Wierd scores in BBO

[majraj]

There are many players bidding 7NTXX

That casues scores to be really starnge

COuld u pls pogram BBO to get rid of cpl of wierd scores ?

[inquiry]

Defining weird scores is not as easy as you might suspect. A lot of people bid 7NT with intention to make. Others are just making spiteful bids to get back at partner, or are jokers who join and jump from table to table doing this.

 

Clearly the jokers are quickly identified and banned. The "normal" people who in a fit of anger or to be cute, take to opening 7NT on nothing or throwing away aces to maximize how much they go down are also subject to being banned.

 

From October 11th to Oct 26 in tourmanents, there were 18 people who opened 7NT on "nothing", and in the main room this occurred 14 times We know who they are, when they did it, etc. The better solution is to educate such people to why you don't do this in a duplicate scoring event, such education might include play restrictions on BBO (blocked from tournaments, blocked from main room, etc).

 

In reality, such things should never end up in the scoring column. These 7NT bids in the main room can be thrown out and the hand redealt by simply removing and reporting the violator. And in tournaments and team games, the director can adjust the score. So SELF=POLICING by members will be very effective at eliminating these actions. Sadly, too many just double, accept the redouble and rack up +7000 points, and never report it to anyone, the director or to the bbo yellows.

[Guest Jlall]

Defining weird scores is not as easy as you might suspect. A lot of people bid 7NT with intention to make.

Indeed, recently I went +4000 against 3N XX (on a speculative X no less) against 2 decent players who were not dumping in the least and were trying to achieve their best result (declarer probably should have cashed winners before taking a key finesse which would have resulted in 1000, or could even cash out for 400 but that is unlikey..instead he just took the finesse...if it was on he would make 1 overtrick but if it was off he was down 7).

 

However, I think instead of trying to throw out weird scores that possibly you could throw out the best score for each direction and then imp it like they do in real life imp pairs. The only problem is sometimes there arent enough comparisons to do this.

[1eyedjack]

Hopefully, in the long term, the 7Nxx "dumpers" will even out, in terms of those sitting in your direction compared with those sitting the other way. To the extent that they do not even out, the frequency should be insignificant to the total number of hands played.

 

Yes, it happens, perhaps more often than we would wish, but the psychological impact tends to assign the event a greater prominence than it probably deserves in the hard light of statistics.

 

I wonder if bridgebrowser can come up with some figures.

[inquiry]

I used BridgeBrowser to look at 25,000 Deals (a total of 142,901 unique auctions) from Tournaments and Team Games on BBO. These auctions took place on oct 12 and 13, i stopped after 25,000 deals. There were 68 7NTX and 7NTxx contracts.

 

That would be this would occur on only 0.27% of all boards (actually 7NTx showed up a couple times on the same hand, so this is a maximum), and only on 0.047% of all auctions. And it isn't as bad as those numbers appear. Of these 68 hands, 16 made (either doubled or redoubled) and 15 were down one, and another 9 down two. Those account for 58.5% of the contracts had a chance, and must have been serious. It is not clear that all the others were intentional. Bidding misunderstandings account for a fair number as well... jump raise as preemptive, partner playing as strong, etc.

 

I just can't see this being a huge problem, but even as rare as this might be, bbo staff looks to stop/educate those who make friviolous bids or plays.

[Gerardo]

Justin (and all):

 

BBO uses CrossIMPs, not Butler, so it can't just discard a result.

 

Looks so much better this way to me (I don't see why anyone would prefer Butler, unless you are doing the computations by hand), though a bigger sample size (currently 16 in MBC) could be useful.

[Guest Jlall]

Justin (and all):

 

BBO uses CrossIMPs, not Butler, so it can't just discard a result.

heh I have no idea what this means...could you clarify? (I don't know what cross imps and butler are).

 

I thought BBO just figured out the average score then imped everyones score vs this average. If that was correct I thought it could just throw out the top/bottom result then figure out the average score,

[cherdano]

Justin (and all):

 

BBO uses CrossIMPs, not Butler, so it can't just discard a result.

heh I have no idea what this means...could you clarify? (I don't know what cross imps and butler are).

