How frequently do different auctions occur

[hrothgar]

Barmar and Uday were nice enough to share some bidding information from BBO with me for a project that I was playing around with. I thought that some of the results might be of general interest.

 

I did a quick and dirty analysis looking at how common various auction might be.

 

Barry shared a data base show an extremely large number of auctions that were by someone (the data was anonymized, but all of the auctions involved a single BBO player)

 

I filtered the data to only look at auctions that had finished, and then calculated how many times each auction cropped up. (Note: the auction All pass did not show up in the data base)

 

Here are the top 10 auctions (As you can tell different variants of 1N - All Pass is relatively popular)

 

Auction Percentage Cumulative %

1Nppp 0.640% 0.640%

p1Nppp 0.633% 1.272%

pp1Nppp 0.564% 1.836%

1Np3Nppp 0.420% 2.256%

p1Np3Nppp 0.316% 2.572%

ppp1Nppp 0.258% 2.830%

pp1Np3Nppp 0.230% 3.060%

1Np2Hp2Sppp 0.179% 3.239%

2Sp4Sppp 0.179% 3.418%

1Sp2Sppp 0.172% 3.590%

 

Few quick points:

 

The top 500 or so auction comprise 25% of all observed bidding sequences made by this individual.

 

The top 3.3K auctions, comprise 50% of all sequences.

 

After this point, you're looking at sequences that only occur once in a blue moon (by which I mean out of 15K unique auctions and 30K total auctions, the individual sequences only occurred once)

[Tramticket]

He/she plays an awful lot of bridge! Do we know the system used? I would expect, for example, the sequence 1NT-P-P-P to occur more frequently in a weak NT context.

[hrothgar]

He/she plays an awful lot of bridge! Do we know the system used? I would expect, for example, the sequence 1NT-P-P-P to occur more frequently in a weak NT context.

 

While I don't have any direct knowledge who this individual might be, I do know that they play a lot.

Sadly, I don't have any information about what system they were playing.

 

Please note, I was looking at auctions that ended with pp, followed by a third p (there were more in the file, but I did this to avoid double / triple counting results)

 

If folks are interested, I might be able to get a much broader swatch of hands and look at something in the aggregate

[FelicityR]

He/she plays an awful lot of bridge! Do we know the system used? I would expect, for example, the sequence 1NT-P-P-P to occur more frequently in a weak NT context.

 

And if a weak NT were used, the opponents would surely have more opportunities to overcall. The data is useful but as with all bridge statistics you need to know the level of the players at the table and their methods. Less experienced players tend to have fewer conventions to call on including conventions to overcall 1NT whether weak or strong.

 

To get a realistic feel of whether the statistics hold up, you probably need to analyse the actions and auctions of a few hundred similarly-experienced players and see whether they are comparable.

[hrothgar]

I had originally hoped to get all the records for JEC because

 

1. Single source

2. Talented individual

3. Single system used by Cayne

 

However Barry (rightfully) had concerns about anonymity...

 

If anyone knows JEC and can get permission, I'd love to look at this set of hands.

[tomkron]

It would interest me, at least, to get statistics about auctions from a large number of hands by many players all using the same system and same type of scoring. Frequency of various contracts, longest auction, some metric of similarity of different auctions if anyone could think of how to define and measure such a thing (i.e, 1H-4H and 1S-4S are different, but very similar in some sense), etc.

 

That might be possible here since the robot individuals are all using the GiB 2-over-1 system. A bunch of matchpoint hands could be looked at, and/or a bunch of imp hands. Of course 3/4 of the bids would be GiB rather than Human, but the analysis would be interesting to me anyway.

[SelfGovern]

> Please note, I was looking at auctions that ended with pp, followed by a third p

Forgive me for being dense, but doesn't every auction end with pass, pass, followed by a third pass?

[barmar]

> Please note, I was looking at auctions that ended with pp, followed by a third p

Forgive me for being dense, but doesn't every auction end with pass, pass, followed by a third pass?

More specifically, these are just auctions where the player in question made the last pass.

 

The data we sent Richard was all of this player's bids, along with the auction that led up to each bid. By filtering the data down to just the ones where the bid was Pass and the last 2 bids in the auction were also pass, he got the complete auctions.

[Vampyr]

And if a weak NT were used, the opponents would surely have more opportunities to overcall.

 

Possibly not. Yes, the opponents will, on average, have more HCP than when a strong NT is opened, but they need that extra strength to overcallver a weak NT. After all, over a strong NT you can come in on any old crap, but over a weak NT you can’t (well, you shouldn’t!).