Matchpoint Expectation Question

[AAr]

I overbid my hand slightly and we reached 3NT. Most of the Field will be in 1NT, we expect and am pretty sure on. Our game depended on a finesse, which lost, and therefore we went down. We expect most of the Field to be in 1NT.

 

My partner says we should expect a very poor MP score because I made an anti-field anti-odds decision that did not work out while most of the field will have better scores. (We were -50 and that does lose to +120, which my partner thinks is the normal score, so my partner might have somewhat of a point.)

 

I say that we would have had a good score had the game made (and would have if the finesse would've won), +400 would've been better than +150 had the game made, and we might still get a good board anyway.

 

So, who'se right? Should I expect a good (or at least average board). Or, is my partner right?

 

Thanks!

[pclayton]

An actual hand would help.

 

For arguments sake, lets say you upgraded a 14 count to a 1N opening and your pard put you in 3N with a good 9 or 10.

 

The hands mesh well, but game is dependent on a finesse, which loses.

 

I'd say you can look forward to about a 10-20% score on this board. You would have had an 80-90% score if the hook won.

 

If variance is your thing, then go for it. But if you want to keep you results in the 40-60% range on hands like this, you should play with the field.

[AAr]

Actually, we ended up with a cold bottom. We would've gotten a 60% board had I went most with the Field, which played at 1NT, scoring +120, beating my -50. B) So, apparently, my partner (and pclayton) were right, and I was wrong.

 

Was the bottom to be of any surprise? Or, would a good board have been a surprise?

[skjaeran]

I don't understand what you're after here.

 

You say that you expect the field to be in 1NT. By that premise, it's obvious that you'll get a near top for making 3NT and a near bottom when you go down.

 

This is rather obvious if you make an anti-field action. Wheter your action was wrong is hard to say, lacking the hand and bidding. The result proves that it was anti-field though.

 

Thus, failing in 3NT, a bottom was hardly surprising.

[Stephen Tu]

Geez, how many years have you been at this, and you still haven't figured it out yet? You are the king of silly questions here and on rec.games.bridge.

 

If you do something not many people will do, and end up scoring less than a normal action would do, of course you should expect a poor score.

 

What is so hard to believe about that? 50% of your questions are of the approximate form "I did [something abnormal/anti-percentage] that resulted in a worse bridge score than the action most others would take. This translated to a poor [matchpoint/IMP] score. Is this surprising?"

 

No, it's not, and never will be. Bad scores in bridge come when you do worse than the field, and you should never be surprised about this. Good scores come when you achieve a score that you think is *better* than what most people would score, not worse. Average scores come when you think you've scored the same as most people. Certainly going minus when you think most people will be plus will never qualify as a good score. You'd expect a good score if you had made the game, and think that no one else would bid it. Being surprised at a bad score when going down in such a game is just completely illogical. MP and IMPS are directly correlated to the size of your score vs. the other pairs. If your score is better than most, it will be good. If it's worse, it'll be bad.

 

Take percentage actions, your scores will improve. Even if you manage to make 9 tricks, at MP it won't be necessary to bid game if it's a clear overbid; taking 9 tricks will still beat other people if they only take 8. Only expect good scores when you take more tricks than other people would in normal contracts, bid more accurately than most would, or get to abnormal contracts that score better for your side than the normal contract. Hoping for a good score when you've screwed up somehow is just kind of loony. Bridge is mostly a game of trying not to make mistakes, and doing stuff to hopefully cause the opponents to make mistakes. If you make any kind of mistake, just expect a poor score, be surprised if it turns out average, not if it comes back low. Try to make fewer mistakes in the future, that's all you can really do.

[gwnn]

Most of the Field will be in 1NT, we expect and am pretty sure on. (...)We expect most of the Field to be in 1NT.

My partner says we should expect a very poor MP score because I made an anti-field anti-odds decision that did not work out while most of the field will have better scores. (...)

I say that we would have had a good score had the game made. So, who's right? (...)

So basically you know everyone or most will score better than you. Thus, you know that you'll get a bad score.

 

Your partner says you'll get a bad score. You say you would have received a good result had the game made. You're both right ;). You didn't actually make contradictory points.

