BBO Discussion Forums: Strategies when outclassed - BBO Discussion Forums

Jump to content

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

Strategies when outclassed any good answers?

#1 User is offline   uday 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 5,808
  • Joined: 2003-January-15
  • Gender:Male
  • Location:USA

Posted 2006-September-15, 10:19

So you're a keen bridge player and enter a strong team event like the ACBL Vanderbilt with your usual crew. The seeding system in these events means you' quickly face a team that is far out of your league. This enemy team will somtimes contain one player who is considerably weaker than the others.

This isnt about seeding or sponsorship Start your own thread if you want to to grump about those things.

What I'm wondering is: what is the "correct" thing to do when faced with a team that rates to beat your own team 995 times out of 1000. Assuming the goal is to win 6+ out of 1000 (as opposed to being respected by peers or losing by a respectable margin), what should your team do?

Should it attempt to define and use a very high variance system/strategy so that the outcome is more based on chance than the differential in skill levels ?

Should it attempt to emulate meckwell and bid up every hand?

Should it attempt to play "good bridge" to get better for future events, treating this event as a learning experience?

Should it attempt to play a style that deliberately creates nothing but "no agreement" sequences (if such a style is legal and possible)

Should it attempt to explicitly cater to the presence of the weak player (if there are legal ways to "aim" at one player on the opposing team)

Something else?

#2 User is offline   Gerben42 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 5,577
  • Joined: 2005-March-01
  • Gender:Male
  • Location:Erlangen, Germany
  • Interests:Astronomy, Mathematics
    Nuclear power

Posted 2006-September-15, 11:09

First if there is a sponsor you have to hope he gets to make all the decisions.

Second I would just try to do as well as possible and get lucky :)

What you should NOT do is try to put in many new agressive conventions that you don't have much exp. with. They will probably know more about it than you...
Two wrongs don't make a right, but three lefts do!
My Bridge Systems Page

BC Kultcamp Rieneck

#3 User is offline   glen 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,637
  • Joined: 2003-May-11
  • Gender:Male
  • Location:Ottawa, Canada
  • Interests:Military history, WW II wargames

Posted 2006-September-15, 11:34

Generally you want to do some things that will create possible swings for your side but that are not low percentage operations.

There are six areas to operate in:

1) Many slam type hands are about 50% to make, give or take a bit. If you can get on the other side of the coin flip, then you will have a swing. One easy way to get a coin flip is to dump using RKCB against a good team, and switch to straight Blackwood! You may get to a slam off the trump king, but finesse for it and cheers!

2) You want to use a different opening notrump range than the opponents, to take advantage of the random effect this has. For example, if the opponents are playing 15-17, play 14-16 and vis-versa. Don’t spend a lot of time implementing this – just agree with partner that you will have 1 more point/1 less point for all actions on balanced hands.

3) You want to be a little conservative making game invites if playing against two strong players. Your opponents will defend extremely well so a lot of stuff will not make. At the other table, the players with your cards will push since they will assume your teammates will not always defend optimally. So on a day where the cards do not reward game bidding, one can pick up a string of swings.

4) Avoid revealing sequences, carding, and tells as much as possible – your sharp opponents will spot everything, even if they continue to look like they could care less. By “tells” I mean how you behave at the table – for example dummy comes down and you grimace for a second.

5) Look for opportunities to slightly distort in the bidding on not vulnerable hands. This does not mean psyche, but just a small deviation from the norm – perhaps having one card less in a suit than you should have or having one more point than you should have.

