A statistical approach to Usain Bolt's times.

obnoxious

obnoxious

You are wrote:

obnoxious

NOW I'm obnoxious! You are still and ass bag!

I will necessarily pile on to say that the a-hole here, sir, is you.

You may find life more enjoyable if you let go of the idea of having to be right all the time. Oh, and don't dismiss the joy to be found in good spelling either.

piler onner wrote:

I will necessarily pile on to say that the a-hole here, sir, is you.

You may find life more enjoyable if you let go of the idea of having to be right all the time. Oh, and don't dismiss the joy to be found in good spelling either.

Yes, uh...Sir, I'm being an asshole for a purpose. You may find life more enjoyable if you don't take the time to point out flaws without understanding. Regardless of what they may be.

Enjoy your anally retentive spelling theory.

dude, let it go. go for a run or something.

Woah Woah woah!

Please, this tread was for serious statistical analysis. Not petty arguments. You all are like those people a few months ago who really got into it on that one thread.

Do it backwards. How fast can bekele run the 100, or komen, or geb?

It's been a while, but wouldn't you rather use calculus to figure out how the curve changes given different datapoints and extrapolate. It should be obvious to you if you are a runner that your formula is too simplistic. I seem to remember there was a russian guy in the 70's that ran from the 200m to the marathon at with excellent times his entire career as he moved up. He'd be a good person to use for multiple datapoints. Also you have to have a bit of reality here. Usain isn't moving up to the distances ever. No money in it.

Hey now, Blind Eye's post was fine. I'm sure the same poster has been posting under different names to "make his point" against him. Weird, but then again skinny, frustrated runners like to take their rage online. Yippee!

for distances 800m and longer, runners' pace at various distances can be approximated reasonably well by a power law, meaning a formula of the form

time = c*distance^p

with "c" and "p" constants. All runners will have p>1, meaning they slow down as the distance gets longer. The closer a runner has to p=1, the less they slow down. The lower the constant "c", the faster the runner.

This works decently for 400m as well, but it messes up significantly on 100m and 200m. Evidently the slowdown is different at those distances. You can use it to do thing like predict Geb's 800m time or Tegenkamp's 10k time.

A while ago I built a running calculator online that actually does this, so it could predict your time at some distance using your time at two or more other distances (but it wouldn't work if you only put in one distance, because you need at least two data points since the model has two free parameters). The server I was hosting it on went down and I don't have the old code, but if a few people are interested, I'll build a new one. It's not too complicated, and would take about one day, but it will have to wait a month or so until I finish the school year.

Yes, Build it!

Does anybody know, if there have ever been any scientific human speed and acceleration studies done, not using competition as the stimulus ? As in, top running speed and acceleration in "lab" conditions?

As in, lets NOT get the 10m splits for Bolt AFTER his race but lets try to create different conditions for him to run as fast as possible...

I suppose the obvious answer is that competition is the best known, ideal condition to measure top values for speed and acceleration. But is it?

Buster Keaton movies come to mind...

At the risk of stating the obvious, the more fast twitch fibers an athlete has, the more they will slow down as the race distance increases. My guess is the Bolt is very close to 100% FT fibers, so he is going to slow down a LOT as the distance increases. (If you watched the last 50 meters of his 19.30 WR, the effort was obvious on his face, and at that point he had only been running for 14 seconds.) With sufficient training, he might be able to break the world record in the 400m. The 800m WR is probably beyond his fiber capacity, and everything longer than that almost certainly is.

Sean Nunn

Raytown South

On a related note, I'm going to win the lottery tomorrow. I bought a lottery ticket. I could either win or lose. That means I have a 50% chance of winning. I then bought another lottery ticket, adding a second 50% chance of winning:

50% + 50% = 100%

Tomorrow I will be a millionaire. Later suckers.

Sure, you could build that--but can you determine the precise rate at which the difficulty of running a sub-5:00 mile increases throughout a 24-hour relay for a 4:22 miler wearing tights and being distracted once per lap by women's track-team members' comments about his ass?

Arc Icon wrote:

Sure, you could build that--but can you determine the precise rate at which the difficulty of running a sub-5:00 mile increases throughout a 24-hour relay for a 4:22 miler wearing tights and being distracted once per lap by women's track-team members' comments about his ass?

Are the women also wearing tights?

mathamatician wrote:

Ok, bolt can run a 19.30 200m.

Michael Johnson ran a 19.32 and a 43.18 400.

By pure mathematics Bolt should run 43.14.

Coe ran a 46.87 400m and a 1:41.73 800m.

By pure mathematics Bolt should run 1:34.27.

El Guerrouj ran 1:42.7 800m and 3:26.00 for the 1500m

By pure mathematics Bolt should run 3:09.14

Bekele has run 3:32.35 for the 1500m and 26:17.53 for the 10000m. By pure mathematics Bolt should run 23:43 (rounded).

There you have it, done purely mathematically. What do you all think?

Usain Bolt gives tips to Manchester United Football Club.

this is just for a laugh, but if we assume bolt couda gone 19.20 ( his 19.30 was into a wind & he belted the curve too fast & paid for it in stretch despite breaking WR ) & say simultaneous 43.0 - 43.5, -> for 800m, 1k,1500m, 1 mile, 2k, 3k, 5k,10k, 1/2M & M

~ 1'35.2 - 1'37.2 ,

2'02.7 - 2'05.60

3'14.2 - 3'19.6

3'30.2 - 3'36.3

4'28.4 - 4'36.8

7'02.8 - 7'17.5

12'27.1 - 12'56.1

26'49.1 - 27'59.6

1"00'56 - 1"03'54

2"09'58 - 2"16'45

interestingly, by about 5k race distance, the WRs have dried up

Sean Nunn wrote:

At the risk of stating the obvious...

Don't worry - Blind Eye already has that one locked down.