥. I freely concede that the two sets are not identical, however they are extremely close and it would be very hard to think of a better definition than 'single raise' of the latter two auctions. Just as

1♥-p-2♥ is called a single raise and
1♥-p-3♥ (in SAYC or SEF) is called an invitational raise,

it makes sense to call

3♥ in the above auctions as a raise or a single raise and the cues (3♣ and 2♠ respectively) as invitational raises.

[hotShot]

 gwnn, on 2011-January-03, 13:53, said:

people basically bid

2♣-2♥-p-3♥
or
1♠-2♥-p-3♥

on the same set of hands as 1♥-p-2♥. I freely concede that the two sets are not identical, however they are extremely close and it would be very hard to think of a better definition than 'single raise' of the latter two auctions.

I tend to disagree.
Playing precision 2♣ and 1♠ are limited bids and RHO's pass is a limited bid. Why should someone holding a "single raise" bid 3♥ if his side can buy the contract in 2♥ and he can still bid 3♥ if LHO decides to compete?
Ain't it more useful to have an invitation and some sort of a GF/ slam interest bid available?

[gwnn]

It ain't. Partner can have a lot of hands where game is excellent opposite a single raise.

[rhm]

 gwnn, on 2011-January-03, 15:09, said:

It ain't. Partner can have a lot of hands where game is excellent opposite a single raise.

True of course, but the mistake you are making is that you seem to assume anything can give you back the one level. Nothing can.
The precision 2♣ opening has a lot of drawbacks, but it also has serious preemptive value, which a 1♣ opening does not.
A raise of an overcall from 2♥ to 3♥ is not the same as from 1♥ to 2♥.
Raising to the three level risks going down, where 2♥ might have made.
So raising 2♥ to 3♥ is essentially a calculated risk between missing game and going down in 3♥.

I would certainly not raise with all hands, which raise from 1♥ to 2♥.
A raise from 2♥ to 3♥ is closer to something between a single raise and a limit raise.
Also whether a cue bid of 3♣ should be reserved for stronger ♥ support can be argued.
For example should a new suit by advancer, e.g 2♠, be forcing? Lawrence recommends no. If not how do you force and how do you aim at 3NT?

Rainer Herrmann

[gwnn]

If I'm making a mistake, it is not a mistake "I" am making. It is the mistake almost all top players today are making. Nobody really good plays 3♥ as anything substantially different than a single raise. I don't have any reliable statistics to back this up, unfortunately, but it's just my impression from vugraphs and other kibitzing.

As I said, the two sets of hands are not perfectly identical, since there will be some hands that pass a 2 level overcall that would have raised 1♥. Also sometimes you have to raise to 3♥ on a doubleton.

Whether a cue bid of 3♣ should be reserved for stronger ♥ can be argued, but perhaps not in this thread. The opening poster's system is crystal clear - 3♣ is a strongish raise, but by no means game forcing or a slam try. Therefore 3♥ is not a limit raise. It might make sense to play 2♠ in this auction as non-forcing and 3♣ as a forcing cue (I think it makes sense but I don't like it), but that is not what the opening poster plays so why do we have to discuss it here?

edit: by the way, I don't think passing from east is completely losing bridge all the time, but when vulnerable at imps it looks quite irresponsible.

This post has been edited by gwnn: 2011-January-04, 04:43

[gnasher]

gwnn said:

Nobody really good plays 3♥ as anything substantially different than a single raise.

I think that depends on what you mean by a "single raise". If you're talking about the constructive raises made by people who have two ways to raise 1♥ to 2♥, you're probably right.

If you're talking about the full set of hands that bid 1♥-2♥ in a system where there is only one raise, I think something like the bottom 1/3 of 1♥-2♥ raises would pass the 2♥ overcall. When you raise a 1♥ opening to 2♥, you have several ways to gain - either side might be able to make either a game or a partscore. After (2♣) 2♥ (pass), the only way that a raise gains is by getting you to a making game - both opponents are limited, and responder, having failed to raise or make a negative double, is unlikely to want to compete. Most people would raise 1♥ to 2♥ with xxxx Kxx Qxxx xx but would pass this 2♥ overcall.

This post has been edited by gnasher: 2011-January-04, 05:38

[gwnn]

Yes and I have written about this in my previous post(s). I guess I was understating the differences between the two sets, but it should be closer (in my opinion) to a single raise than a classical constructive raise of 8-10. But we're (gnasher and I) splitting hairs here - rhm thought 3♥ is best played as an invitation, accepted by 50% (and probably more than 50%) of the hands. I disagree and I think you (gnasher) disagree too.

[gnasher]

Well, personally I usually play the 3♥ as invitational, and the cue-bid as a non-specific game-force, but I'm aware that I'm in a minority, and also that what I do might not be best.

[gwnn]

In that case I recount my statement that started with "nobody really good" and will substitute it with "only a small minority of really good players"

[rhm]

 gwnn, on 2011-January-04, 06:38, said:

Yes and I have written about this in my previous post(s). I guess I was understating the differences between the two sets, but it should be closer (in my opinion) to a single raise than a classical constructive raise of 8-10. But we're (gnasher and I) splitting hairs here - rhm thought 3♥ is best played as an invitation, accepted by 50% (and probably more than 50%) of the hands. I disagree and I think you (gnasher) disagree too.

Basically gnasher said what I said. You are splitting hairs not him.

And to cite you

"I don't have reliable statistics to back this up, unfortunately, but it is my impression from vugraph and other kibitzing" that 3♥ is rarely passed after an overcall of 2♥ has been raised. If anything an estimate of 50% is on the conservative side. Nobody likes to stop in 3♥. It is rather similar to a raise of 1NT to 2NT. Otherwise a raise would hardly make sense, since you risk going down with infrequent gains.

It comes down to the following philosophy

Do you invite heavy and accept often or do you invite weakly and accept infrequently.

The first philosophy misses a few games but has few minus scores.
The second may miss fewer game (not clear, because the ♥ raise is wide ranging) but what is certain it will incur many more minus scores .

Rainer Herrmann

[gwnn]

Please just answer me this: if 3♣ is played as a raise which could include a 10 point hand with no shortness, what do you think 3♥ is played as? How often do you think it is accepted?

[hotShot]

rhm showed that a hand that is better than OP's west hand (because it has 6+♥) will only have a 29.5% chance for game opposite likely east hands when opener has a precision 2♣ opening and his partner passes.
His results indicate that that such a west hand should bid over 2♣ but east should usually not bid on.

It is logical to conclude that east should pass, if west can have weaker hands than this.

The simulation does not say if the actual west hand should bid over 2♣.

My problem with this simulation is that the posting says nothing about the the east hand.

 rhm, on 2011-January-02, 17:09, said:

I decided to do a simulation, with fairly favorable assumptions about West (e.g. always more than 5 cards in ♥):

South: Precision club opener as described 11-16 HCP, either at least 6 cards in ♣ or 5 cards in ♣ and a 4 card major.
West: No 5 card ♠ suit, either 6 cards in ♥ and 10 to 17 HCP or 7 cards in ♥ and 10 to 15 HCP. But otherwise West could be very distributional.
North: Less than 4 cards in ♣ (no raise), no further restrictions

If the east hand is set to what it actually is on the OP, than the best move for East is to pass 2♥.

If the east hands are unrestricted in this simulation, than the results contain a lot of east hands that are weaker than the posted one.