[Vampyr]

FrancesHinden, on 2012-March-12, 15:42, said:

You mean, just like the difference between scoring 60% at matchpointed pairs and winning and scoring 60% and coming in 8th? The difference still escapes me.

I think that the size of the field and the variance of the scores are relevant in both IMPs and MPs.

[awm]

Vampyr, on 2012-March-12, 20:22, said:

I think that the size of the field and the variance of the scores are relevant in both IMPs and MPs.

Well the size of the field certainly effects your chances of winning, but I'm not sure it has much effect on your expected percentage. I mean, if I expect to score 55% on average in a small game, I'd expect to score the same in a larger game of the same average caliber right? Maybe variance goes down.

The variance might be an issue in the short term, but again if you're trying to measure my "average score" it should even out over enough sessions (which obviously takes a while).

[phil_20686]

awm, on 2012-March-12, 20:48, said:

Well the size of the field certainly effects your chances of winning, but I'm not sure it has much effect on your expected percentage. I mean, if I expect to score 55% on average in a small game, I'd expect to score the same in a larger game of the same average caliber right? Maybe variance goes down.

The variance might be an issue in the short term, but again if you're trying to measure my "average score" it should even out over enough sessions (which obviously takes a while).

I think the point Stefanie was trying to make, is that the variance on a set of hands scored for MP is probably not as large as the variance of a bunch of IMP hands.

E.g. If I deal two sets of boards, the first set all partscores, and the second set all games and slams. Then I might be able to rack in a huge "equivalent MP" score on the second, while being unable to on the first, while I would have the same score had I scored them at MP.

There are lots of things you could in theory do for this, like scale the imp scores onto the average normal distribution for a MP score, but it would probably be a lot of effort for a minimal game, since there are not all that many imp pairs sessions anyway.

[mchristie]

Just to say that, due to a display quirk, (well, ok, a bug) it is somewhat random whether you will see your NGS result for any multi-section pairs event, but it will have been scored, and if so it will affect your grade. Hopefully the display will be corrected later this month.
Various IMP scored results, especially in Oxfordshire, have not been graded due to a bug in the reporting of these results by the scoring system.
IMP results are compared to MP results and the conversion factor is essentially the ratio of the observed variances of IMP and MP scores for fields of equal players. (That doesn't read very clearly, sorry.) I'd like to use our large database of both MP and IMP events to refine these observations. Some day, year...

[Vampyr]

phil_20686, on 2012-March-13, 11:28, said:

I think the point Stefanie was trying to make, is that the variance on a set of hands scored for MP is probably not as large as the variance of a bunch of IMP hands.

Yes; I wish I had expressed it half as well as in this post.

Quote

it would probably be a lot of effort for a minimal game, since there are not all that many imp pairs sessions anyway.

But for a lot of people IMP pairs represents a large majority of the club bridge, or (EBU affiliated) bridge period, that they play. Even though it is only once a week.

Also, there seems to be a lot of "this effect is small, so we will ignore it" and "we can't do anything about that effect, but it only affects about 10% of the people" etc. This all adds up.

[Blue Uriah]

Vampyr, on 2012-March-13, 18:05, said:

Yes; I wish I had expressed it half as well as in this post.


But for a lot of people IMP pairs represents a large majority of the club bridge, or (EBU affiliated) bridge period, that they play. Even though it is only once a week.

Also, there seems to be a lot of "this effect is small, so we will ignore it" and "we can't do anything about that effect, but it only affects about 10% of the people" etc. This all adds up.

Since Christmas there have been over 10,000 club sessions submitted to the EBU and 0.7% of those have been Butler Pairs, 0.5% Cross-IMPs. So I'd be very surprised if there were many people for whom this represented a "large majority". Still, to suggest that the EBU doesn't care about this tiny group is a little unfair. The conversion tables have been put together by some very clever people over years of study and trial and, personally, I'm at least willing to give it a chance before dismissing it so readily.

[mchristie]

At both MPs and IMPs some boards are more variable (wilder) than others which are flatter, and there is less opportunity to score well/badly on the flatter boards. We know that there are comparatively few extremely significant boards at IMPs whereas at MPs it is the extremely flat boards that are less common. What we compare when converting IMP scores to MP% equivalents is the average variability of IMP boards and the average variability of MP boards, based on actual data from a large set of boards, so that we CAN say the average variability of a bunch of MP boards is equal to the average variability of a bunch of IMP boards.
Stephanie, I think this means we treat IMP pairs events fairly. It certainly isn't meant to be in the group of small effects that we ignore.