6) When in doubt about what to do, bid 3NT if a possible contract, and smile whatever dummy hits the table. If a couple of flaky 3NTs roll home, not only do you have the swings in the bank, but you have the opponents petrified they will lose the match due to “unbelievable” 3NTs.
'I hit my peak at seven' Taylor Swift

#4 User is offline   hrothgar 

  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 15,343
  • Joined: 2003-February-13
  • Gender:Male
  • Location:Natick, MA
  • Interests:Travel

Posted 2006-September-15, 11:45

Assuming that

1. You are thoroughly outclassed and
2. Your goal is to win as often as possible and
3. You are ethically constrained not to pull a Harding and kneecap members of the opposing team

I'd argue that your only real chance is to neutralize the superior skills of the opposing team by maximizing the variance of the results. Play sound bridge, but try to maximize the role that luck will play in the proceeding. Here's a classic example that I first saw discussed by Woolsey:

Assume that you are in 4 and you contract depends on picking up the following trump suit for no losers

AKJ87 opposite


Consider playing for the drop rather than the finesse. The finesse is technically the better play, however, the odds aren't that far apart. If you play for the drop, you'll dramatically increase your variance while having a small impact on your expected score.

These types of arguments can easily be extended to bidding. If the opponents are playing 5 card majors and a strong NT, you should be playing 4 card majors and a weak NT to minimize the chance that you'll play the same contract from the same direction with the same inferences available.

Personally, I think that its a mistake to change your methods in the middle of a tournament. Even subtle shifts to your NT range could rebound nastily. I suspect that the losses from bidding misunderstands would far outweigh the gains from increased variance. However, the Vanderbilt is still a couple monthes out. This might give you enough time to learn a new bidding system that seems theorectically sound but very unlikely to be played by any of the teams that you might face... I'd recommend MOSCITO, however, you can't play it in the Vanderbilt.

You might want to consider Fantoni - Nunes if you could find a good write up.
Alderaan delenda est

#5 User is offline   awm 

  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 8,281
  • Joined: 2005-February-09
  • Gender:Male
  • Location:Zurich, Switzerland

Posted 2006-September-15, 11:46

One of the best ways to swing boards in this situation is to be conservative. Most players bid very aggressively at IMPs. They bid a lot of games that are less than 50-50 to make. Of course, in expectation this will work out for them, but there's definitely a chance that all three 40% games are going down in any particular match. If you bid really pushy games that the other team doesn't bid, you might get lucky, but more often making them will hinge on a misdefense or tough declarer play. If you don't bid games they do bid, then all you have to do is be a bit lucky.

If/when you can, it helps to manipulate the lineups. If there is one weak player on their team, seat your strongest player so he/she will hold the same hands as this weak player. If a high percentage of the tough decisions are made by the person in that seat, you maximize your advantage. If the cards lie in such a way that the strong team's weak player basically doesn't have to do anything but follow suit, your chances aren't good anyway.
Adam W. Meyerson
a.k.a. Appeal Without Merit

#6 User is online   mikeh 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 12,716
  • Joined: 2005-June-15
  • Gender:Male
  • Location:Canada
  • Interests:Bridge, golf, wine (red), cooking, reading eclectically but insatiably, travelling, making bad posts.

Posted 2006-September-15, 12:00

1. Play your own game.

2. If you and partner (or your teammates) are comfortable with minor changes, try to play a different 1N range than the opps: an ideal change would be to play 14-16 if your opposite numbers play 15-17. There are many examples of swing hands in high-level competition caused by such minor differences: you hold 16 hcp... partner holds a poor 9.. he may decide to pass opposite 14-16 while he cannot do so opposite 15-17.. the two pairs may end up 2 tricks apart, with both pairs bidding perfectly: this is a great randomizer that does not entail any anti-percentage action, and thus is as likely to win as to lose, unlike trying to create swings through judgment or play decisions

3. Take mildly, the emphasis on mildly, anti-percentage lines in some contracts. Years ago, in the finals of a NABC Mixed BAM, where my partner and I were playing with our spouses, where we were all arguably outgunned and certainly our spouses were in unfamiliar territory, I had to play K10x opposite AJ8xx for 5 winners. The percentage play is to cash the K and then run the 10. I cashed the A and then played low to the 10... there were no clues in the hand, and when my LHO peevishly asked why I had misplayed the hand (winning, else I am sure she would have silently gloated), my honest answer was that I was not interested in tying the board: we needed to win as many as we could.

