Join Bridge Winners
All comments by Nicolas Hammond
You are ignoring the author of this comment. Click to temporarily show the comment.
Here's the truth: real bridge players say they don't care about MPs, but they do.

When the Pairs formula was changed to increase the Pairs MP reward by 25% about 4 years ago, the results was that many players migrated to playing Pairs and not Teams.

There are now lots of attempted solutions to ‘fix’ the dearth of teams in the KOs.
Jan. 17
You are ignoring the author of this comment. Click to temporarily show the comment.
In Gatlinburg they will check and throw you out if you mis-report. The unofficial guideline is that your estimated points are within 5% to 10% of your actual points depending on total points, and also if you are in the wrong bracket. Those with less than 300-500 MPs per team typically don't need to worry.
Jan. 15
You are ignoring the author of this comment. Click to temporarily show the comment.
There is also a group of players, mostly on the cusp of bracket 1/2 that do not want to play in Bracket 1. I've probably worked on more KO starts than any non TD. Those players will hang out by the TD booth and will constantly ask if they are going to be in B1 or not. If they will be in B1 then they go buy a pairs entry.

Playing up/playing down is a policy decision. Get the ACBL to modify, or to allow the TDs to accommodate reasonable requests.

Junior national squad members should be allowed to play up.

Allow some teams to play up iff they will not affect the bracketing, i.e. if each of the affected brackets already has less than 16 teams then this does not really affect the bracketing.

I personally don't care what the policy is. Whatever the policy is, it is something that can be easily implemented in software. Therefore the “it's difficult to implement” argument should not apply.
Jan. 15
You are ignoring the author of this comment. Click to temporarily show the comment.
@Randy: Software can handle the details, but the decision on what to do it is currently at discretion of ACBL TDs. Bridgescore+ is certified by ACBL so can be used by TDs.
Jan. 14
You are ignoring the author of this comment. Click to temporarily show the comment.
@Randy: Bridgescore+ (BS+) has done Gatlinburg regionals for the last 5 years. It can handle teams that ask to play up (or down). It easily handles 300+ teams with a myriad of stationary table requests. Assignments are typically up in under 2 minutes from last sale/MP correction.

What is at issue is a policy decision. It is trivial for software the implement the policy.

@Ray: Timing for Gatlinburg is under 2 minutes while 1200+ people are waiting around. Without Bridgescore+ is used to take 23-27 minutes. BS+ has started 100s of KOs without any problems; there is nothing for Mark to prove. Already been done.
Jan. 14
You are ignoring the author of this comment. Click to temporarily show the comment.
@Art. In general (there are exceptions) a doubled contract is more likely to go down than an undoubled contract. Therefore you should double most contracts because statistically they are then more likely to go down. See the argument for bidding 4NT over 3NT above.

1X is more likely to make than 1 (at IMPs). But 1XX is even more likely to make than 1X.

1 is likely to make, 1X is likely to go down, but 1XX is very likely to make (80%) so… if you are in 1X you should redouble. (See argument for bidding 4NT above).

I haven't found any 4XX or 4XX making. Am sure there are out there just not in my IMP database.
Jan. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
Here are the only two ACBL 5NTxx I could find:

http://live.acbl.org/event/1701020/0601/1/board-detail/G?board_num=16
http://live.acbl.org/event/1607004/0202/2/board-detail/H?board_num=17

One of the six ways of scoring 1240.
Jan. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
To clarify, a simple contract is <level><suit>. It includes undoubled, doubled, redoubled.
At the top level, 5NT is the rarest simple contract. It occurs 0.03% of the time, i.e. given 10,000 hands it occurs 3 times.

There are some logical auctions to get to 5NT. 1NT-4NT (invite to slam)-5NT (return the invite)-Pass. I know there are other possible meanings for 5NT in that auction. If you look at the expert players that end up in 5NT, they don't get there using simple methods. Not as good a story to tell at the bar.

3NT is the most common (19%). Question: what's the second and third and why?

The most likely to make is 4NT (70%). As someone once pointed out, if you are not sure if you can make 3NT, bid 4NT because it has a higher success rate.

The least likely to make? Rather than tell you, I'll let you guess. And then tell me why. It's not 1.

The above is data from top level IMP games.

If you look at data from ACBL events, it is different. But I don't trust the data. Too often you will see some smart Alec playing 3NT and making 5 and put 5NT= in the Bridgemate trying to be cute. According to the ACBL data, each 7 level contract is less likely to be played than 5NT. I don't believe that. 3NT is still the most likely contract to be played (just under 16%). Adding the smart alecs that enter 4NT or 5NT when they really played 3NT puts 3NT at almost exactly 16%.

To be clear, the original data is from top level events, nearly all IMP games. The ACBL data is from regionals/sectional and includes all levels - 299er, Gold Rush, Open etc. The ACBL data is all MP games.