[ahydra]

Wow I'm a King! (59%)

Doubt it'll last though.

ahydra

[billw55]

I think it would be an interesting experiment to send a large number of predealt boards to matchpoint games in one region, and IMP pair games in another, and then analyze the results.

[IanPayn]

FrancesHinden, on 2012-March-12, 15:42, said:

You mean, just like the difference between scoring 60% at matchpointed pairs and winning and scoring 60% and coming in 8th? The difference still escapes me.

++++In the first instance you feel fate has smiled upon you. In the second instance you feel hard-done by. It all comes back to feelings.

[JLOGIC]

Gnasher up to 63 makes the ratings seem way more legit somehow lol bbfanboism.

Anyways, I think this is a great idea, I hope it works and USA implements it. Ratings make people play more, not less.

[billw55]

JLOGIC, on 2012-March-14, 12:49, said:

Anyways, I think this is a great idea, I hope it works and USA implements it. Ratings make people play more, not less.

I wonder. I think most likely you are right. But I also think that there will be a nontrivial minority who begin engaging in rating protection - various metatactics designed to maintain an otherwise unsustainably high rating. In some cases this would probably include playing less, as well as careful selection of opponents, partners, events, etc.

The upside of nondecreasing masterpoints is that there is never anything to lose by playing.

[gnasher]

A friend of mine has just asked me for a game. In round numbers, I am rated 63 and he is 70. The average field at the Young Chelsea is about 55. Does that mean we have to score 61.5% (or +43 IMPs) just to break even? Help!

[WellSpyder]

gnasher, on 2012-March-15, 04:58, said:

A friend of mine has just asked me for a game. In round numbers, I am rated 63 and he is 70. The average field at the Young Chealsea is about 55. Does that mean we have to score 61.5% (or +43 IMPs) just to break even? Help!

Yes, it does. Does that mean you won't agree to play? I doubt it! In any case, once the grades have settled down you will probably find that around half your sessions fail to break even in the sense you have used it above.

[gnasher]

We decided that the best thing to do was to play on Friday 13th.

[billw55]

gnasher, on 2012-March-15, 04:58, said:

A friend of mine has just asked me for a game. In round numbers, I am rated 63 and he is 70. The average field at the Young Chealsea is about 55. Does that mean we have to score 61.5% (or +43 IMPs) just to break even? Help!

This is the sort of thing I was talking about. Some players will look at such a situation and choose not to play, to protect their rating. Not you, of course, and probably not anyone who posts here. But some.

I wonder if a rating of 70 is sustainable at all, by anyone.

[gordontd]

billw55, on 2012-March-15, 06:22, said:

I wonder if a rating of 70 is sustainable at all, by anyone.

All but one of those who have a rating of 70+ on the EBU's NGS are "evolving" - ie they haven't played the 1000 boards to provide a stable rating, so I imagine your are right.

I did know someone on OKbridge who had a Lehman rating of 70 & rising. I used to play with him when my rating was low, but as it improved he stopped asking me, and seemed to play less and less although he was usually online. He did however eventually achieve the maximum rating of 75. I think most people would rather play bridge though.

[phil_20686]

gordontd, on 2012-March-15, 06:37, said:

All but one of those who have a rating of 70+ on the EBU's NGS are "evolving" - ie they haven't played the 1000 boards to provide a stable rating, so I imagine your are right.

I did know someone on OKbridge who had a Lehman rating of 70 & rising. I used to play with him when my rating was low, but as it improved he stopped asking me, and seemed to play less and less although he was usually online. He did however eventually achieve the maximum rating of 75. I think most people would rather play bridge though.

On Ok bridge didnt people do this to counteract grade inflation? I never played but I heard that some players considered it a public duty to build up their ranking then create a new login.

[Vampyr]

mchristie, on 2012-March-14, 05:04, said:

we CAN say the average variability of a bunch of MP boards is equal to the average variability of a bunch of IMP boards.
Stephanie, I think this means we treat IMP pairs events fairly. It certainly isn't meant to be in the group of small effects that we ignore.

Is the average variability of a set of boards more accurate than the variability of the set of boards in question?

[Vampyr]

billw55, on 2012-March-15, 06:22, said:

Some players will look at such a situation and choose not to play, to protect their rating.

I worry particularly about people showing up without a partner. In many clubs there are normally a fair few such people.

When I wrote a couple of years ago, I suggested letting any number of pairs opt out of their game being rated, rather than just the actual host. This suggestion was ignored, which I think is a shame.