BBO Discussion Forums: MP vs IMPs - BBO Discussion Forums

 nige1, on 2014-December-20, 17:26, said:

Yes, they would. But there aren't too many of them.

IMP-Pairs is arguably the most popular form of online bridge play -- compare the number of IMPs vs MP tables in the BBO MBC. That's probably the most telling evidence that it's an easier, or less stressful, form of the game. If you're playing with a random, you're less likely to get called an "idjit" for dropping an overtrick.

cherdano,

Maybe the link I posted was too tough for you to understand. Try this
one from the Khan Academy.
https://www.khanacad...sample-variance

Helene thinks double dummy analysis is an exact science. It isn't.
She thinks confidence intervals look like this ||. In bridge they
really look like this |______________|. In 7 board swiss matches
flight C teams occasionally get lucky and beat world class teams.
90% slams go down 10% of the time.
Feel free to delete your inane post.

 jdeegan, on 2014-December-19, 03:40, said:

That may have been true in the past. Today with Salman Khan teaching
statistics for free in his Khan Academy, a bright high school can solve
bridge hand evaluation in little pieces at a time. None of the discrete
bridge models require knowledge of calculus.
The key in using hypothesis testing is choosing the appropriate side
conditions. When testing opening 1S vs 1NT every data entry must
have a 5332 pattern. Do not include the 5422 pattern.

Quote

Double dummy analysis is always biased.

Sigh. This study has nothing to do with DD analysis. I just evaluated two different scoring methods for mitchell movements of different size. You could apply it to figure skating or jeopardy etc, it is not specific to bridge.

And this talk about "bias". LOL. I just simulated some normal distributed data. If there is any bias in those data it can only be because there is a bug in the random number generator I used. I think jdeegan just thought he could impress someone by demonstrating that he can spell the word "BIAS", regardless of whether he can use it in a context where it makes sense.

Finally, jogs, you don't need to teach Arend and me high school statistics.

helene, let's put it this way: you can't teach "standard deviation" to a person... the person is either born with it built-in.. or not

 jogs, on 2014-December-21, 12:26, said:

 whereagles, on 2014-December-21, 19:40, said:

helene, let's put it this way: you can't teach "standard deviation" to a person... the person is either born with it built-in.. or not

Correcting our own or others' mistakes is fine but there's no need to impugn each others' abilities. I believe that those who are interested can understand the concept of standard-deviation, with the help of Jogs' link.
Khan's Descriptive Statistics Course

 nige1, on 2014-December-20, 15:01, said:

You mean generating an mp score based on all tables involving non overlapping pairs being voided? Yes that is probably a good idea

 nige1, on 2014-December-22, 06:05, said:

Probably, but I would doubt that just anyone can understand algebraic geometry with the help of cherdano's published papers on the subject. (Algebraic geometry makes my head hurt.)

 helene_t, on 2014-December-22, 06:14, said:

No, I wish I'd meant that. Helene_t's suggestion is better. Perhaps, you could also follow her protocol with the cross-imp tournaments.

 nige1, on 2014-December-22, 06:51, said:

What about using cross-IMP scores from a teams tournament?

 nige1, on 2014-December-22, 06:51, said:

Yes, you would have to follow the same protocol for XIMPs and for MPs since the purpose of the study is to make a fair comparison between the two.

 nige1, on 2014-December-22, 06:05, said:

I just wanted to say it's far easier for those with an analytic mind to understand the concept. The rest is rhetoric

The point was also that some people simply refuse to understand the concept. I see this happening over and over again: "Why were my two finesses wrong? At least one should have been right. The deck was fixed. Damn these computer hands!!!!"

 mgoetze, on 2014-December-22, 06:15, said:

You only need a complete understanding of algebra. No need to know geometry.

 jogs, on 2014-December-22, 09:34, said:

Q.F.T. I don't think there is anyone on this planet with a complete understanding of algebra.

I didn't attempt to read Arend's thesis so I can't say that you are wrong. I suspect that you don't know what algebraic geometry is, though. (Hint: it doesn't have much to do with geometry except in a very abstract sense).

 helene_t, on 2014-December-22, 10:09, said:

Go to your profile.

Click on "manage ignored users."

Adjust as appropriate.

 helene_t, on 2014-December-22, 10:09, said:

As half an algebraic geometer, I'd like to dispute this! If you don't think algebraic geometry has much to do with geometry, you learned it from the wrong perspective.

(Okay - maybe the Bridgebase forums are the wrong place for mathematicians to be fighting over mathematics. )

 helene_t, on 2014-December-22, 10:09, said:

You're right, only need partial understanding of algebra. The terms never shift from the right side
of the '= sign' to the left. It helps to understand Excel. The functions are in Excel.
Enter your set of data into column A. Click the function you want.

fx

For mean click average.
For variable click var.
For standard deviation click SQRT for square root of variance.

 jogs, on 2014-December-22, 17:45, said:

While watching you make an idiot of yourself *IS* funny, you know she literally has a PhD for her work in this area? Right?

 Cthulhu D, on 2014-December-22, 18:26, said:

You mean knowledge of Excel functions doesn't give me a complete understanding of maths?!!?

You will be telling me next that I cannot get a complete understanding of bridge by using a pseudo-mathematical equation and some big words...

FWIW I ran the simulations again but with the per-board skillfactor proportional to the noise level. As discussed earlier, I would expect this to favour matchpoints. I simply multiplied the skill factor with the noise - some rescaling of the parameters in order to give similar correlations as before may be appropriate, as you can see the correlations are much weaker now:

The average Spearman correlations between strength and IMP scoring was (as a function of shape parameter of the skill factor distribution and number of tables):

  .1;3  .1;9   .1;27  1;3    1;9    1;27   10;3   10;9   10;27 
  0.316 0.376  0.394 0.503 0.584   0.617   0.562 0.640   0.675
 

For MPs:

  .1;3  .1;9   .1;27  1;3    1;9    1;27   10;3   10;9   10;27 
 0.348 0.423   0.452 0.506 0.610   0.651   0.543 0.645   0.690

As before, maatchpoints are favored by large fields and imps by small fields when the board population is homogenous. This is what you would expect since when the board population is homogenous it doesn't matter much whether the noise and skill factor are correlated or not since both are roughly constant. But for heterogenous sets of boards (shape parameter for the skill factor equal to 0.1 or 1), matchpoints are generally favoured. Again, this is what one would expect.

Obviously the two scenaria - noise proportional to skill factor, and noise independent of skill factor - are two extremes, and a mor realistic scenario would be somewhere between the two. The again, we really need some real data.