 

I thought BBO just figured out the average score then imped everyones score vs this average. If that was correct I thought it could just throw out the top/bottom result then figure out the average score,

What you describe is Butler. Instead, BBO IMPs against every other result, and only then takes the average of those IMPs.

Theoretically certainly superior, as IMP odds for bidding vulnerable games etc. change somewhat with butler, while with IMP-across-the-field they are exactly the same as in a team match. Also, it automatically reduces the effect of extreme scores, due to the IMP scale, instead of tossing them out somewhat artifically.

 

(Cavendish uses the same.)

 

Arend

[pclayton]

Defining weird scores is not as easy as you might suspect. A lot of people bid 7NT with intention to make.

Indeed, recently I went +4000 against 3N XX (on a speculative X no less) against 2 decent players who were not dumping in the least and were trying to achieve their best result (declarer probably should have cashed winners before taking a key finesse which would have resulted in 1000, or could even cash out for 400 but that is unlikey..instead he just took the finesse...if it was on he would make 1 overtrick but if it was off he was down 7).

 

However, I think instead of trying to throw out weird scores that possibly you could throw out the best score for each direction and then imp it like they do in real life imp pairs. The only problem is sometimes there arent enough comparisons to do this.

Nah, I was just dumping. :P

[kenrexford]

At matchpoints, where "dumping" is not so much a problem, I once managed to find the elusive -7600 "intelligently." MANY MANY years ago.

 

I had opened 1NT (15-17) with two Aces, two Kings, one Queen, and a Jack under an Ace-King.

 

LHO doubled, a one-suited hand.

 

Partner had a problem. We played "systems on" after the double, but she forgot that we played a weird 4♦ convention (her idea). For some bizarre reason, she was also blessed with that rare hand where for some similarly bizarre reason a four-level diamond preempt looked right, vulnerable at that. So, she jumped to 4♦.

 

It turns out that this was an asking bid, asking for strength up the line. I responded 4NT, showing 17.

 

She then answered aces, 5♦, as she held the diamond Ace.

 

I then answered her 5♦ call, which was strangely enough also Ace-asking, depending upon who remembered the agreement. My answer? 5NT of course.

 

She answered kings, 6♦, which strangely enough also asked me for Kings.

 

I held two, so 6NT.

 

With a void, she opted to correct this to 7♦.

 

This may sound fabricated, but 7♦ actually asked for Queens. The theory was to decide between the grand in a major or 7NT. When I answered 7♠, her void, she bid the grand in notrump.

 

This was doubled. Being matchpoints, I saw no harm in sending it back.

 

So, phase one was met -- bidding 7NTXX vulnerable.

 

I cannot remember the play, but suffice it to say that it takes some true skill to manage to lose all thirteen tricks after bidding 7NT XX intentionally. However, given the contract, this seemed the only course that a player would take. It took several jettison plays, but I managed to reach that Holy Grail. -7600.

[Gerardo]

Thanks Arend. Unlike the Cavendish, BBO takes the extra step to divide the final sum of IMPs by the number of comparisons (= tables - 1), so the results look like Butler, but they are not.

[Guest Jlall]

Thanks Arend. Unlike the Cavendish, BBO takes the extra step to divide the final sum of IMPs by the number of comparisons (= tables - 1), so the results look like Butler, but they are not.

Thanks, now I know the difference :P

[hotShot]

...

I don't see why anyone would prefer Butler,

...

CrossIMPs create a IMP difference where no difference should be.

 

Assume a board where 4♠ (420) was made 15 times and 1 table made 6♠-2(100).

 

Assume you defended 4♠ so your score is:

Those who played 4♠ have produce 14 times a score of 0.

One time you loose 620 => 12 IMPs because your virtual teammates lost.

 

So you loose 0.8 CrossIMPs against your table opps for making the best possible play your side can make.

 

(Side note: MP scoring produces the same artificial difference!)

 

While calculating Butler style:

The best and the worst result gets eliminated and the average score is used as reference.

