Inside the World of Big-Money Bridge (2013) | Hacker News Reader

Inside the World of Big-Money Bridge (2013)

observer.com

23 pointsluu12 comments

> While bridge is infinitely analytical, poker is more psychological: In high-level matches, every player at the table can compute the odds instantaneously, and what separates the best players from the pack is the ability to pick up “tells,” such as the furrowing of the brow as an indication of bluffing.

This is largely incorrect, and I wish this stereotype of poker (esp. high-stakes poker) would not be propagated so often.

Pros playing against fish in a private cash game may get a huge advantage off of physical tells, but that’s not the most consistent information that almost all players give away.

The consistent information that players give away is how they play specific kinds of hands in terms of position, frequency, aggression, and/or bet sizing. This extends even to very high levels of play. This type of information cannot be hidden, it can only be varied.

A classic low-stakes live tell is that players bet small on weak made hands and draws, but they bet bigger with their stronger made hands. Once a good player gets a sense of what those ranges are and what type of pressure (if any) can make the weak player fold, it just becomes a field day for the skilled player.

Highly-skilled players will have gears that they shift through such that their opponents have a tough time defining the skilled player’s hand range. Once the skilled player shows down some unexpected hands and gets into the weak player’s head, the game again becomes extremely easy for the skilled player.

Contrary to what the author says, I’m not actually convinced that the gaming skill set of high-level poker players and high-level bridge players is actually that much different.

If any high-level bridge players (I am not one) want to chime in and compare and contrast high-level strategies and tactics in these two games, I would love to extend the discussion.

Your description sounds very psychological, which is exactly what the article is saying as well? In other words, you're not concerned with the specific cards (analytical), but what your opponent thinks about their cards.

I would characterize high-level poker playing as very analytical, much in the same way that bridge (or spades in my case) is analytical.

I would say that the psychological game — especially getting into someone’s head — exists in bridge, spades, poker, and even games like chess and go (high stakes go games between a pro and a rich and skilled amateur often end with the amateur being brain-fucked and playing far below their actual skill level).

I will use spades as an example moving forward due to the fact that I have much more experience with it than bridge, and the gameplay is similar (as well as simpler).

A skilled poker player is always putting opponents on a range of hands, and the betting becomes largely very mechanical once the skilled player is dialed into a weaker player’s range.

A skilled spades player is going to make a guess based on bids and cards played about how the bidder plans to make their bid — specifically, with what cards — and then discard accordingly to attempt to create the highest probability of setting the opponent. The same can be said of winning the bid and potentially choosing to play a specific gambit if the hand doesn’t seem to be a sure thing.

The cards always matter to some extent in both games. That said, the analysis and mind games can be used in both as well.

Does that answer your question?

[edit: The author uses the term “psychological” to refer to physical tells. Online poker is played with no physical presence, so this form of psychological advantage does not exist. What about bet sizing info? They are called betting tells in the poker world. Is that psychological or analytical? What about learning about hand reading and playing tendencies?

My point is simply that I think that the skill sets of high-level players in each game are actually more similar than is widely believed and represented. Whether these similarities are called psychological or analytical is really just semantics in this particular context.]

I've played a lot of high level bridge, and a fair amount of mid-level poker.

You are definitely correct that there are a number of transferable skills between the two games. Interestingly, a lot of these concepts are much more elaborated in one game or the other, such that playing one game will make you a better player at the other. In fact, I actually picked up poker to improve my bridge skills.

For example, counting an entire deck is a critical and basic skill on bridge, and gave me a significant head start in poker. On the other hand, the psychological game is definitely applicable to both games, but is indeed much more developed in poker than in bridge. For example, in poker the concept of 'tilt' (being in a state where emotions are clouding judgment) is quite well developed. There are treatises on how to recognize that you're on tilt, how to get off tilt, how to deliberately play to put your opponents on tilt, etc. There is nothing written about this in the bridge literature, but at high levels it's hugely applicable.

As a teen I worked as a caddy at duplicate bridge tournaments (moving the cards from one table to another, collecting score cards &c.). There were a lot of interesting people involved there.

I actually worked a qualifier for the Women's National Yeam once. At that level they placed a divider diagonally across the table to prevent partners from being able to see and signal each other.

Personally I think it would be very hard to get to that level by cheating, as you end up with anomalous results (not only were they recorded, but pretty much any competitive player at the regional tournaments could rattle off every single hand and result after an afternoon of 26 hands of bridge).

It was good work, paid really well for a 13-15 year old, and I even learned enough bridge to play on the rare occasion I'm with 3 other like minded people.

side note: for those not familiar with duplicate bridge, you play the same hands as other players. The hands are scored similarly to rubber bridge, but your actual points for ranking in the tournament are based off of how much better or worse you do than other people.

In matchpoint play, many people play the same hands, rotating opponents and there is a formula for calculating the score.

In team play, 2 from your team play against 2 from the other team, and then you switch the hands so that you get the cards that the other half of your opponents team just had. Typically this is done in a sequence of rounds in a Swiss format[1]

1: https://en.wikipedia.org/wiki/Swiss-system_tournament

can you say any more about how cheating is prevented? I always wondered about that as the incentive is high and since teams enter as a team, it seems there is a lot of opportunity for planning communication. I guess I always assumed there was some kind of mixing of partners to prevent collusion (and judge the contribution of a particular individual rather than a team), but it doesn’t seem like that is the case.

