BBO Discussion Forums: Has statistical analysis/data science arrived in Bridge performance yet - BBO Discussion Forums

Jump to content

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

Has statistical analysis/data science arrived in Bridge performance yet

#1 User is offline   thepossum 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,362
  • Joined: 2018-July-04
  • Gender:Male
  • Location:Australia

Posted 2019-March-14, 23:15

Dear all

Do top players and teams use statistical analysis/data scientific approaches of their own or opponents play at the top level. I'm just asking since I'm trying to analyse how I go in tournaments, trying to identify my weaknesses, areas mainly in need of improvement and was wondering about how much theory and application there is at the top level of bridge - as is happening in other sports now

At a simple level I suppose I'm looking at the changes in mean, variance on different types of hands, different types of tournaments, different sets of opponents etc. Mainly however, just analysing my own hands and trying to get rid of the worst x % of hands, given the skewed nature of the distributions. I dont have enough data on other players to analyse more broadly. I'm also considering analysing time of day, declare vs defense, seat, vulnerability, IMPs/MPs etc etc.

I would be very interested to know if people do it outside their own private and personal analysis :)

regards P

Note, its quite difficult searching for this stuff since there is a huge amount of information about analytics in relation to bridges, contracts, performance, bridging gaps etc. The few reference relating to contract bridge, analytics and performance is the performance of simulation models, not performance of the players :) Matthew Kidd seems to be one of the few people who have published any such analysis on the internet - in searches to date - at a broad level of hand analysis, rather than individual performance analysis
0

#2 User is offline   helene_t 

  • The Abbess
  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 17,068
  • Joined: 2004-April-22
  • Gender:Female
  • Location:UK

Posted 2019-March-15, 00:58

I don't think most top players do.

Emile Borel, the famous mathematician, wrote a book about probabilities in card games, and there has been some work by Pavlicek and the Bird/Anthias book. And lots of research on th LOTT. A few top players have mentioned statistical analysis of their own performance, for example to monitor gains/losses from extremely aggressive preempts.

The butler score from team events is sometimes used by the press to monitor performance of individual pairs and I wonder if captains make decisions based on it. If so, I would hope they adjust for the strength of the opposition.
The world would be such a happy place, if only everyone played Acol :) --- TramTicket
0

#3 User is offline   hrothgar 

  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 15,372
  • Joined: 2003-February-13
  • Gender:Male
  • Location:Natick, MA
  • Interests:Travel
    Cooking
    Brewing
    Hiking

Posted 2019-March-15, 04:21

Back in the day, I actually spent a bunch of time tracking the MP scores that my partner and I received based on

1. Opening bid
2. Seat
3. Vulnerability

While this analysis might seem pretty simple, it actual had a pretty profound impact on my thinking about a number of different aspect of the game.

In particular, while the data was noisy, it seemed pretty clear that

1. Our expected results varied dramatically by opening bid
2. Our overall results depended a lot on the relative frequency of those hands this particular night

(At the most basic level, the more strong club openings we were dealt, the worse our score)

At the time, our first/second seat strong 1!C opening occurred about 15% of the time
We were playing 27 boards a night, so we expected about 3.5 strong clubs an evening, with a standard deviation of about 3.5.

On evenings when the card gods didn't smile, that number could get a whole lot higher, and conversely our score would go lower.
Alderaan delenda est
1

#4 User is offline   hrothgar 

  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 15,372
  • Joined: 2003-February-13
  • Gender:Male
  • Location:Natick, MA
  • Interests:Travel
    Cooking
    Brewing
    Hiking

Posted 2019-March-15, 04:38

In my experience, the "powers that be" in bridge don't value expertise in this area, and consequently, there isn't much in the way of serious work being done:

None of the organizations is willing to invest in the type of comprehensive record keeping that is necessary to actually address issues like

  • Developing accurate player ratings
  • Analyzing whether people are cheating

It would be easy to do some work to improve upon tournament schedules and re-allocate boards to improve on the accuracy of the events, but no one is much interested in that.

Don't get me started on what happened when I attempted to convince people that we should generate confidence bound on the results of team matches and use this to tweak master point allocation...

And, of course, there was a whole lot of interesting work being done around bidding system design, up until the regulatory authorities decided that they needed to make the world safe for inferior methods and started banning strong pass openings, relay methods, anything other than conservative single suited preempts.
Alderaan delenda est
1

#5 User is offline   Hanoi5 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 4,080
  • Joined: 2006-August-31
  • Gender:Male
  • Location:Santiago, Chile
  • Interests:Bridge, Video Games, Languages, Travelling.

Posted 2019-March-15, 12:55

Unfortunately a lot of work is involved in analyzing bridge data. And it is even more difficult to find reliable data on the internet. Can you trust vugraph archives 100%? I don't think so. I had some archives from Hamman and he appeared as B Hamman, R Hamman, Bob Hamman, Robert Hamman and a couple more I don't remember and that's just the name, how can one trust the spots played to each trick?

View Postwyman, on 2012-May-04, 09:48, said:

Also, he rates to not have a heart void when he leads the 3.


View Postrbforster, on 2012-May-20, 21:04, said:

Besides playing for fun, most people also like to play bridge to win


My YouTube Channel
0

#6 User is offline   PrecisionL 

  • PipPipPipPipPip
  • Group: Full Members
  • Posts: 941
  • Joined: 2004-March-25
  • Gender:Male
  • Location:Knoxville, TN, USA
  • Interests:Diamond LM (6700+ MP)
    God
    Family
    Counseling
    Bridge

Posted 2019-March-15, 13:09

I have used Match Point Analyzer from Bridge Buff 19 to evaluate different opening bids MP results history for several partnerships. I find this useful since I play several Strong Club Systems and we frequently change our opening 2-bids based on this analysis.
Ultra Relay: see Daniel's web page: https://bridgewithda...19/07/Ultra.pdf
C3: Copious Canape Club is still my favorite system. (Ultra upgraded, PM for notes)

Santa Fe Precision published 8/19. TOP3 published 11/20. Magic experiment (Science Modernized) with Lenzo. 2020: Jan Eric Larsson's Cottontail . 2020. BFUN (Bridge For the UNbalanced) 2021: Weiss Simplified (Canape & Relay). 2022: Canary Modernized, 2023-4: KOK Canape.
0

#7 User is offline   thepossum 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,362
  • Joined: 2018-July-04
  • Gender:Male
  • Location:Australia

Posted 2019-March-15, 17:22

Thanks everyone

Very interesting thoughts

regards P
0

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users