In the example case the reference score is 420. You and your table opps get 0 IMPs for playing 4♠= as it should be.

 

So basicly:

CrossIMPs create artificial CrossIMP differences between pairs at the same table, awarding/punishing you for sitting on the same/opposite side with an "unfortunate" pair. Over a few hundred boards that might even out, but over a few boards it usually won't.

 

So whenever one score is dominant by far, Butler scoring (eliminating enough noise) is superior.

[paulg]

...

I don't see why anyone would prefer Butler,

...

CrossIMPs create a IMP difference where no difference should be.

 

Assume a board where 4♠ (420) was made 15 times and 1 table made 6♠-2(100).

 

Assume you defended 4♠ so your score is:

Those who played 4♠ have produce 14 times a score of 0.

One time you loose 620 => 12 IMPs because your virtual teammates lost.

 

So you loose 0.8 CrossIMPs against your table opps for making the best possible play your side can make.

 

(Side note: MP scoring produces the same artificial difference!)

 

While calculating Butler style:

The best and the worst result gets eliminated and the average score is used as reference.

In the example case the reference score is 420. You and your table opps get 0 IMPs for playing 4♠= as it should be.

 

So basicly:

CrossIMPs create artificial CrossIMP differences between pairs at the same table, awarding/punishing you for sitting on the same/opposite side with an "unfortunate" pair. Over a few hundred boards that might even out, but over a few boards it usually won't.

 

So whenever one score is dominant by far, Butler scoring (eliminating enough noise) is superior.

 

 

Butler scoring is superiour here, but there are other

I disagree.

 

Clearly the 6♠-2 could be the result of poor play. It could also be the result when the defending pair found a good 6♣ sacrifice and the other pair misjudged. In this case you would wish to include the score.

 

In a field of equal standing I prefer X-IMPs and not to discard any scores. In a field of unequal standing it matters little which you play (and is 0.8 IMP really worth worrying about).

 

There is further discussion of butler methods on David Stevenson's web site.

 

Paul

[1eyedjack]

...

I don't see why anyone would prefer Butler,

...

CrossIMPs create a IMP difference where no difference should be.

 

Assume a board where 4♠ (420) was made 15 times and 1 table made 6♠-2(100).

 

Assume you defended 4♠ so your score is:

Those who played 4♠ have produce 14 times a score of 0.

One time you loose 620 => 12 IMPs because your virtual teammates lost.

 

So you loose 0.8 CrossIMPs against your table opps for making the best possible play your side can make.

 

(Side note: MP scoring produces the same artificial difference!)

 

While calculating Butler style:

The best and the worst result gets eliminated and the average score is used as reference.

In the example case the reference score is 420. You and your table opps get 0 IMPs for playing 4♠= as it should be.

 

So basicly:

CrossIMPs create artificial CrossIMP differences between pairs at the same table, awarding/punishing you for sitting on the same/opposite side with an "unfortunate" pair. Over a few hundred boards that might even out, but over a few boards it usually won't.

 

So whenever one score is dominant by far, Butler scoring (eliminating enough noise) is superior.

This has been repeatedly thrashed out in this forum in previous threads, and the above argument is no more convincing now than it was then.

 

For all we know, 6S might be a priori odds-on and yet fail, with all but two pairs failing to appreciate the potential on the hand.

 

But the par result is rather irrelevant. You are playing effectively 15 separate one-board teams matches. If you are numbered among the 4Spades making then you rightly score flat against the other 13 tables that score the same. If at two tables the contract is 6S -2, what reason should you have to suggest that those mini-matches are any less signficant? When you play in a REAL teams match, do you elect to discard all the results where team-mates happen not to find the par result?

[hotShot]

I disagree.

 

Clearly the 6♠-2 could be the result of poor play. It could also be the result when the defending pair found a good 6♣ sacrifice and the other pair misjudged. In this case you would wish to include the score.

 

In a field of equal standing I prefer X-IMPs and not to discard any scores. In a field of unequal standing it matters little which you play (and is 0.8 IMP really worth worrying about).

