[Scarabin]

In researching past work on simulating bridge play I have trawled through many papers by academics, mostly theses, and was surprised to fond that many feel that bridge play is inadequately researched and analysed. Many lamented a tendency to rely on examples and skimp on formulating general rules. One writer, and remember these are highly intelligent people, took the view that bridge is under developed by comparison with chess, is of recent origin and hence relatively unstudied. Because of its infancy there has very little analysis of play, and hence its development is stagnant relative to the progress on bidding.

Now its probably true that since the 1930's there has been greater development in bidding than in play but I think this could merely mean that play was already analysed under whist, plafond and auction bridge?

I don't expect many posters can recall the 1930's but in your experience is it true that development in play hangs behind bidding, or is bidding development merely catching up?

[chasetb]

We had a discussion awhile back (See this thread). I don't know whether it was in that thread or not, but Justin asked Bob Hamman what the biggest differences between the top players in the 60s and today. Apparently, Bob said it was the bidding. If the famous Blue Team in their prime were given a month or two to learn a modern system and treatments, he thought they would again be among the best, because their card play was that good.

Based on Bob Hamman and that statement, I think I can safely say bidding development is still catching up.

[Free]

Depends a lot what you consider theory. The play of the hand has been around for almost a century (and even longer if you consider other games with 52 cards as well) and the theory remains unchanged, while bidding theory evolves rapidly. Therefore I'd say play theory is way ahead of bidding theory.

The success rate of possible lines of play of a hand can relatively easily be calculated. Those odds haven't changed. Bidding theory is much more complicated, it changes all the time, and has made a lot of progress because of that. But percentage-wise I'd say play theory may be in the high 90s while bidding theory is completely uncertain. For all we know, bidding theory may still be in its infancy and we're on the wrong track without knowing it. (Perhaps) system regulations stand in our way to develop the optimal bidding system. Who knows...

Edit: added brackets around 'perhaps'

[awm]

There seems to be quite a bit of room for improvement in opening lead and defensive carding theory. I feel like this has been less well-developed than bidding and play, despite some interesting recent work.

[32519]

Free, on 2014-February-16, 12:30, said:

Perhaps system regulations stand in our way to develop the optimal bidding system. Who knows...

Great thought!

[Lorne50]

You only have to look at the myriad of different contracts reached on the same cards by players in top competitions to see that there is a huge gap between the optimum contracts and what is reached in real life.

[whereagles]

Bridge is obviously under-studied. The academic work on it is scarce. Here are some of the milestones I know of:

1. "The Mathematical Theory of Bridge", by Emile Borel and Andre Cheron. A bridge statistics book from the 1940s. It's out of print in english, but you can order it in french.
2. Vernes article on The Law of Total Tricks. Foundation of many modern bidding ideas.
3. "I Fought the Law", by Mike Lawrence and Anders Wirgren. Criticism of law of total tricks abuses.
4. "Winning NT/suit contracts leads", by David Bird and Taf Anthias. Major modern work on opening leads. Validates some of the results of Borel & Cheron.
5. "Partnership Bidding at Bridge", by Andrew Robson and Oliver Segal. One of the very few systematic books on bidding. A masterpiece.
6. "Capitancy for Advancing Bridge Players", by Denis Klein. Leadership principles on bidding and defense debated.

[billw55]

I think bidding theory has much more content, both explored and unexplored, than declarer play does. Not sure this is the same as saying that play is more advanced.

[Scarabin]

Yossi Nygate, author of Python (bridge squeezes) and Aspen-Bassinet (rule based declarer play program) said in his paper on Aspen:

"To use these techniques one must have a good evaluation fubction; but no one has been able to find such a function for Bridge......the overall quality of card playing has not improved substantially since the invention of Bridge......I believe this stagnation is due to lack of research.If more time was spent in developing card playing theories both humans and computer programs would benefit."

Here's a link to Aspen (if I've got it right):

My link

Sorry I do not know why it does not work. Can anyone help, please?

Best to just Google aspen+nygate

This post has been edited by Scarabin: 2014-February-18, 19:55

[whereagles]

Evaulation funcion? As in, objective function?

If so, I'm not sure whether cardplay can be cast into multi-attribute problem form. It it's not like chess, which is purely inductive. Cardplay has inferences.. it's an inductive-deductive procedure.

[Zelandakh]

There was a time when people said something similar about chess, that you could not quantify such factors as additional space or the different value of tempo depending on the position. And yet it was successfully done.

In bridge you have to value the effect of having additional information against closing off some layouts where you will make. And also weight probabilities depending on a range of factors. There is nothing here that could not be done with enough investment and involvement of top players though, at least imo. Funnily enough, if bridge had its surge of popularity now rather than after the war this investment would almost certainly be forthcoming and possibly allow the game to leap forwards. But bridge is not only in decline in terms of numbers but is also played primarily by the one of the least marketable demographic groups. Add to that that the most important national body actively discourages any change or advancement and you have a very bad combination for attracting investment.

[akwoo]

Bridge is more random than chess (at least over a session or a day of play), so I think the natural human tendency of loss aversion has something to do with lack of advances.

A typical theoretical advance in chess might win you 1 game out of 30. It will never cost you a game.

A typical theoretical advance in bridge wins you 4 boards out of 100 but also costs you 3 boards out of 100. Over a week-long event the theoretical advance is pretty sure to gain, but when playing events on the order of 50 boards, the advance will cost you 3 events for every 4 it wins you. Most people seem to feel more pain over losing 3 events they would have won than they feel pleasure over winning 4 events they would have lost.

[Scarabin]

That's fascinating: I take this as saying top players do evaluate and monitor progressive card states but the process is not well documented. Agreed deriving information is the major factor but others would be creation & management of entries,stops, and ducking and unblocking. The factors writers tend to classify as basic technique but do not list and that players at my level tend not to recognize as necessary.