4. Do not go nuts: the opps will have the edge on you in most aspects of the game. However, if you are entering the event, presumably you have sufficient technique so as to play and defend the hands reasonably well: if you lack that basic competence, no advice will help lack even a 1 in a 1000 chance. Where you are absolutely going to be outgunned, probably even by the sponsor, is in the area of high-level judgment. They will double you more effectively than your teammates will be doubling at the other table and, especially the non-sponsor pair, they will maximize the defence. So don't go nuts... stay within your comfort level in terms of preempting, sacrificing etc.

5. If you are faced with a truly close bid game or not decision, stay low. Your opps will be bidding everything that moves, and they play the hands better than you do. If you reach the same contracts as your opposite number on every board, you lose. You have to hope that some of the close games go down. You hope for a series of 4-1 or 5-0 trump breaks etc. While the odds of that happening are low, they are higher than the odds that you will outplay them as declarer or out-defend them.

6. Be quiet. Don't engage in banter with the opps, no matter how flattering it is to be treated as an equal by a star player. Don't apologize to partner, and never say anything when you lay down dummy or inspect partner's hand as dummy other than 'good luck' or 'thank you partner'. Don't recriminate or remonstrate. Keep a poker face and an even tempo. If you feel this is becoming difficult, go to the bathroom or get a glass of water.

7. Play every hand as if you are ahead by 1 imp, no matter what the score may feel like.... in terms of focus, bearing in mind the above suggestions for creating mild chances for swings.

8. If the sponsor is on play or defence, try to make the sponsor guess early.

9. Do not make close penalty doubles of one of the stars... he will play the hand too well for this to be wise, especially if warned by your double.

10. Pace yourself. Some star players are exceptionally quick.... and they use this as a tool. Most of us feel pressure to keep up with that tempo...don't! Insist on taking your time before playing to the first trick and thereafter. This can be very difficult... but once you establish that this is how you are playing, the opps have to respect this.... and they will respect you as well. Other star players are very slow.... don't lose focus. Use the time you are given to look deeper into the hand rather than sit there getting impatient and angry. Any emotion impairs your ability to concentrate.

11. Have a rule that, when comparing with teammates, one player reads his socre, a member of the other pair announces the imps, and NO-ONE mentions the contracts nor any expression of fault or credit. The time to post-mortem is after the match. I once won a major (for me, anyway) event after a 72 board final and, when asked to write an article about it, was unable to say anything about the results in the other room because I never heard what their results were: I only knew the imp differential each hand. It is tough enough to worry about one' s own game without worying about your teammates, and the less you know, the less will be the distraction... and the fewer the recriminations if you lost imps on a board.
'one of the great markers of the advance of human kindness is the howls you will hear from the Men of God' Johann Hari

#7 User is offline   Al_U_Card 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 6,080
  • Joined: 2005-May-16
  • Gender:Male

Posted 2006-September-15, 13:23

Have you never played in an event against a hopelessly outclassed opp who managed to squeek by you by the narrowest of margins????? Of course you have. Did you assign it to any one factor? If so, was it that they seemed outclassed so you decided to play recklessly (or cautiously)?

A little psychology goes a long way. My only experience was playing as a relative newcomer in a CNTC against a team of top-flight opps. We were nervous, respectful and somewhat embarrassed. I even apologized when I bid a grand in the middle of the second set and it came home on a nice break. I was assured by the opps that it would be bid at the other table.......but nonetheless the opps started to bid and play as if they were "worried".

Well we didn't finish last but we did win our match against the big guys because the grand was not bid at the other table.....they didnt think that they had to bid it to beat us!

The moral of the story is...on any given sunday.....give thanks for small favours...
The Grand Design, reflected in the face of's a fluke!