 

...

 

Paul

The beauty /correctness of scoring is that equal performance should lead to equal score.

 

Perfect scoring (team fight):

Your table scores NS 4♠= and the other table scores 4♠=.

So both teams score 0 IMPs here.

 

If 30 pairs perform equal to this team result and 2 pairs have a different result, there is some beauty giving those 30 pairs 0 IMPs for their result.

Some just benefit from the fact that they where at the same side as some "unfortunate" pair the others suffer from being on the other side.

 

I prefer to get / loose IMPs for skill and not for luck in seating.

[1eyedjack]

The beauty /correctness of scoring is that equal performance should lead to equal score.

.... which result is still achieved in X-Imp scoring. You score equally with all of the other players who score the same result as you. It is just not a zero score, because in common with all of the others who also score the same, you all also score the same non-zero result against those in 6S

[hotShot]

For all we know, 6S might be a priori odds-on and yet fail, with all but two pairs failing to appreciate the potential on the hand.

But the par result is rather irrelevant. 

You introduced the par result to the discussion, and is is usually used to describe what a double dummy solver could make.

If i were talking of a par result concerning scoring i would define it as:

 

The par result of real life bridge is the result produced by the majority of the player.

 

You are playing effectively 15 separate one-board teams matches.  If you are numbered among the 4Spades making then you rightly score flat against the other 13 tables that score the same.  If at two tables the contract is 6S -2, what reason should you have to suggest that those mini-matches are any less signficant?  When you play in a REAL teams match, do you elect to discard all the results where team-mates happen not to find the par result?

 

If real life bridge my team-mates are my choice and i must accept to get a bad result because of this choice. With CrossIMPs (especially at BBOs MBC) i win my team-mates by seating.

 

With CrossIMP's the seating creates a noise of about 1 IMP/Board.

I consider that to much.

[hotShot]

The beauty /correctness of scoring is that equal performance should lead to equal score.

.... which result is still achieved in X-Imp scoring. You score equally with all of the other players who score the same result as you. It is just not a zero score, because in common with all of the others who also score the same, you all also score the same non-zero result against those in 6S

Bridge scoring is far from being fair.

It does make a difference what hand you play against which opps.

It does make a difference which position you sit at the start of a tourney.

With CrossIMPs and MP it also matters which side of the deal you are.

 

People have put tremendous efforts in reducing the noise that is introduced into the result by movement. Letting you play against as many pairs as possible and changing directions as often as possible are the means used to approach equity.

The best movement so far is the 7 table Howell. But even in this best movement, two pairs of equal strength can end up e.g. 49% and 46% just because of the position they started.

 

Since side matters CrossIMP results of EW and NS should not be compared.

 

In the MBC a score for NS and OW is given in a way that creates the impression that the results are comparable. At BBO tourneys CrossIMPS are used to declare one winner.

 

So if you want to compare NS and EW result of a board, you have to award both sides 0 IMP's, if they made what the majority made. CrossIMPs as used now fail to do that.

[paulg]

As noted in the link from my previous post, a median variant of Bulter scoring is certainly a viable method scoring and mode (most common result) would work too, but they also have flaws.

 

According to David Grabiner of the USA, Robin Barker of England and Jeff Goldsmith of the USA the disadvantage of the median method of scoring [taking the middle result as the datum, or the arithmetic mean between the two middle results with an even number of frequencies] is primarily because of "polarised" boards. If there is a close 3NT that everyone will bid but might make or go off, then you might get a set of scores such as:

 

+1100 +600 +600 +600 +600 +600 +600 -100 -100 -100 -100 -100 -200

 

Using the median method the datum is +600 and the imp scores become

 

+11 0 0 0 0 0 0 -12 -12 -12 -12 -12 -13

 

Now, if one score is changed from +600 to -100 the datum becomes -100 and the imp scores become

 

+15 +12 +12 +12 +12 +12 0 0 0 0 0 0 -3

 

It does not seem reasonable that changing one score should make such an enormous difference. However, when the boards are not polarised this looks quite a fair method. It has the advantage that the datum score is a "real" score for people to imp against, and on non-polarised boards people quite like it.

