[jogs]

awm, on 2014-May-08, 18:09, said:

It seems to me that even a small study would be useful. If Woolsey-Stewart results on preempts are significantly positive (and better than their average) it seems like a strong signal.

Of course there will also be some effect on the hands where they don't preempt (hard to measure) and some noise (can be bounded statistically) so a near-zero result won't tell us much.

The key is statistical significance. A small study where nearly every result is good suggests it is a good strategy. Be aware of purity. Your points need to be in your long suits. 6421. HCPs in the two short suits is unlucky.

[whereagles]

Well, I actually have a story of Woolsey-Stewart preempting me and ending up down -500 vs nothing.. lol

[mycroft]

At least part of the issue with undisciplined preempts (what can I say, I play EHAA) is that because of it, there are things you can't do opposite partner's preempts that a more disciplined side can do - like jump to game, or double the push, or even jump to slam with trump Kx and a bunch of tricks (yeah, thanks GIB).

So I think you can't just look at hands where there's a preempt at one table vs a something else at the other (pass, one level lower preempt, maybe even a "10"-HCP opening in a strong club system); you have to look at hands where both pairs did the same thing (with a "standard" preempt) and see how the style affects partner's bidding. Oh and the opponents; when I open 2♦ EHAA, the opponents with the exact same hand will have a different decision to make than the field, even if this time I have AQJxxx and a stiff.

[FM75]

PhilKing, on 2014-May-08, 17:21, said:

As many of you know, I have a searchable database of a bit less than 20K hands form top-level play. I am pretty certain this is not remotely large enough to provide an answer to the questions posed. When I get it up to 100k (and with more functionality) I will start looking at such things, and see if the data pool is large enough. Any methodology that suggests the use of DD analysis is, in my opinion worse than useless - preempts are all about pressure, and even world-class pairs underperform in this area - no pair can reach the par contract against a preempt without the use of coughing and sneezing.

I agree with the answers that point out that looking at whether (say) Woolsey/Stewart show positive results with an aggressive strategy are almost irrelevant for several reasons.

I was not suggesting the use of par contracts or the determination of par contracts. The role of DD is to determine the likely score after the bidding has been modeled. This could be done on a scale that is statistically significant. I agree that 20,000 hands will not contain enough examples of "loose" preemptive bids for meaningful analysis. Bird and Anthias generated 100,000 deals that matched their auctions to determine the average suit distribution for each table position. They randomly selected 5,000 of those for double dummy analysis of opening leads.

I am suggesting that this approach could be done fairly readily, with the caveat that the WS bidding needs to be modeled, and tested against various competitive auction strategies. The results of auctions meeting WS preempts can then compared with the results of other strategies on that set of deals. Ideally you program up the engine(s) and run them until you get enough results that the standard deviation of the differences starts to reach an asymptotic level.

[jogs]

I forgot to add that undisciplined preempts work well when non vul. Would be more careful when vulnerable.

[MickyB]

jogs, on 2014-May-09, 13:56, said:

I forgot to add that undisciplined preempts work well when non vul. Would be more careful when vulnerable.

In before CSGibson flips

[CSGibson]

MickyB, on 2014-May-09, 14:29, said:

In before CSGibson flips

I had decided to let it be - I tried to keep the thread headed in a useful direction for me and not on these stupid tangential points, but one admonishment is all I had in me.

[Mbodell]

bluecalm, on 2014-May-08, 21:03, said:

Maybe it would be useful to email Richard Pavlicek so he could run something along the lines of "preempt with 6carder at one table, pass at the other" on his database.
He did similar studies already, link to one: http://www.rpbridge.net/9x13.htm

I agree this style of study could be useful. I was thinking of http://www.rpbridge.net/9x16.htm which is open 1H versus 4H (4H better in small sample size). But you might want to know preempt at one table versus more passive action at the other (lower preempt/pass) compared to preempt at one table compared to stronger action at the other (higher preempt, opening 1 bid) and compare those. As others have suggested if there is a pair in question you want to study (or even a set of pairs) for this sort of thing you probably also want to look at hands where the same action is taken at both tables because a wide ranging style should be behind a more constrained style when both take the same action.

