Rating systems (again?)

[HeavyDluxe]

From what I've seen, Jack would massacre GIB in the bidding. Computer card play theory however has not improved much since GIB other than increase in computer strength.

 

GIB was the first of a new generation of computer bridge programs and at the time miles ahead of the rest. The Jack team realized that the biggest gain was still in the auction (although I'm sure Jack is also better in the card play than GIB, simply because it was based on it and some improvements have been made since then).

 

It's interesting you mention that... I wound up buying BridgeBaron a handful of months ago when I started to get serious about learning.

 

At least part of the motivation to pick that product was based on looking at the results of one of Richard Pavilcek's bidding polls (like this, for example).

[helene_t]

From what I've seen, Jack would massacre GIB in the bidding. Computer card play theory however has not improved much since GIB other than increase in computer strength.

The Jack team claims that they have improved the defensive card play substantially. I haven't played that much with version 4 but I agree with their observation that the defensive play was prone to improvement in earlier versions.

[hotShot]

How do you measure errors?

I recalculate the number of tricks that NS can make after each card played.

If the number after the play is smaller NS made an "error", if it's greater EW must have made one.

 

The blame always goes to the one who played the card, although because of signals, lead directing doubles and things like that, the real responsibility must be given to his partner.

[jtfanclub]

How do you measure errors?

I recalculate the number of tricks that NS can make after each card played.

If the number after the play is smaller NS made an "error", if it's greater EW must have made one.

 

The blame always goes to the one who played the card, although because of signals, lead directing doubles and things like that, the real responsibility must be given to his partner.

Ah, well then. Luckily, most signalling errors don't do any harm, so they won't factor into your calculation of errors.

 

Heck, I once played a game where we agreed on up-side down attitude on signals and discards and then partner forgot for the entire game and signalled normally (and took my signals as normal). It made a difference once in 12 boards.

[Gerben42]

And it's certainly a wrong judgement of Zia to give up his famous bet when gib was invented.

 

Although GIB is not world class, it was revolutionary and solved some problems in computer bridge that allows in principle world class computer programs, so in a way Zia was correct recognizing that.

[awm]

I'd make the point that there aren't really "great bidders." Bidding is so much a function of partnership and system. Take the best player in the world and force him or her to play a totally unfamiliar system and we'll see a lot of trouble in the bidding.

 

You can't really rate an individual's bidding acumen by looking at hands. If you put two "good bidders" together but they're accustomed to different agreements and style, they're likely to have a lot of problems and reach a lot of poor contracts. If you put two "mediocre bidders" together but they're a well-established partnership with lots of agreements (and quite possibly their notes in front of them if playing online), they will do quite well.

 

If you want to rate bidding, I think you have to do it by partnership.

[helene_t]

I don't think so. Take two World Class players with very different systems and style, say Sabine Auken and Huub Bertens, and they will just play some middle-of-the-road expert style which they both know reasonably well. Of course partnership harmony will not be the same as with their regular pd's, but it will still be quite good. I've seen some casual partnerships doing well in top competition. Last year one of the bigger Dutch tourneys was won by Bauke Muller and Jan Jansma. Marion MIchielsen has played successfully with just about everybody in the Dutch sub-top.

[nige1]

How do you measure errors?

I recalculate the number of tricks that NS can make after each card played. If the number after the play is smaller NS made an "error", if it's greater EW must have made one.

Thank you hotshot.

 

I am amazed that GIB and "World-class" players average less that one "double-dummy" play-error in every two boards.

 

I do accept your figures but I'm surprised. I would have thought that, for example, an expert declarer could often finesse the wrong way; or finesse when the drop would have worked. At teams, he would also play to make his contracts when a successful risky play for overtricks was available.

 

I wold expect defenders to make even more "mistakes", especially on the opening lead.

 

More questions...

• Does the average include when a player is dummy?
  
• How do defender and declarer "errors" compare?
  
• In particular, what is the range and variance of such "errors". For example, do you find many boards with more than 5 errors
  

[hotShot]

Board where a player is dummy are eliminated.

Well the software keep extra records for errors in declarer play, defense and leads, but I did not investigate much. I don't have enough recorded boards for a full/serious statistic analysis. If I have 400 recorded deals with a player, he's dummy in 100, so they are lost. He's declarer in about 100 and defender in about 200, which is not much for serious analysis.

[nige1]

Board where a player is dummy are eliminated.

Well the software keep extra records for errors in declarer play, defense and leads, but I did not investigate much. I don't have enough recorded boards for a full/serious statistic analysis. If I have 400 recorded deals with a player, he's dummy in 100, so they are lost. He's declarer in about 100 and defender in about 200, which is not much for serious analysis.

Oh Well. Thanks anyway. You are lucky with your pick-up partners hotshot :) Mine often make three or four trick costing errors on a single board :D Fortunately opponents often chuck the tricks right back :)