 

The more interesting question is: is it the percentage action or a non-percentage action to bid an exactly 50% game @ MP? But this question has already been answered (in this thread also by Phil): in the long run it makes no difference, as long as the game is indeed 50% and everybody plays in the same denomination. For example 120 beats 110 regardless of the chances of 3NT.

 

I hope any of this helps.

[Free]

It depends. If you expect 60% if you go with the field, then your option is losing. If you expect 40% if you go with the field, your option is winning in the long run.

 

I mean, bidding this 3NT is clearly going to be a top or a bottom, nothing in between. So if the gains are 40% and the losses are 60% (you expect 60% on average) on a 50-50 decision, you took a wrong action.

 

Always hard to calculate how much the field will win ofcourse ;)

[mcphee]

Playing pairs you simply do not make over aggressive bids in random situations. While there are times that your action will result in better scores, the majority of times it will fail because your action is anit field.

 

An example you may understand better, is a person who is playing a hand the can take a particular number of tricks, it is not important how many. The line of play selected has 2 ways to success, one line is 100% and another line is 50%, a finesse or a strip and end play. On this particular hand the hook works and the scores are therefore equal. Surely we do not believe the 50% line of play was ok only because it worked. This type of thing happens all the time and sadly many players overlook what took place because they SCORED the same. This is just about the same as a person who just played a hand in part score and made 10 tricks although the hand could have been held to 9 tricks, and they say, we should have bid game, or berate partner for failing to bid game because "other people made it" . Resulting will not help anyone learn this game.

[bid_em_up]

I overbid my hand slightly and we reached 3NT.

Ok, so you know you have overbid.

 

Most of the Field will be in 1NT, we expect and am pretty sure on.

You evidently know that you are in an anti-field contract.

 

Our game depended on a finesse, which lost, and therefore we went down. We expect most of the Field to be in 1NT.

 

My partner says we should expect a very poor MP score because I made an anti-field anti-odds decision that did not work out while most of the field will have better scores. (We were -50 and that does lose to +120, which my partner thinks is the normal score, so my partner might have somewhat of a point.)

Somewhat of a point? What part of being plus +120 is always better than being -50 are you failing to comprehend?

 

I say that we would have had a good score had the game made (and would have if the finesse would've won), +400 would've been better than +150 had the game made,

Yes, you would have had either a top or close to top had 3N made but.....

 

and we might still get a good board anyway.

You don't have a snowballs chance in hell of receiving a good board anyway. You have already acknowledged that the majority of the field will be +120. You are -50. How many people do you expect to outscore? Very few. It isn't possible to receive a good board. Statements like this make no sense.

 

So, who'se right? Should I expect a good (or at least average board). Or, is my partner right?

Your partner is. And you should expect a terrible score for the result.

[kenrexford]

Actually, it depends upon the actual field.

 

In my home club game, you might get a fairly good score. Sure, the "field" action might be 1NT, making two. However, you might beat a few people, like 4♦ redoubled down two, 1♥ the other way, doubled, making six, and 1NT down two.

[pclayton]

If I'm in a pairs event where I don't expect to be the favorite, or even close, I WANT to be in games like this. 40-60% scores on this will only cement my standing in the middle.

 

Say in a four session event I expect to make the cut, but still finish about 4 boards behind the winners if we play our normal game. That result will put us in the low overalls. We'll make plenty of judgment errors and a few mechanical errors. Plus, there's the celebrity factor. A lot of name players will get some gifts from some of the weaker players by just showing up at the table. I can't expect to get these.

 

A few 50-50 boards that go my way can do wonders for my scores. On the given hand, I'm guessing I will pick up 1/2 - 3/4 board by just being 'lucky' and being on the right side. If I can get 6 of these decisions right in the last two sessions, and tighten up some of our decisions, we give ourselves a chance of winning.

 

Gaging issues like this are pretty tough. Sometimes it just isn't a straight coin flip, sometimes by swinging you put yourself in a underdog position.

 

I think I heard that Greg Raymer won 28 'coin-flips' when he won the WSOP a few years ago.

[bid_em_up]

Actually, it depends upon the actual field.

 

In my home club game, you might get a fairly good score.  Sure, the "field" action might be 1NT, making two.  However, you might beat a few people, like 4♦ redoubled down two, 1♥ the other way, doubled, making six, and 1NT down two.