#8 User is offline   joshs 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,082
  • Joined: 2006-January-23

Posted 2006-September-15, 16:00

While I agree with most of the comments made before, let me comment about the places were IMPs can be swung your way, fairly randomly:

As has been mentioned before
1. be slightly more conservative than the opps on bidding game, when there are two pros defending at your table. There are many situations where there are 35-40% games and you gain 10 if you make and lose 6 if you go down. As you will go down more often than you will make, if there are 3 such deals in your match and the opps bid all 3, and you don't, here is the matrix:

Lets say they are 40% games:

All 3 make: 6.4% of The Time, Lose 30
2/3 Makes: 28.8%, Lose 14
1/3 Makes: 43.2%, Gain 2
0/3 makes: 21.6%, Gain 18

You will gain big 21.6% of the matches, and lose bid 35.2% of the matches. So if you thought your head to head chances were much worse than 40% this is a reasonable chance to take.

Note this is the losing strategy in the long run by IMP expectation (-1.2 imps/bd), but only by a little bit. And if the games were really 35% games it becomes winning strategy.

2. Be VERY agressive for part scores. One way expert teams pick apart their opponents is by winning lots of 3 and 5 imp gains in part scores. Further even in the dangerous situations where many experts will not bid (such as marginal balancing situations over 2S), it is often the case that bidding will gain 2-3 imps 3 times, while losing 10 the 4'th (IF they remember to x you). Obviously this is a long term losing strategy, but you will often have matches where you made this decision 4-5 times and never ran into the lose 10 layout (which only happens 1/4 times on the very agressive part score competative bids). So on a good day, you can be plus 10-15 from this strategy, and the good day occurs maybe once for every 3 matches.

3. Be a little less disciplined with your (prior to the opps bidding) pre-empts. This introduces a great deal of randomness in your result. For instance, white vs white consider opening 3C on: x Kxx xxx QT8xxx in first seat, which is an action that favorites usually don't take, because its highly swingy.

4. Consider bidding a 50% grand slam early in the match, if its an obvious slam hand and the other teams client isn't at the other table

5. Play a different 1N opening bid range.

1-5: I probably wouldn't over do it and take too many of these actions, after gaining 15-20 imps from these types of swings, you probably want to get more mainstream, since the odds of you winning have now gotten much better (assuming your team is not totally hopeless)

6. Bid more agressively to game/slam whenever the client is on opening lead.

7. Do not sacrifice very much when the client will play the hand. If the client opens 1N, don't overcall as agressively as normal, until the declarership of suit contracts is determined.
(E.G. IF the client opens 1N, With KQJTx xx AQxx Jx I would pass 1N but bid 2S later if responder x-fer into hearts. You want to let the client declare, rather than yourside declarering, whenever the decision is close. Similarly you want to client to defend rather than the pro declares, whenever the decision is close)

8. And if you haven't built a 20 imp lead when the client was in, its now time to prey.

#9 User is offline   PhantomSac 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,488
  • Joined: 2006-March-23
  • Gender:Male

Posted 2006-September-15, 18:12

Uday, the key is for there to be different decisions, inferences, and judgements available. A different NT range from your counterparts is perfect. This creates a lot of different situations and inferences which is exactly what you want. Different decisions on preempting or overcalling are also very good. Swings at the highest levels can be created from very small things like this. There is no need at all to overdo it.

I think not bidding close games is a huge mistake against anyone. This is the bread and butter of the top teams, and the situations arise so often in a match that they WILL beat you on these boards if you don't bid your close games. Rather, I think it is better to bid or defend these games after a slightly different auction where different inferences are available. Both declarers may then make the "right" play and one may go down while the other makes.

Slam swings can often swing the match. I do think trying to do the opposite of the other team on slams in the 45-55 % range is a good policy. There will probably be just 1 or 2 of these in a match, and if you are on the winning end of both of them you will become the favorites. Don't overrate how outclassed you are, the top players are human too and I happen to know you are quite a good player yourself.

