[ulven]

Using http://www.bridgecap...downloadDD.html and sampling 3614 deals from BBO Vugraph archives, some people here might be interested in statistics using the DDS filter.
Especially those that look less favorably on Precision 1D/2C openings in a 5-card M framework.

I've just picked some samples from stats (teaser).

Meckwell overall (3614 deals): +0.39 imps/board (compared to teammates)
MW open (1899): +0.42
Opps open (1711): +0.33

1C-opening (360): +0.54
no overcall (73): +0.44
simple overcall (147): +0.93
they overcall/they declare (41): +2.49

1D-opening (521): +0.70
no overcall (168): +0.57

Both 1C and 1D is bigger imp-gainer when opps interfere. Simple overcall vs strong 1C especially bad tactic .

1NT-opening (321): -0.31
2C-opening (66): +1.36

Opps open bidding, MW don't overcall (702): +0.21
Opps open, MW make single jump overcall (51): -0.57
Opps open, MW overcall, but opps declare (515): +0.73

Single jump overcalls seems significantly ineffective. Getting in there with simple overcalls on lesser values seems big winner.
It's possible to dig deeper with time and reasonably fast computer and maybe find some more interesting facts behind the numbers.

[wclass___]

Wow. It is an excellent amazing and free software that i didn't know about. Thanks for sharing.

I don't really care about DD, but ''filter deals'' is a monster.

[Zelandakh]

It seems that MW do better than their opps on deals where more judgement is required - who'd of thunk it?

[robdixon87]

Did anyone else see the thread title and get excited that it might be to do with something like an Erdos number, or was that just me?...

[wyman]

robdixon87, on 2012-August-09, 08:08, said:

Did anyone else see the thread title and get excited that it might be to do with something like an Erdos number, or was that just me?...

Ha, yeah that was my thought too. "Well, I beat X, who beat Y, who beat Z, who beat Meckwell in Gatlinburg!"

[ArtK78]

While I believe that the Precision 2♣ opening bid is one of the weaknesses in the system, it stands to reason that if it comes up there will be a net gain, as it does describe a very specific hand type. Most times that an unusual bid comes up which describes a specific hand type, it will result in a gain over players who are not employing that call.

Typically, if MW open 2♣, their opponents would be opening a standard 1♣. Since the 2♣ opening is far better defined than a standard 1♣ opening, it stands to reason that it puts MW at an advantage ON THAT HAND.

[Free]

ArtK78, on 2012-August-09, 08:43, said:

While I believe that the Precision 2♣ opening bid is one of the weaknesses in the system, it stands to reason that if it comes up there will be a net gain, as it does describe a very specific hand type. Most times that an unusual bid comes up which describes a specific hand type, it will result in a gain over players who are not employing that call.

Typically, if MW open 2♣, their opponents would be opening a standard 1♣. Since the 2♣ opening is far better defined than a standard 1♣ opening, it stands to reason that it puts MW at an advantage ON THAT HAND.

A standard precision 2♣ opener (4M-5♣ or 6+♣) is indeed a weakness, but MW play it as 6+♣. This improves the opening imo, because they can preempt more accurate and the chance of missing a 4-4M fit is a lot smaller.

[semeai]

Free, on 2012-August-09, 12:13, said:

A standard precision 2♣ opener (4M-5♣ or 6+♣) is indeed a weakness

Do you (or ulven) have data on this?

From the Fantunes Revealed thread and book, it seems as though their 2♣ opener does well, even though it's similar to what you describe. Maybe comparing "all precision 2♣ openers by pairs that allow 5 clubs - 4 other" and "Fantunes 2♣" we can get deduce something about the contribution of Fantunes bidding judgement, such a commodity in the other thread. (This would involve subtracting off the contribution of their card play, which we at least have a suggested number for in that thread.)

[semeai]

robdixon87, on 2012-August-09, 08:08, said:

Did anyone else see the thread title and get excited that it might be to do with something like an Erdos number, or was that just me?...

wyman, on 2012-August-09, 08:11, said:

Ha, yeah that was my thought too. "Well, I beat X, who beat Y, who beat Z, who beat Meckwell in Gatlinburg!"

Your collaboration distance (as partner or teammate) from Meckwell might be interesting too.

Meckwell to Hamman to Lall gives us an in on the forums.

[JLOGIC]

You have a sample size of FIFTY ONE on their jump overcalls. lol.

[ulven]

JLOGIC, on 2012-August-10, 03:26, said:

You have a sample size of FIFTY ONE on their jump overcalls. lol.