The most likely contract to make in ACBL MP games is 1 followed by 1 then 4NT.

Doubling contracts is not done anywhere near enough. Only 65% of contracts make (IMPs), 66% for MPs. More than 1 in 3 contracts will go down.
Jan. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
The formula for the maximum number of legal auctions to reach a non-doubled contract is F(n)=(21 * sum(f(n-1) … f(1)) + 4 where n is 1..35 for the contracts 1, 1, 1, 1, 1NT, 2 etc.and F(1) is 4.
1 1 4
1 2 84
1 3 1764
1 4 37044
1NT 5 777924
2 6 16336404
2 7 343064484
2 8 7204354164
2 9 1.51291E+11
2NT 10 3.17712E+12
For a doubled contract, D(n) = 2 * F(n). For a redoubled, R(n) = 2 * D(n).
The numbers quickly become quite large.

Edit: Use word auctions not bids per comment below.
Edit: Added word legal
Edit: Corrected formula
Jan. 10
Nicolas Hammond edited this comment Jan. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
@Ken: From the data I have, 5NT is only played .03% of the time. 1 is the next then 1. You are 3 times more likely to play 7 than 1.

Here are two 5NTx contracts.

http://www.bridgebase.com/tools/handviewer.html?bbo=y&linurl=https://www.bridgebase.com/tools/vugraph_linfetch.php?id=10611
Board 5

http://www.bridgebase.com/tools/handviewer.html?bbo=y&linurl=https://www.bridgebase.com/tools/vugraph_linfetch.php?id=11016
Board 27.
Jan. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
5NT is the rarest simple contract in top level bridge. X and XX would be even rarer.
Jan. 9
You are ignoring the author of this comment. Click to temporarily show the comment.
Deliberations yes; data used for it not necessarily so.
Jan. 8
You are ignoring the author of this comment. Click to temporarily show the comment.
Can you use the BBO handviewer URL?
Jan. 6
You are ignoring the author of this comment. Click to temporarily show the comment.
Ask Georgia…if she denies there was anything, let me know…
Jan. 5
You are ignoring the author of this comment. Click to temporarily show the comment.
Georgia Heth rejected the available information.
Jan. 4
You are ignoring the author of this comment. Click to temporarily show the comment.
@Jonathan: Anyone can edit Wikipedia. If you have data you can add, including references to published sources, please add.
Dec. 31, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
@Ed: “Of course, we do know that the sexes differ :)”
“We” don't. I do. You may suspect, but I have the real world data to back up my statements.
I have data on the women's events and open events and can compare the two. The results are … interesting. At some point, if/when I have time, I might publish.
Dec. 31, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Couple of points from the report: https://bridgefeed.acbl.org/lanzarotti-readmitted-to-acbl

“Lanzarotti was convicted of a one-time violation, not a pattern of behavior, and there was no evidence AT THE TIME of a pattern of behavior.” The capitalization is my emphasis. A very, very carefully worded sentence. Georgia Heth is aware of more evidence since then but chose to block it.

" stated that he had kibitzed thousands of hands played by Lanzarotti and Buratti prior to 2004, and more after that with the most recent having been the 2005 Cavendish Pairs won by Lanzarotti and Buratti.“

Assuming ”thousands" means >= 2000 this means he kibitzed 2,000+ hands from 1996-2003 and even more (2000++) from 2003-2005. 2005 Cavendish Pairs was held in May 2015. So Coker kibitzed 2,000++ hands by L/B in a 2 year period. I find this very hard to believe. At NABCs, a top pair plays < 50/hands a day.
Dec. 24, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Just some 7NT facts - nothing to do with OP.
There are ~3.7906E+45 (~3.8 x 10^45) ways of bidding to 7NT. It is unlikely anyone has system notes to cover all of those auctions.
The number of 7NT contracts that are easily found through public data is in the hundreds, not thousands, making this data set a little small for analysis. Following data is from high level team games, not pair games.
About 5% of the time when you are on opening lead against 7NT you will have an ace.
About 10% of the time when you are on lead against 7NT with an ace you will forget to double.
One example:
http://www.bridgebase.com/tools/handviewer.html?bbo=y&linurl=https://www.bridgebase.com/tools/vugraph_linfetch.php?id=52986 (board 14).
About 5% of BBO files have errors in them.
About 3% of the time against 7NT your partner has an ace.
About 40% of the time against 7NT when your partner has an ace, they forget to double.
If they don't double, you have approximately a 1 in 3 chance of leading their suit.
If they do double, you have approximately a 1 in 2 chance of leading their suit.
About 1 in 3 times against 7NT you will have a king. For those times:
If you have AK, you will lead A (or K).
If you have KQ you will lead K (or Q) (1 exception).
About 1 in 7 times you will lead away from your unsupported king.
Dec. 24, 2018
.

Bottom Home Top