As for the client factor, sit your most active pair against him. I always try and put a lot of pressure on the client in both the bidding and play to exploit his weakness. Definitely bid em up against the client, and make him use his judgement as much as posible.
The artist formerly known as jlall

#10 User is offline   hotShot 

  • Axxx Axx Axx Axx
  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,976
  • Joined: 2003-August-31
  • Gender:Male

Posted 2006-September-16, 07:50

One key to beating a better team is using the "wishfull thinking defensive method".

A strong team as you describe it, is used to bid slams and games that have a success rate of 45-50%. So don't look for the best defence, imagine a holding consitent with your cards and opps bidding that will make opps go down. Then play as if this was the actual holding until you know better. That way will often give away a trick but it will create a big swing in favor of your side if it works.

If you think you need more than that:
We all know that declarer play is a little easier than defence. If your side is good in declarer play, try to play more often. One way to do that, is using a weak NT (12-14) and 5-card weak two's (1st seat, 7-11 HCP, second suit possible) in alls suits. This is constructive bidding and the risk is not to high. You force opps to start their bidding at the 2 level or even above it. Bidding will often be: 2 (something) - all pass

In any sport it ethical to select a stategie against opps weakest spot. So if you know that a team has a weaker player, you can try to use that. But this is usually dangerous, because opss will expect that and you are temped to try that even if it's not neccessary or even unhelpfull.
E.g. you need to exit and you have the choice, exit to the weaker player, he might make an error more often.

#11 User is offline   Tola18 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 333
  • Joined: 2006-January-19
  • Location:Sweden
  • Interests:Cats.

Posted 2006-September-16, 09:16

hotShot, on Sep 16 2006, 08:50 AM, said:

E.g. you need to exit and you have the choice, exit to the weaker player, he might make an error more often.

A variation of this is the finessing. Say you have A10xx to KKn9x. You dont have any clue where the queen is, and it dont matter who gets in.
Finess into the weaker player!

Another variation is, there come out an unharmful lead. You do see one particular switch is deadly. But the leader apparently didnt see it, otherwise he would already lead this suit.
Finesse into him.


A strategy strong players usually have: If they understand they are leading, they play very conservative, letting you do your mistakes as much you want...

This they also do sometimes already from the beginning. They know they wont make stupid mistakes, but a weaker player do ALWAYS some...
Cats bring joy and a feeling of harmony and well-being into a home.
Many homeless cats seek a home.
Adopt one. Contact a cat shelter!
You too can be an everyday hero. :)

#12 User is offline   keylime 

  • PipPipPipPipPipPipPip
  • Group: FD TEAM
  • Posts: 2,735
  • Joined: 2003-February-13
  • Gender:Male
  • Location:Nashville, TN
  • Interests:Motorsports, cricket, disc golf, and of course - bridge. :-)

Posted 2006-September-16, 11:37

I do the following:

1. Overcalls and preempts are sound and disciplined; due to my adjustment period playing in A/X here I went to intermediate jump overcalls. Now it's a staple on my CC regardless of bridge level or area. I am VERY careful of my 2 suited overcalls.

2. Stay within yourself. Do not let the natural emotions that swing us back and forth effect you adversely. Definitely be careful of your speed of play; I tend to play naturally quick but Larry is quite methodical in approach.

3. Preparation, preparation, preparation. I feel this is the most important area to address - it breeds confidence and promotes partnership togetherness.

4. Fight like hell for the partscore.

5. Psyching is a thorny issue; discuss with partner.

6. Lastly, always remind yourself that pros are quite capable of having bad outings and that this is your day to shine. Leave it all out there on the table.
"Champions aren't made in gyms, champions are made from something they have deep inside them - a desire, a dream, a vision. They have to have last-minute stamina, they have to be a little faster, they have to have the skill and the will. But the will must be stronger than the skill. " - M. Ali

#13 User is offline   pclayton 

  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 9,151
  • Joined: 2003-June-11
  • Location:Southern California

Posted 2006-September-16, 12:34

If I am on a truly hopeless team, I might try some of these methods. If I'm on a decent team that happens to be playing a top 16, I would try some of these methods as well.

