# Thread: The 1000th best performers

1.
Originally Posted by 26mi235
So you are going to compare the 1000th best 10,000 mark with the 1000th best 100m mark via some scoring table? Really? There are many dozens of opportunities for the sprinter's to get their best mark; the 10,000m guys may never get good conditions.

It is well known that the nth order statistic, especially where n is large, is not reliable in such a case. And so the asserted purpose of the statistics is not valid; other things are not held constant in obtaining the observations. Yes, that might seem like a technical statistics thing, but since we are talking about statistics, it goes to the core of a comparison.
I agree that you can't really compare the 10000 rankings with the 100m or most other events. Amongst other things, many athletes, for one reason or other, are only in world-class form for a year or two and may only get to run one or two track 10000s. If they get difficult conditions (e.g. heat/humidity, gusty winds) or are physically under-par due to something like a mild infection that is the chance of a good time gone, probably for quite some time. Also the 100m rankings are boosted to some degree by specialists from other events such as 200, 400, sprint hurdles or long jump. Fewer non-specialists will run a 10000 on the track.

2.
26mi365. Perhaps you should read my query.

It was about the 1,000th best performer, not performance.

I am well aware that the 1000th best performance could possibly be someone who is the 10th best performer in an event like the 100m, or the the 300th best performer in an event like the marathon.

Mine was a simple query. And while I take your word for the "n"th time, that your" n" statistical theories are well known, this issue does not need to be complicated with "n" this and "n" that.

3.
Same holds for performer and in fact that was the notion that was in my head. The reason frequency comes in is because there are so few opportunities and thus the observed times have a bias at the top for great conditions while at the bottom less conducive thus the gradient is steeper so that there is a selection bias across conditions both for performance and performer.

4.
We are each entitled to our own opinions.

I happen to think your logic is flawed, to say the least. I just don't buy into the statistical mumbo jumbo that you roll out all the time in an attempt to beat those who disagree with you into submission.

And obviously you think my logic is flawed.

C'est la vie.

5.
You can have an opinion but having an opinion on how statistics is done is not really a thing. Having an opinion on what meaning you take from it certainly is 'a thing'.

6.
Originally Posted by 26mi235
You can have an opinion but having an opinion on how statistics is done is not really a thing. Having an opinion on what meaning you take from it certainly is 'a thing'.
So when an opinion is not a thing, which part of speech is it?

7.
the problem is that 26mi is talking Statistics, while the vast majority of people here are talking statistics and there's a world of difference.

8.
You can over-analyze something ....this is one of those times....

9.
Originally Posted by gh
the problem is that 26mi is talking Statistics, while the vast majority of people here are talking statistics and there's a world of difference.
Some people explained how to manipulate statistics to get them sorted in a way that got to the question that started the thread. Then Statistics made an appearance and rather clumsily fogged up the picture.

Page 3 of 3 First 123

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•