I used the phrase 'seems' when commenting on the statistic.
If that statement qualifies as an 'lol' then you must be spending a significant number of hours each day having a real blast. Happy to contribute to that bliss.

[han]

ulven, on 2012-August-10, 03:57, said:

I used the phrase 'seems' when commenting on the statistic.

You did well with the phrase "seems", but you did poorly when you wrote "significantly".

I estimate that -0.57 average IMPs on 51 jump overcall hands is less than half a standard deviation. So it does not seem significant to me.

[JLOGIC]

ulven, on 2012-August-10, 03:57, said:

I used the phrase 'seems' when commenting on the statistic.
If that statement qualifies as an 'lol' then you must be spending a significant number of hours each day having a real blast. Happy to contribute to that bliss.

so significantly was the opposite of what you were trying to say?

Go onto a poker forum and tell them that you win 10BB/100 hands over even a 1,000 hand sample and say that you "seem to be a significant winner." You will get similar reactions. If you tried that for a 50 hand sample...

Sorry if I offended you by thinking that it was funny that you seemed to think there was significant meaning to meckwell losing 29 imps over 51 hands where they made a weak jump overcall!

[JLOGIC]

Zelandakh, on 2012-August-09, 07:36, said:

It seems that MW do better than their opps on deals where more judgement is required - who'd of thunk it?

Also, this.

I think ulf was joking when he suggested that this data suggests overcalling is a bad tactic, but for those who don't get it the whole point is:

If meckwells opps are overcalling, it's likely that meckwells teammates are overcalling. Often that will lead to a competitive auction *at both tables*, maybe RHO will preempt to a high level etc. So part of this is that meckwell is judging better than the other table when the auctions are competitive, not a shocker. This does *not* mean that not overcalling would have been a better strategy, meckwell would probably be winning even more imps if one didn't overcall against them with a normal overcall since uncontested auctions are easier than highly contested ones.

Then there's the whole play thing, maybe meckwells cardplay gains them more imps when the auction has been contested because their cardplay ability combined with their balls allow them to take better advantage of the info provided to them by the bidding to pick you off in the play than the opps at the other table are able to do.

Or maybe their system just prepares them for competitive situations better/handles competitive situations better, either by having MORE agreements, or BETTER agreements in contested auctions.

We do not know which of these things are happening but it is certainly not the case that we should stop overcalling against meckwell because they do better compared to the other table when they have a contested auction rather than an uncontested auction. Before I offend ulf again, let me point out that I'm not saying he was implying this, however some people might have taken the presentation of this data that way.

[Trinidad]

han, on 2012-August-10, 05:23, said:

You did well with the phrase "seems", but you did poorly when you wrote "significantly".

I estimate that -0.57 average IMPs on 51 jump overcall hands is less than half a standard deviation. So it does not seem significant to me.

Let's cut the statistics (or worse: guessing statistics) and get back to something that bridge players can understand.

We let Meckwell play 2 matches. In the first match, they play the 51 boards with a weak jump overcall. In the second match, they play 51 "overall boards".
In the weak jump overcall match, they lose 0.57 IMPs per board. That means that they will lose this match by 29.07 IMPs. On a 51 board VP scale (=32 board scale, multiplied by sqrt(51/32) ) this converts to a 19-11 loss in VPs.
In the "overall boards" match, they win 0.39 IMPs per board = 19.89 IMPs. On the same VP scale this converts to a 17-13 win.

Now bridge players can decide for themselves whether they consider the difference between winning a match 17-13 or losing it 11-19 "significant", without a further need for statisticians.

Rik

[awm]

One thing that this might refute is the widely-held opinion that "strong club pairs have trouble when opponents bid over the 1♣ opening" or that a strong ♣ system is "great when you open a limited 1M or an unobstructed 1♣, but lousy when you open 1♣ and opponents bid." At least in the case of Meckwell, their results when opponents bid over 1♣ are quite good. This holds true even if we normalize for the fact that Meckwells' results are always quite good -- they are doing better on the "strong club with overcall" boards than they do on average by a fairly significant margin.

It's true that you can't directly compare their results from "strong club with overcall" to "strong club without overcall" and conclude that you shouldn't overcall (they are different sorts of hands)... but at the same time, presumably opponents at the other table were not playing strong club on most of those "strong club with overcall" hands and Meckwell still came out ahead (and by an even bigger margin than they come out ahead in general on random hands). I think it would be safe to say that this is not the weak point of their system!