During the auction, the players are screened from each other – no eye contact, and the screens extend to the ground.

https://en.wikipedia.org/wiki/Screen_(bridge)#/media/File:Br...

Also, verbal communication is banned. All bidding is communicated through the use of bidding boxes:

https://en.wikipedia.org/wiki/Bidding_box

which both provide an unambiguous record of the auction and remove the need for players to say anything.

Cheating prevention is a serious issue at the highest level of Bridge. The basic tenet is to have screens that prevent a player from being able to see their partner's face and torso and rules that seriously limit speaking. The screens have just enough space at the bottom to be able to see cards played and nothing else. Eliminating voice, body language, hand and feet action reduces ability to pass signals.

In the historic world of high-stakes rubber bridge, you're right - players would cut for partners after each 'rubber' game so your partner in crime might soon be your opponent. I'm sure there were occasional big-money scams like all situations involving gamblers and money generally. Roald Dahl wrote an amusing short story about one possibility.

In the modern world of competitive duplicate bridge, there are occasional collusion-based cheating scandals - the most recent one I'm aware of involved a communication system via position of the pencils with which the pair involved kept track of their score. By far the most famous scandal was a pair of English players back in the sixties communicating the number of hearts in their hand (enough to give a winning edge) with hand signals.

Nowadays good records are kept and cheating tends to come to light eventually. The apparatus used in top tournaments is different from a casual game and meant to prevent collusion, and also, with the diagonal screen, make it harder to tell whether any hesitation came from your partner (reducing deliberate or unintentional inferences).

In big tournaments with lots of players, because hands are duplicated to be played lots of times, it's not uncommon for players to be on a 'wire' i.e. deliberately or by overhearing learn some features of a hand they haven't played yet. It causes a lot of outrage and controversy but it doesn't ultimately matter that much. The most important matches are between two teams and things are arranged to reduce the opportunity for the pairs in a team to communicate with one another.

Since I'm writing a pretty long post here, I'll also mention that computers still don't play bridge especially well, so the problem of players getting computer assistance, such a concern in chess, doesn't feature.

All in all, preventing cheating in bridge is kind of an ongoing but not particularly high-profile activity. There's an obvious solution to make cheating even harder - keep players physically separate and make them do everything from a monitored computer terminal, which could also do things like introduce delays to make everything happen in tempo. But the administrators obviously think that would be pretty unpopular and prefer to keep it a card game.

In most formats partners play as a team of two (or as part of a team of four or six) and are not mixed.

Cheating at bridge has some interesting information theory elements. Communication between partners is explicitly allowed during the bidding phase of bridge via bids (e.g. a bid of "2 clubs" may imply a lot about the cards the bidder holds) but must follow a convention which is known to your opponents (so they learn just as much from that bid as your partner does). Communication during the playing phase of bridge is explicitly forbidden, e.g. you are not supposed to take a long time making a decision about which card to play, from which your partner could infer which cards you are probably choosing between. Most cheating schemes occur between the two phases, after the bidding but before the play starts where the defenders try to tell each other some critical feature(s) of their hands (e.g. that they hold the queen of trumps). Because a tiny amount of data can make a difference (a single bit could communicate the above fact) cheating strategies come in many forms. Making a noise (e.g. coughing or tapping) were classics, but can be overheard and may be obvious in review. The position of a body part, the choice of which hand to touch something with, the angle that a pencil or the bidding tray are placed at can all communicate information.

If you want to see some analysis of a relatively recent cheating scandal (and some of the anti-cheat devices that are used at top tournaments) YouTube has a few videos on the Israeli pair Fisher & Schwartz: https://www.youtube.com/watch?v=831tJ4EHLBY

It is prevented through a number of mechanisms

Bids are no longer spoken, but signaled through cards

At higher levels there is a screen across the table (as I mentioned in my original post).

Players are not allowed to vary the tempo of their bids (to allow for longer thought in certain situations, some bids are defined as "surprising" and the next player is required to pause longer).

In terms of detection:

Records are kept both officially and by other players looking to improve their own game. Players are required to define their bidding system ahead of time, so any pattern of "psychic" bids will be noticed very quickly.

Defense play is considerably harder, but a single win after making the "wrong" (i.e. not playing the odds) decision will be memorable for the opponents, so in any given circle of players you can only do it so many times (particularly when it's just enough to win the match) before suspicions get raised.

In several years of my working at sectional, regional, and national tournaments there were two times total that players were caught cheating, and in both cases it was at smaller games (one was at a regional tournament, but not at a particularly competitive level; the partnership at that regional tournament weren't just cheats though, they were really bad cheats and the motivation is still unclear since they were nowhere near professional level).

Fun fact: actor Omar Sharif was a high-stakes Contract Bridge player and organized many international competitions[0].

[0] https://en.wikipedia.org/wiki/Omar_Sharif#Contract_bridge_ca...

Reply on news.ycombinator.com