Under the scenario given by the OP, he "knows" that the majority of the field will likely be in 1N.

 

This, to me, precludes such silly contracts as 4D xx'd -2 or 1H x'd making 6. But even assuming those two results on a 12 top, you would likely have something similar to the following.

 

3N=......400

1N+2....150

3S=......140

1N+1....120

1N+1....120

1N+1....120

1N+1....120

1N+1....120

3D=......110

1N=......90

3N-1.....................50

4Dxx -2................600

1Hx'd+5...............?? whatever number it is.

 

Now score it.

 

Pair 1 gets 12.........100%

Pair 2 gets 11.........91.6%

Pair 3 gets 10.........83.33%

Pairs 4-8 gets 7......58.33%

Pairs 9 gets 4.........33.33%

Pair 10 gets 3.........25.0%

Our pair gets 2.......16.67%

Pair 12 gets 1.........8.33%

Pair 13 gets 0.........0%

 

Clearly, it cannot be right to bid 3N, when 1N+2 would have resulted in 87.5% (10-). If we make three NT, we are getting the majority of the matchpoints, even without bidding it.

 

We have turned a win/win situation (1N+2 or 1N+1 for an above average score in either case) into either an absolute but unlikely good result vs. a probable poor result by bidding 3N.

[bid_em_up]

If I'm in a pairs event where I don't expect to be the favorite, or even close, I WANT to be in games like this. 40-60% scores on this will only cement my standing in the middle.

 

Say in a four session event I expect to make the cut, but still finish about 4 boards behind the winners if we play our normal game. That result will put us in the low overalls. We'll make plenty of judgment errors and a few mechanical errors. Plus, there's the celebrity factor. A lot of name players will get some gifts from some of the weaker players by just showing up at the table. I can't expect to get these.

 

A few 50-50 boards that go my way can do wonders for my scores. On the given hand, I'm guessing I will pick up 1/2 - 3/4 board by just being 'lucky' and being on the right side. If I can get 6 of these decisions right in the last two sessions, and tighten up some of our decisions, we give ourselves a chance of winning.

 

Gaging issues like this are pretty tough. Sometimes it just isn't a straight coin flip, sometimes by swinging you put yourself in a underdog position.

 

I think I heard that Greg Raymer won 28 'coin-flips' when he won the WSOP a few years ago.

Phil, this is all well and good, and I can even agree with you. But...you "know" you are swinging when you take this approach and that it is pretty much a top or bottom approach (in respect to this particular board).

 

However, the original poster leaves me with the impression that he is attempting to justify his bid on the basis of "well, I know I overbid but it might make" and "It is still possible that we can get a good score even if it goes down".

 

This doesn't strike me as someone who is taking a swinging approach during a session, but instead as someone who made a bad bid, and he knows he has made a bad bid, his partner knows it, and everyone else knows it, and instead of simply saying "I made a bad bid, sorry", he is now is looking for sympathy, solace, support, justifications or rationalizations for his bid.

 

I won't provide him with any.

[pclayton]

Actually, it depends upon the actual field.

 

In my home club game, you might get a fairly good score.  Sure, the "field" action might be 1NT, making two.  However, you might beat a few people, like 4♦ redoubled down two, 1♥ the other way, doubled, making six, and 1NT down two.

Under the scenario given by the OP, he "knows" that the majority of the field will likely be in 1N.

 

This, to me, precludes such silly contracts as 4D xx'd -2 or 1H x'd making 6. But even assuming those two results on a 12 top, you would likely have something similar to the following.

 

3N=......400

1N+2....150

3S=......140

1N+1....120

1N+1....120

1N+1....120

1N+1....120

1N+1....120

3D=......110

1N=......90

3N-1.....................50

4Dxx -2................600

1Hx'd+5...............?? whatever number it is.

 

Now score it.

 

Pair 1 gets 12.........100%

Pair 2 gets 11.........91.6%

Pair 3 gets 10.........83.33%

Pairs 4-8 gets 7......58.33%

Pairs 9 gets 4.........33.33%

Pair 10 gets 3.........25.0%

Our pair gets 2.......16.67%

Pair 12 gets 1.........8.33%

Pair 13 gets 0.........0%

 

Clearly, it cannot be right to bid 3N, when 1N+2 would have resulted in 87.5% (10-). If we make three NT, we are getting the majority of the matchpoints, even without bidding it.

