[charles311]

How when I see a partnership win a contract even with over tricks they lose IMPs?

[Vampyr]

Someone recently posted the same question.

The simple answer is that some pairs scored more points on a board. Is it possible to view the scores achieved on a board? If so, do that and you will have your answer.

[NickRW]

At teams, if you contract for 3♠ and make and over trick you score 4*30+50 = 170. If the opponents on the other table play in game (i.e. 4♠) making the same 10 tricks they score 4*30+300 = 420 (more if they are vulnerable). Which makes your team minus 250 overall. I never remember the scoring chart for raw scores to imps, but it is about minus 6 or so

If it is imp scored pairs the detail is slightly different (more complicated), but the upshot is the same.

[charles311]

NickRW, on 2017-January-26, 11:10, said:

At teams, if you contract for 3♠ and make and over trick you score 4*30+50 = 170. If the opponents on the other table play in game (i.e. 4♠) making the same 10 tricks they score 4*30+300 = 420 (more if they are vulnerable). Which makes your team minus 250 overall. I never remember the scoring chart for raw scores to imps, but it is about minus 6 or so

If it is imp scored pairs the detail is slightly different (more complicated), but the upshot is the same.

[charles311]

Thanks to you both. I am aware about the scoring for teams. It was just a casual pairs game I was watching. That's what confused me.

[rmnka447]

NickRW, on 2017-January-26, 11:10, said:

At teams, if you contract for 3♠ and make and over trick you score 4*30+50 = 170. If the opponents on the other table play in game (i.e. 4♠) making the same 10 tricks they score 4*30+300 = 420 (more if they are vulnerable). Which makes your team minus 250 overall. I never remember the scoring chart for raw scores to imps, but it is about minus 6 or so

If it is imp scored pairs the detail is slightly different (more complicated), but the upshot is the same.

IMP scoring is much more like rubber bridge where scoring game contracts are worth more.

In rubber bridge scoring, the important scores are the scores associated with the exact contract you have bid. If the value of the contract you make is at least 100 points, you score a game toward rubber. If you can score two game contracts before your opponents do, then you win the rubber. When one side or the other accumulates 100+ points for the contract(s) made, they score a game and both sides start from scratch to try to win the next game.

Let's take your example. At one table, 4 ♠ was bid and made. The value of a 4 ♠ contract is (4*30)=120. Since it is worth more than 100, it means a game has been scored and with the next hand both sides would contest for the next game. At the other table, the contract was 3 ♠ making 4. The value of a 3 ♠ contract is (3*30)= 90. Since this is less than 100, it is scored toward game, but does not score a game. So that side has scored only a part score toward game. They would receive credit for the extra trick elsewhere on the score sheet, but it doesn't count toward game. (In rubber bridge, both sides would continue to contest for game with the next hand. If the side with the part score was able to bid and make any contract, then there cumulative total would over 100 and they would score the game. If instead, their opponents bid and made a game contract, they would score the game instead and both sides would start over towards the next game.)

The bonuses added in duplicate of +50 for part scores or +300 for non vulnerable games reflect the amount that would be received for them at rubber bridge in an unfinished rubber. The +500 bonus for a vulnerable game reflects the rubber bridge bonus for winning a 3 game rubber. At rubber bridge, if you win a rubber by scoring 2 consecutive games, you would receive +700 bonus. But, for whatever reason, duplicate awards just +500 for any vulnerable game.

So to recap --

3 ♠ making 4 reflects a value of (3*30)for the contract + 50 for the part score + 30 for the overtrick =170
4 ♠ making 4 reflects a value of (4*30) for the contract + 300 for making a non-vulnerable game= 420

IMP scoring goes one step further. It takes the scoring one step further to adjust for outlying results in team matches. I believe at one time team matches were scored on a total points basis. But that makes it possible for one or two huge results to decide a match virtually no matter what else happens. So at some point, the IMP scale was developed to dampen big total point swings have so that matches will be decided more on the skills of the players throughout the match than on one outlaying result.