Of course, there are second and third order effects and some bidding style might be good primarily through the negative inferences when it doesn't come up (which could happen in the bidding or the play). The hope is these would be smaller in effect than the "main" bid being studied, but that isn't clear. For instance a bid that is -0.1 IMPs/hand when it comes up but adds +0.01 IMPs/hand when it doesn't come up through negative inference could be a big winner when its frequency causes it to only come up on 1-2% of hands. There is also how well the style/bid fits into the rest of the system (it could be wide ranging preempts are good in a 2/1 context but bad in precision - or vice versa, say).

[MrAce]

bluecalm, on 2014-May-08, 21:03, said:

Maybe it would be useful to email Richard Pavlicek so he could run something along the lines of "preempt with 6carder at one table, pass at the other" on his database.
He did similar studies already, link to one: http://www.rpbridge.net/9x13.htm

Richard Pavlicek himself is one of the most aggressive players I have ever seen. I partnered him online a lot and off line in Reisinger bam. I looked like conservative compared to him when it came to preempts.

I agree with Philking that simulations are silly and that preempts are all about pressure. If you look at the pars both side can make it will lead you to inaccurate conclusions.

[PhilKing]

If anyone wants to do the work I have filtered out 90 hands where one side opened 3♣ and the other did something else.

[hotShot]

DD Simulations are tricky.
You need to be absolutely clear about what question you want to get answerd and once the simulation has finished you need to recheck if the result you got is really an answer to the question you wanted to ask.

It is possible to answer the question what percentage opps could make game/slam on deals you would preempt. (But it won't tell you if your preempt applies any pressure on opps.)
If that percentage is "low" your preempt style might be to solid.

You could ask for the percentage of preempts in minors where your side could take e.g. 7 tricks.
If that percentage is "low" your preempt style might be to aggressive.

If you ask for the percentage where opps find their best spot after your preempt. DD simulations can't give you the answer.
As for this study DD-Simulations don't seem to be very helpful.

But I think it is obvious that a preempt is reducing the bidding space and by that limiting the amount of information that can be transported by bidding. This will usually result in less accuracy in opps bidding, giving them an opportunity to guess wrong.

[NickRW]

PhilKing, on 2014-May-08, 17:21, said:

Any methodology that suggests the use of DD analysis is, in my opinion worse than useless - preempts are all about pressure, and even world-class pairs underperform in this area

I agree. DD results are reasonable, if not 100% perfect, given that you know what the contract is (should be). The whole point of a pre-empt is that the proper contract will not always be reached.

Another point is that you have to factor in the matter of, when opps do find their contract, you have leaked considerable information about your shape. They might now make their contract when they wouldn't have without the leakage.

And yet another point is that pre-empts sometimes propel opps into making slams that they might well not have bid under their own steam.

Not saying that aggressive bidding, pre-emptive or otherwise, is not winning bridge. Just that there are a *lot* of factors to weigh up.

Nick

[Zelandakh]

One obvious way of using DD would be to run an auction for the hands manually and then assume the DD result for the resulting contract. Obviously not perfect either but faster than also trying to work out the play manually. Could you produce the hands as an East set and a West set plus opponent bidding should one of our pairs feel up to trying this as a challenge, Phil?

[helene_t]

You could also let some SD software like Jack play a zillion boards between three or four teams, playing the same system but different preempt style.

[cherdano]

helene_t, on 2014-May-12, 02:59, said:

You could also let some SD software like Jack play a zillion boards between three or four teams, playing the same system but different preempt style.

Which would tell you very accurately whether it is beneficial to preempt against Jack if your bidding works exactly like Jack's...

[FM75]

cherdano, on 2014-May-12, 05:30, said:

Which would tell you very accurately whether it is beneficial to preempt against Jack if your bidding works exactly like Jack's...