 

We have turned a win/win situation (1N+2 or 1N+1 for an above average score in either case) into either an absolute but unlikely good result vs. a probable poor result by bidding 3N.

I don't think this is a relative scatter of results.

 

The poster mentioned a scenario where the field took 8 tricks, but they bid 3N.

 

If there was some variability amongst the abilities of the declarer, and either 8 or 9 tricks were avilable, then playing the field contract, and garnering most of the matchpoints (though not an absolute top) for the uptrick makes sense.

[pclayton]

If I'm in a pairs event where I don't expect to be the favorite, or even close, I WANT to be in games like this. 40-60% scores on this will only cement my standing in the middle.

 

Say in a four session event I expect to make the cut, but still finish about 4 boards behind the winners if we play our normal game. That result will put us in the low overalls. We'll make plenty of judgment errors and a few mechanical errors. Plus, there's the celebrity factor. A lot of name players will get some gifts from some of the weaker players by just showing up at the table. I can't expect to get these.

 

A few 50-50 boards that go my way can do wonders for my scores. On the given hand, I'm guessing I will pick up 1/2 - 3/4 board by just being 'lucky' and being on the right side. If I can get 6 of these decisions right in the last two sessions, and tighten up some of our decisions, we give ourselves a chance of winning.

 

Gaging issues like this are pretty tough. Sometimes it just isn't a straight coin flip, sometimes by swinging you put yourself in a underdog position.

 

I think I heard that Greg Raymer won 28 'coin-flips' when he won the WSOP a few years ago.

Phil, this is all well and good, and I can even agree with you. But...you "know" you are swinging when you take this approach and that it is pretty much a top or bottom approach (in respect to this particular board).

 

However, the original poster leaves me with the impression that he is attempting to justify his bid on the basis of "well, I know I overbid but it might make" and "It is still possible that we can get a good score even if it goes down".

 

This doesn't strike me as someone who is taking a swinging approach during a session, but instead as someone who made a bad bid, and he knows he has made a bad bid, his partner knows it, and everyone else knows it, and instead of simply saying "I made a bad bid, sorry", he is now is looking for sympathy, solace, support, justifications or rationalizations for his bid.

 

I won't provide him with any.

If someone is looking for justification for bad bidding, I won't provide him with any either.

 

If I'm 'swinging', then I'm swinging, thats all there is to it. I might make a spec double trying for a good lead; or maybe trying for 200. I might psyche. I might make an anti-field 1N opening on a 4333 15.

[cherdano]

If I'm in a pairs event where I don't expect to be the favorite, or even close, I WANT to be in games like this. 40-60% scores on this will only cement my standing in the middle.

 

Say in a four session event I expect to make the cut, but still finish about 4 boards behind the winners if we play our normal game. That result will put us in the low overalls. We'll make plenty of judgment errors and a few mechanical errors. Plus, there's the celebrity factor. A lot of name players will get some gifts from some of the weaker players by just showing up at the table. I can't expect to get these.

 

A few 50-50 boards that go my way can do wonders for my scores. On the given hand, I'm guessing I will pick up 1/2 - 3/4 board by just being 'lucky' and being on the right side. If I can get 6 of these decisions right in the last two sessions, and tighten up some of our decisions, we give ourselves a chance of winning.

Well, having all 6 of these going your way is a 1.6% chance. I would rather try to play well and get a decent result as often as possible than shoot for a placing in the top 3 one out of 60 times I take part.

 

Edit: % typo fixed

Edited May 1, 2007 by cherdano

[bid_em_up]

I don't think this is a relative scatter of results.

 

The poster mentioned a scenario where the field took 8 tricks, but they bid 3N.

 

If there was some variability amongst the abilities of the declarer, and either 8 or 9 tricks were avilable, then playing the field contract, and garnering most of the matchpoints (though not an absolute top) for the uptrick makes sense.

I don't think its a relative scatter of the actual table results either. It's just some numbers I made up as a response to kenrexford's comments regarding his home game. ;)

 

Of course they could easily go in a different direction, but in the context of the original post, combined with kens post, seem to be a reasonable approximation of what "could" happen, especially at ken's home game.