The result of the example you used is a net of +250 or +6 IMP swing for the team that bid game.

[mycroft]

On BBO, for some reason, pair games, even "casual pair games", are almost universally IMP pair games. "For some reason" because the rare time we have an IMP pairs game face to face, the rank and file hate it. But "nobody" plays MP on BBO.

On BBO each hand is scored 32 (I believe) times (at most; sometimes there are hiccups that limit). So, were it a MP table, top on a board would be 31 (62 for yer Yeropean types) and we'd divide your MP score by 31x100% to get a percentage score. We use percentages partly because a random number is just that "what's average?" and partly because the number of playings on a board is not constant (either because it's still in play, or because it got hiccupped to end with less than 32 results).

But there aren't many of those - regular "random" games are IMP scored. Again, the hand is played a maximum of 32 times, and if you're N/S on the board, *every E/W pair that plays that board, except for your opponents, are your teammates*. So, for each other table, we take your score, add it to the E/W score at that table, and convert to IMPs. We add all the IMPs, positive and negative, up, and (because of comprehensibility and non-constant board plays) divide by the number of comparisons. That gets you the IMP score, positive or negative, you see in the results.

You may only be playing "a casual game at one table", but really , you're part of a massive, ongoing, randomly sectioned club game, and the BBO universe picks out 31 other tables to play against you on this hand, then dips back into the swirling mass to find 31 more for the next board, and so on.

So, you might be +170 on your board, which is nice, but if most of the other tables that played that one got to game and made it, you're going to be minus IMPs, even with the overtrick.

A small hint I usually give people new to IMP pair events: on one board, -2 to +2 is "average". Don't fret the small stuff. Law of Large Numbers means that over 30 boards, though, +2 IMPs/board, even +1 IMPs/board, is massive.

[msjennifer]

charles311, on 2017-January-26, 13:07, said:

Thanks to you both. I am aware about the scoring for teams. It was just a casual pairs game I was watching. That's what confused me.

In A Pairs IMP game,your score is compared to the average scored by other pairs sitting in your position Ew or NS as the case may be .While finding the average,depending upon the number of entries ,a certain number of top scores and an equal number of lowest scores are omitted.The sum of remaining scores divided by the number of Pairs ( excluding the number of pairs omitted).This average is then compared with each pair and the difference is converted to IMP as per the chart.So you may get a plus ,minus or zero IMP.

[Vampyr]

msjennifer, on 2017-January-27, 00:49, said:

In A Pairs IMP game,your score is compared to the average scored by other pairs sitting in your position Ew or NS as the case may be .While finding the average,depending upon the number of entries ,a certain number of top scores and an equal number of lowest scores are omitted.The sum of remaining scores divided by the number of Pairs ( excluding the number of pairs omitted).This average is then compared with each pair and the difference is converted to IMP as per the chart.So you may get a plus ,minus or zero IMP.

I think that cross-IMPs has replaced Butler almost everywhere. I am not sure about BBO, but there seems no reason for it to be different.

[msjennifer]

Vampyr, on 2017-January-27, 02:32, said:

I think that cross-IMPs has replaced Butler almost everywhere. I am not sure about BBO, but there seems no reason for it to be different.

Sir,thanks for your valued information.

[1eyedjack]

msjennifer, on 2017-January-27, 09:47, said:

Sir,thanks for your valued information.

My sarcasm filter is malfunctioning, but for the benefit of the doubt Vampyr is correct.

The incentive to compute an average datum against which to IMP the results stems from the days of manual scoring without computer aids.

But it distorts the result, because the logarithmic nature of the IMP scale results in a different weighting to results than with cross-IMP. The latter matches the intention of those who devised the IMP scale, where the former does not.

BBO sticks to the cross-IMP method.

[mycroft]

Also, for future reference, Vampyr is female. And wonderful. And, despite our frequent disagreements, a fount of useful information.

And, last I checked, I am not Marie of Roumania.