No method is perfect, we all have our preferences, and most threads on this topic finish with everyone agreeing to disagree :P

 

Paul

[hotShot]

As noted in the link from my previous post, a median variant of Bulter scoring is certainly a viable method scoring and mode (most common result) would work too, but they also have flaws.

 

According to David Grabiner of the USA, Robin Barker of England and Jeff Goldsmith of the USA the disadvantage of the median method of scoring [taking the middle result as the datum, or the arithmetic mean between the two middle results with an even number of frequencies] is primarily because of "polarised" boards. If there is a close 3NT that everyone will bid but might make or go off, then you might get a set of scores such as:

 

+1100 +600 +600 +600 +600 +600 +600 -100 -100 -100 -100 -100 -200

 

Using the median method the datum is +600 and the imp scores become

 

+11 0 0 0 0 0 0 -12 -12 -12 -12 -12 -13

 

Now, if one score is changed from +600 to -100 the datum becomes -100 and the imp scores become

 

+15 +12 +12 +12 +12 +12 0 0 0 0 0 0 -3

 

It does not seem reasonable that changing one score should make such an enormous difference. However, when the boards are not polarised this looks quite a fair method. It has the advantage that the datum score is a "real" score for people to imp against, and on non-polarised boards people quite like it.

No method is perfect, we all have our preferences, and most threads on this topic finish with everyone agreeing to disagree :P

 

Paul

Paul i don't understand what you are trying to say.

 

Gerado asked why anybody would want to prefer Butler to CrossIMP's.

 

The answer to that question is, CrossIMPs are very unfair to almost half of the field, if a board has a very unusual result, especially if there is a dominant score. It does not even matter whether that unusual result was caused by playing perfect bridge or producing utter nonsense.

 

I was not answering the question why someone would not want to use Butler scoring.

 

On BBO especially in the MBC we have a unusual score on most boards. So most boards are scored unfair. Since there are only 16 scores available the unfairness is significant. If there were 100 scores it would be much less relevant.

For those boards Butler scoring would definitely be an improvement.

 

I know this modified Butler scoring method, but i don't think it has enough benefits to use it. Original Butler scoring compares everyones score against the average not against the median.

Using a median only makes sense if there is one. Using the average is more stable even if there are no scores close to the average.

[1eyedjack]

If real life bridge my team-mates are my choice and i must accept to get a bad result because of this choice.

And yet you may not have the luxury of choosing who your opponents are at the other table. 50% of the so-called skewed results that arise at the other table, perhaps more, may be attributable to their action.

 

So I ask you this. In a one-against-one teams match, if you bid and make a normal 4S, and your opponents sitting your way at the other table overreach to bid 6S going down, are you suggesting that you should only get zero IMPs, on the grounds that you took the normal action?

[cherdano]

Guys get a life. :) One unusual result just gives you less than an IMP. If you compare this to the 3-5 IMPS you get in main room for bidding an obvious game on 23 hcp...

 

Also, sometimes the one extreme result is just the only table who found a good 6♠ making. Why would you throw this out?

 

I am like Gerardo, I really fail to understand why anyone would prefer any scoring method over CrossIMPs.

[hotShot]

Guys get a life. :) One unusual result just gives you less than an IMP. If you compare this to the 3-5 IMPS you get in main room for bidding an obvious game on 23 hcp...

 

Also, sometimes the one extreme result is just the only table who found a good 6♠ making. Why would you throw this out?

 

I am like Gerardo, I really fail to understand why anyone would prefer any scoring method over CrossIMPs.

When we discuss rating, winning an IMP/board is considered a strong performance.

Loosing an IMP/board by sitting on the "wrong side" of table does not even matter to you?

I will remember that :).

[Blofeld]

But the IMP/board is not consistently in the same direction ... in fact it's going to be pretty random.

 

I can't imagine preferring anything to cross-IMPs. Though a wider field of comparison would be good, if possible.