 

What probably "did" happen was:

 

1N+2 (one or two times) +150

1N+1 or 2N= (10-11 times) +120

3N -1 (one or two times). -50

 

Still the same zero or almost zero, where by just being in 1N+2 you would tie for top, and at least average for 1N+1.

 

That was all I was trying to point out.

[SoTired]

Why do most people "swing" by overbidding? It is a stupid gamble. If you want to swing, try underbidding instead. When you swing by overbidding, you only win when the overbid works. If you swing by underbidding, you still get your top if the hand can't make, but even if wrong, you have a positive score and beat the people that got lost in the wrong suit. For example, you bid stayman, find a fit and would normally bid 4M, but decide to swing by only inviting and end up in 3M. Your underbid fails, but you still beat those that skipped stayman and get to a hopeless 3N.

[bid_em_up]

Well, having all 6 of these going your way is a 60% chance. I would rather try to play well and get a decent result as often as possible than shoot for a placing in the top 3 one out of 60 times I take part.

Arend, I swear I am seeking info here. I know you are probably better in math than I am from some of your other posts. I freely admit that I'm not a math whiz. But I don't see how are the chances of all six going "your" way equal to 60%? Is this a typo?

 

Assume that the odds of a contract are exactly 50%. Say there is a finesse to take, and there are 8 tricks if it loses, 9 if it wins.

 

Since each board was originally 50%, you would expect to make 9 tricks on three of them and 8 tricks on the other three, on overall averages.

 

But, aren't the odds of all 6 being right consecutively calculated as:

 

0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 = 1.56% of actually being right 6 times in a row.

 

Or have I totally forgotten my statistic/probabilites calculations?

[gwnn]

1♥x+5 NV is 660.

[awm]

I think Arend means "one in 60" and not 60%. In fact the odds are 1/64 but "one in 60" is close enough.

 

There is a general question about swinging in long pair events. When you play in such an event, is your goal to win the event or is your goal to do well in the event, and what is doing well anyway?

 

If your goal is to win and you don't really care if you finish 2nd or 20th or 200th because they are all losing then it makes sense to take a lot of swingy actions. After all, even if you are in fact the best pair in the field your chance of placing first is not very good (well unless all the other pairs are really bad). You will need a lot of luck to go your way.

 

On the other hand, there is something to be said for consistently high finishes, especially in important or prestigious events. It's not clear that given the choice between placing in the top ten 20 times in a row (but never winning) versus winning once and being far down in the rankings the other 19 times, that we'd all choose the one win. After all, winning once could be the result of luck whereas being in the top ten every single year almost has to be based on skill. Then again, if there is a big cash prize for winning (but nothing for second) maybe it's better to win once...

 

In any case, there is nothing wrong with taking anti-field actions at MP. Sometimes these actions are even "right" in that you are more likely to outscore the field than to lose to the field. But any time you take an unusual action, you should expect a near-top or near-bottom result (yes there will be other outliers who did something bizarre, but you will generally get a different score from the majority of the field). If your unusual action is "right" you will get a near-top more often than a near-bottom and if it's "wrong" you will see the opposite, but in any case there is luck involved. Whether you want to increase the amount of luck in your score depends on what you want to do and what you expect to manage by "playing down the middle." But in any case, it should be no surprise that if you take an anti-field action and it doesn't work out because you went down in your pushy game, or the close contract you doubled made, or the close game you avoided by being conservative was cold... you will get a near-zero. That's not to say anti-field actions are necessarily to be avoided in all circumstances, because when they do work out you should expect the near-top. It's all a game of percentages.

[kenberg]

If I overbid and go down am I apt to get a bad score? Let me do a little research on this and I'll get back to you. Right now I am working on: If I drive 90 mph down a highway without wearing a seatbelt and crash my car, am I apt to get hurt?

 

I think we are being played.

[mycroft]

Please note that Phil is simply saying if he takes a few coin flips in 100 boards, and he's up 6, he has a shot at winning.

 

That's not a 6-way parlay. It could be 7 of 8, 8 of 10, or otherwise.

 

That's more than 1.6%. Still not good, but hey, if your goal is to Win The NABC+ Event against The Best, and your chance of doing it playing straight up is zero.whatever%, play for the 3,4% chance. After all, for GrandLM you only need one...

 

Michael.