Which is why I suggested that an effective study would need to be done against a range of competitive bidding systems. Of course, your results likely would vary against each, which would show you when you get your best results, and when you get your worst - or conversely, how best to compete from the set of systems measured, against the preemptive style.

[rhm]

I do not quite understand why it would be beneficial to restrict this question to a specific pair.

Richard Pavlicek published results from top level tournament play at http://www.rpbridge.net/rpme.htm where one room opened with 4M and the other with 1M

This could be extended to other preemptive scenarios.
Of course this does not answer all questions, but I doubt that any study will ever be conclusive.
For example you do not know from these results what happens when a preempt is obvious, but one pair is better placed, because its preemptive tendencies are more conservative.

Undoubtedly, just like opening bids restrictions on preempts have loosened over the years.
What puzzles me is that there is a corresponding trend to shy away to take the money.
This remarkable reluctance can only reinforce the trend to preempt on thin air.
The success of the Kranyak team in the trials to the Bermuda Bowl in Bali was in part due to the fact that they did not have this reluctance to play for penalty.

---

What do you bid?

This time it was Kranyak, who preempted and Wolpert, who raised the preempt.
A reigning world champion bid 4♥ and played it there.
In the other room with no preempt 6♦ was bid and made
A simple PASS and an easy defence would have netted six undertricks and 1400.
Well, if even a world champion can not take the money when the contract is six down, I am not surprised that preempts get weaker by the day.

Rainer Herrmann

[helene_t]

cherdano, on 2014-May-12, 05:30, said:

Which would tell you very accurately whether it is beneficial to preempt against Jack if your bidding works exactly like Jack's...

True, if the question is if Kit's style works well for Kit then Jack is obviously irrelevant.

But a Jack study is, at least, easy to interpret. You can avoid confounders (aggresive preemptors may tend to have strengths or weaknesses in specific other areas) and take related effects into account (a pass caries a different meaning for an aggresive preemptor than for a disciplined preemptor). You can also avoid a possible gain from inadequate disclosure (partner knows your preempt style better than opps do).

[hotShot]

rhm, on 2014-May-13, 05:21, said:

I do not quite understand why it would be beneficial to restrict this question to a specific pair.

Richard Pavlicek published results from top level tournament play at http://www.rpbridge.net/rpme.htm where one room opened with 4M and the other with 1M

This could be extended to other preemptive scenarios.
Of course this does not answer all questions, but I doubt that any study will ever be conclusive.
For example you do not know from these results what happens when a preempt is obvious, but one pair is better placed, because its preemptive tendencies are more conservative.

Undoubtedly, just like opening bids restrictions on preempts have loosened over the years.
What puzzles me is that there is a corresponding trend to shy away to take the money.
This remarkable reluctance can only reinforce the trend to preempt on thin air.
The success of the Kranyak team in the trials to the Bermuda Bowl in Bali was in part due to the fact that they did not have this reluctance to play for penalty.

---

What do you bid?

This time it was Kranyak, who preempted and Wolpert, who raised the preempt.
A reigning world champion bid 4♥ and played it there.
In the other room with no preempt 6♦ was bid and made
A simple PASS and an easy defence would have netted six undertricks and 1400.
Well, if even a world champion can not take the money when the contract is six down, I am not surprised that preempts get weaker by the day.

Rainer Herrmann

What to bid now is irrelevant, as the problem was caused by North pass last round.
A bidding system that forces North to pass his 12HCP balanced hand with support or at least tolerance for the unbid suits, definitely puts you in an disadvantage playing against preempts.

[rhm]

hotShot, on 2014-May-13, 06:29, said:

What to bid now is irrelevant, as the problem was caused by North pass last round.
A bidding system that forces North to pass his 12HCP balanced hand with support or at least tolerance for the unbid suits, definitely puts you in an disadvantage playing against preempts.

Strong stuff, considering that (Bocchi-Madala) are the current Bermuda Bowl winners and on anybodys shortlist for the best pair in the world.
I at least have much more sympathy for the initial pass than for the second bid.

Rainer Herrmann