In the Denver BAM, my teammate opened 3 w/w on KQJxx and out on a 2353 in 2nd chair. Meckstroth / Johnson got to 3 -1. At our table we defended 2 against Rodwell/Soloway and beat it 1 as well.

Loosen up the preempts, and (more importantly) make them on a wide range of hands. The "I have TWO opponents but ONE partner" strategy creates a lot of variance, which is what you are looking for.

But I think the best advice hasn't been mentioned yet: "Don't beat yourself'.
"Phil" on BBO

#14 User is offline   P_Marlowe 

  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 9,916
  • Joined: 2005-March-18
  • Gender:Male

Posted 2006-September-17, 06:31

pclayton, on Sep 16 2006, 01:34 PM, said:

But I think the best advice hasn't been mentioned yet: "Don't beat yourself'.

Very true.

What I would hate most, is to find out, that playing our
normal bridge, we would have had a chance, since the
opponents did have a weak moment.

With kind regards
With kind regards
Uwe Gebhardt (P_Marlowe)

#15 User is offline   ArcLight 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,341
  • Joined: 2004-July-02
  • Location:Millburn, New Jersey
  • Interests:Rowing. Wargaming. Military history.

Posted 2006-September-17, 13:14

Are there any good books on IMP strategy?

Such as:
Winning Swiss Team tactics in Bridge - Feldheim, Harold
Team Tactics in Bridge - Jago, Willie
Rate your team play - Zaluski, Edward

#16 User is offline   FrancesHinden 

  • Limit bidder
  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 8,482
  • Joined: 2004-November-02
  • Gender:Female
  • Location:England
  • Interests:Bridge, classical music, skiing... but I spend more time earning a living than doing any of those

Posted 2006-September-17, 14:36

Let's make this more interesting: I disagree with virtually everything said above.

All these slightly-swingy tactics (off-centre pre-empts, changing the NT range slightly, dropping Qx offisde instead of finessing) are the way you should play when trying to beat a team who will beat you 19 times out of 20, say.

I've lost to teams I'd expect to beat 19 times out of 20 (roughly 5% of the time, in fact) and it's usually either been due to the type of tactics above, or my own team self-inflicting these tactics on themselves (e.g. bidding a good slam and going off) together with playing badly.

If you are playing a team who will really beat you 995 out of 1000 over a normal, shortish KO match - say 24 boards - then you will usually expect to lose by 50 imps (and if anything I'm being generous here). To generate 50 Imps you need, let's say, 4 13-IMP vul game/slam swings in your direction on top of what would normally happen. There will be a load of boards with no opportunity for one of these swingy actions, and some of them may lose, but even if each one has a 20% chance of success and no cost otherwise, you still only have a 0.2^5 chance of pulling it off, which is about 0.3 in 1000 so still not enough...

To beat a team that much better than you, you need three things:
1. Play over as few boards as possible. Ideally one. (ideally, toss a coin.)
2. Be much more random than given above. Psyche a lot.
3. Hope they screw up.

To be honest, if I were playing a team _that_ much better, I would prefer simply to do the best I could and try to learn from the experience. I will save up my 'swinging' actions for when I'm 20 IMPs down with 8 boards to play, or so, in matches where I have a real chance of winning if I get things right.

For a serious lesson in how to swing, look at the last set of the Spingold(???) final from last year when Meckwell came back from 60-odd down to win. Note in passing how lucky they were, as well as how well they played.

#17 User is offline   EricK 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,303
  • Joined: 2003-February-14
  • Location:England

Posted 2006-September-17, 22:48

So I can get a handle on this, what strength difference corresponds to winning 995 times out of 1000? Suppose one team were the top US or Italian team, what level of players should they expect to beat 995 times out of 1000? It seems clear that this will depend on the length of the match, so answers for various match lengths would be nice!

