[myspauyijbv]

As I see it, these are basically what are called p-values: probabilities that the result could be achieved in true randomness. With a 100% p-value, there is a 100% chance that the draw would be perfectly random. Usually in statistics, a p-value of something less than 5% is needed for the result to be significant. Since this study had one of .3% for the men and 0% for the women, it'd be significant. Hope that makes sense.

If you're right, then their wording is horrible: "How frequently ESPN's simulated draws came up with average difficulty scores that were at least as low as scores for the actual Grand Slam draws. Percentages closer to 0 indicate a lower likelihood that the actual results are strictly due to random chance."

I would expect about half to be "at least as low" and half to be higher. No?

[Les Allen]

They appear to like Serena (faced 126 and 99) more than Maria (faced 60 and 47).

[77chawzence]

I suppose it is worth looking into but I don't find the evidence particularly persuasive. Just because something is unlikely to happen doesn't mean that it will or did happen. It seems to me that they only look at the first two seeds over 10 years worth of tournaments which makes for a very small sampling. Unless I missed it (I read the information last night so I perhaps could have forgotten) it doesn't take into consideration at all who else was in their quarter; who they faced in the second round etc. Additionally rankings are not necessarily the most honest reflection of the desirability of facing a person in the draw -- I think there are a lot of people who would rather face Woz in the first round of the US Open than Serena regardless of their rankings.

The draw isn't an experiment that is looking to be replicated they get the results once and they go with it. The odds of winning a lottery are astronomically small as well (and why I don't play) but to the person who wins I'm sure they don't care about how unlikely it was-- they are going to cash the checks anyway.

I think that it is good that the USTA is willing to look into this but as Woody is pointing out we seem to be doing a lot of guessing about what exactly the information they have released means. I've read the extended "how it was done" portion but I don't think it directly explains all of the information and figures they have provided. I'm sure their results are valid I'm just not sure that it demonstrates anything of importance other than that it is worth looking into. But I am certainly not the most statistically inclined TATuer...

[occafeVes]

To properly discuss this one needs to understand statistics really well. The analysis itself is not a believe or not believe matter any more than astrophysics or thermodynamics. My rather limited understanding says that the analysis is sound (and thus the draws have been rigged to some extent), but I could easily be wrong...

Do we have any professionals here willing to invest a bit of time for proper analysis?

[occafeVes]

I suppose it is worth looking into but I don't find the evidence particularly persuasive. Just because something is unlikely to happen doesn't mean that it will or did happen. It seems to me that they only look at the first two seeds over 10 years worth of tournaments which makes for a very small sampling. Unless I missed it (I read the information last night so I perhaps could have forgotten) it doesn't take into consideration at all who else was in their quarter; who they faced in the second round etc. Additionally rankings are not necessarily the most honest reflection of the desirability of facing a person in the draw -- I think there are a lot of people who would rather face Woz in the first round of the US Open than Serena regardless of their rankings.

That's all true, but I am afraid completely irrelevant for this analysis.

The draw isn't an experiment that is looking to be replicated they get the results once and they go with it. The odds of winning a lottery are astronomically small as well (and why I don't play) but to the person who wins I'm sure they don't care about how unlikely it was-- they are going to cash the checks anyway.

Odds of somebody winning a lottery is 100% or close.

[Romobencience]

"Davenport - Li (37)"? I knew those bitches hated Lindsay.

For real, though, I'm with the person in the article (I forget who) who expressed the idea, "You seriously think the U.S. Open would risk getting caught for something this big just to give its top seeds a 99.5% chance of winning their first round rather than the usual 99% chance?" I'm not buying it.

[Les Allen]

I doubt they sat down and said we are only going to look at seeds 1 and 2. I'm sure they looked at other high seeds. Does using the top 4 seeds bring it within range?

[Vomephems]

To properly discuss this one needs to understand statistics really well. The analysis itself is not a believe or not believe matter any more than astrophysics or thermodynamics. My rather limited understanding says that the analysis is sound (and thus the draws have been rigged to some extent), but I could easily be wrong...

Do we have any professionals here willing to invest a bit of time for proper analysis?

Does getting an A+ in the one stats course I did make me an expert? Anyways, I agree that this is way to small a sampling to be statistically relevant. In the course I did 30 was the bare minimum sample size needed to even start considering results - and even that would have a 90 or 95% at best chance of being relevant at best. I didn't read too much into this myself.

I'm sure they could find other random anomlies...ex. maybe at the French Open 7 & 8 have had crazy hard draws. Actually...they do mention that at the French the #1 and #2 women had significantly harder draws - without anomalies like this it actually wouldn't really be random anyways.

[hapasaparaz]

"Davenport - Li (37)"? I knew those bitches hated Lindsay.

For real, though, I'm with the person in the article (I forget who) who expressed the idea, "You seriously think the U.S. Open would risk getting caught for something this big just to give its top seeds a 99.5% chance of winning their first round rather than the usual 99% chance?" I'm not buying it.

Doesn't have to be a conspiracy, just that there is something wrong with the program being used. Though the report does say that the AO uses the same company: Earley said he would consult with representatives of Information & Display Systems, the company that provides the software that generates the random draw. IDS has been providing the random draw software for the U.S. Open for more than 10 years, and does the draw for the Australian Open. Which is strange as there is such a big difference between the USO and AO results. But we don't know if they use the exact same program, and if it has been the same for every year in the last 10.

I think their explanation of the sample size question (and what Togtdyalttai wrote about this on pg 1) makes sense. Not that this is my area of expertise! I have no expertise

[slima]

Do we have any professionals here willing to invest a bit of time for proper analysis?

Yeah, I'm totally reserving judgment until Dr. rabbit hath spoken.