I think too many books on play are written as entertainments (good for sales) and too few as textbooks. Let me take Forquet's "Bridge with the Blue Team" as an example. As an entertaining puff for the Blue Team it's magnificent but as a textbook it stresses the "shock,golly" at the expense of straightforward explanations.

The first hand describes how Chiaradaia made 6 spades on the following:

---

Forquet, who is one of my favourite writers, waxes eloquent on how Chiaradia "played as if he could see through the backs of the cards" instead of just saying he assumed everything was favourable.

Ch unblocked the spades and created an extra entry to dummy. F describes this as a farsighted unblocking play instead of saying Ch needed another entry to dummy and the unblocking play could not lose and might provide this.

The whole effect is to suggest Ch exercised inimitable genius and not to instruct you how to play a hopeless hand.

The only genuine textbooks I can remember are Love's "Bridge squeezes complete", Mollo & Gardiner's "Card Play Technique" and, perhaps Culbertson's Blue Book.

This post has been edited by Scarabin: 2014-February-18, 22:52

[blackshoe]

whereagles, on 2014-February-17, 15:32, said:

Bridge is obviously under-studied. The academic work on it is scarce. Here are some of the milestones I know of:

1. "The Mathematical Theory of Bridge", by Emile Borel and Andre Cheron. A bridge statistics book from the 1940s. It's out of print in english, but you can order it in french.
2. Vernes article on The Law of Total Tricks. Foundation of many modern bidding ideas.
3. "I Fought the Law", by Mike Lawrence and Anders Wirgren. Criticism of law of total tricks abuses.
4. "Winning NT/suit contracts leads", by David Bird and Taf Anthias. Major modern work on opening leads. Validates some of the results of Borel & Cheron.
5. "Partnership Bidding at Bridge", by Andrew Robson and Oliver Segal. One of the very few systematic books on bidding. A masterpiece.
6. "Capitancy for Advancing Bridge Players", by Denis Klein. Leadership principles on bidding and defense debated.

An excellent list. I would be tempted to get the Borel, even in French, but my French is very, very rusty. I have the third and the last two. I should look for the Bird and Anthias.

I have two other books on leads that might fit in: Journalist Leads, by Rubens and Rosler (aka "the Journalist") and Reese's translation of Vinge's book, the title of which escapes me.

[blackshoe]

Scarabin, on 2014-February-17, 23:08, said:

Sorry I do not know why it does not work. Can anyone help, please?

Best to just Google aspen+nygate

Try this one.

[Scarabin]

blackshoe, on 2014-February-18, 23:23, said:

Try this one.

Thanks

[EricK]

Lorne50, on 2014-February-17, 14:56, said:

You only have to look at the myriad of different contracts reached on the same cards by players in top competitions to see that there is a huge gap between the optimum contracts and what is reached in real life.

But a lot of that is down to interference by the opposition. One of the advances in bridge bidding theory, since the early days of bridge at any rate, is that it is much harder to reach the optimum contract if the opponent's interfere, and interfering is much safer in practice than it might first appear.

[whereagles]

Scarabin, on 2014-February-18, 20:06, said:

The only genuine textbooks I can remember are Love's "Bridge squeezes complete", Mollo & Gardiner's "Card Play Technique" and, perhaps Culbertson's Blue Book.

There are very few textbooks in english, but in french you can find a couple. Bridge is an optional course at secondary school (as chess, teams sports, etc), so they do have textbook-like stuff. French bridge textbooks are good, pedagogically speaking, and have reasonable systematics. See e.g. works by Roudinesco, Lebel, Bessis or Cronier. However, they have a tendency to flee from borderline hands and I found a few systemic lapses (some due to "holes" in the system, others due to different theoretical trends). Textbooks go all the way, up to university level stuff.

In english I would add Willam Root's "How to play a bridge hand" to your list.

I am in the process of writing one myself, in my native language (portuguese). Problem is, work keeps getting in the way LOL

[whereagles]

blackshoe, on 2014-February-18, 23:20, said:

I would be tempted to get the Borel, even in French, but my French is very, very rusty. (...)
I have two other books on leads that might fit in: Journalist Leads, by Rubens and Rosler (aka "the Journalist") and Reese's translation of Vinge's book, the title of which escapes me.

Borel probably classifies as the only academically-oriented work on bridge (that I know of). The amount of effort put on it was gargantuan. Remember that was before pocket calculators, and numbers were put out with 6(!!) significant digits.

"Journalist Leads" is a great book, but it is somewhat hard to digest. Some sentences seem incomplete It's interesting to see that simulations of Bird/Anthias go completely in line with Rubens/Rosler's arguments in favor of Rusinow leads: A from Axx, K from AKx. Borel's book also corroborates that (1) ace leads are less harmful than suggested by standard lore, and (2) king/queen underleads more dangerous.

[nige1]

Free, on 2014-February-16, 12:30, said:

For all we know, bidding theory may still be in its infancy and we're on the wrong track without knowing it. (Perhaps) system regulations stand in our way to develop the optimal bidding system. Who knows...

There's little scope for further innovation in declarer-play theory. Unfortunately, system-regulation constrains advances in bidding and defensive theory (e.g. Strong pass, Encrypted signals). Different Bridge legislatures forbid different methods (e.g. EBU & Moscito) and handicap different conventions (e.g. ACBL & Multi). This frustrates a concerted effort by theorists in what would otherwise be interesting and challenging aspects of Bridge. There's no incentive to develop a more effective method that you can play only in a long-match, at world-championship level, conceding seating rights, with opponents consulting their written defences, at the table. Unclear, over-sophisticated, inconsistent, fragmented rules hasten the demise of Bridge.