#18 User is offline   jikl 

  • PipPipPipPipPip
  • Group: Full Members
  • Posts: 558
  • Joined: 2004-October-08
  • Location:Victoria, Australia

Posted 2006-September-18, 00:46

There is one thing you all haven't mentioned when playing against a client/pro pair. The client will always signal truly, whether it is count or attitude. The pro will play any card since the client will be too busy worrying about their signals. (Depending on the quality of the client) This is actually worth a large amount of IMPs.


Team Tactics in Bridge - Jago, Willie

This is a really bad book, written by an intermediate/advanced player at the time, and printed by his own company or a friend's company.


#19 User is offline   Cascade 

  • PipPipPipPipPipPipPipPip
  • Group: Yellows
  • Posts: 6,756
  • Joined: 2003-July-22
  • Gender:Male
  • Location:New Zealand
  • Interests:Juggling, Unicycling

Posted 2006-September-18, 00:52

EricK, on Sep 18 2006, 04:48 PM, said:

So I can get a handle on this, what strength difference corresponds to winning 995 times out of 1000? Suppose one team were the top US or Italian team, what level of players should they expect to beat 995 times out of 1000? It seems clear that this will depend on the length of the match, so answers for various match lengths would be nice!

This is my attempt to get a handle on this:

1. We need to know the variation in IMPs per board in reasonable level team's matches.

In the recent PABF the standard deviation in the matches that I played was nearly exactly 7 Imps/board.

I will use this since it was accessible to me. I have done similar calculations in the past based on IMPs traded in world championship finals and got similar although slightly smaller values.

2. We need to know the range of skills in a top level event.

In last years World Championship the butler rankings show a range of about +0.5 Imps/board to -1.0 Imps/board.

This suggests a range of about 3 IMPs/board for a team consisting of two pairs from near the top of the butler versus a team of two pairs near the bottom.

Based on this we can estimate the probabilities that each team will win with some fairly standard statistical assumptions - I have no idea how valid they are in this context so we will gloss over that for now.

To win 99.5% of the time the difference in skill would need to be:

8 boards - about 6.4 IMPs/board. This is about twice the range of skill level at the world championships. So Italy playing a team that Guadeloupe would beat comprehensively.

20 boards - about 4.1 IMPs/board. This is a team a little worse that Guadeloupe.

64 boards - about 2.3 IMPs/board. The chance that a team like Chinese Taipei had of beating Italy.

128 boards - 1.6 IMPs/board. Canada or England beating Italy.

160 boards - 1.5 IMPs/board. Canada or England beating Italy.
Wayne Burrows

I believe that the USA currently hold only the World Championship For People Who Still Bid Like Your Auntie Gladys - dburn
dunno how to play 4 card majors - JLOGIC
True but I know Standard American and what better reason could I have for playing Precision? - Hideous Hog
Bidding is an estimation of probabilities SJ Simon

#20 User is offline   mycroft 

  • Secretary Bird
  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 6,940
  • Joined: 2003-July-12
  • Gender:Male
  • Location:Calgary, D18; Chapala, D16

Posted 2006-September-18, 09:54

As a point of interest:

Spingold and Vanderbilt (after the first day, once we're down to power-of-2) are full-day, 64-board matches.

So it's quite reasonable that a decent, but not great team will be sitting in the .5% category on day 2 (round-of 64) or 3 (round-of-32) against the top 3 or 4 seeds.

I'd even say that seed 15 or 16 on day 4 could easily be Chinese Taipei vs. Italy - especially if they got lucky early (how often do you see a 66th seed get lucky and beat a moderate seed early, and then get teams-at-their-level as the "new" 15th seed?)

When I go to sea, don't fear for me, Fear For The Storm -- Birdie and the Swansong (tSCoSI)

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