Endurance.Net Ridecamp Archives

terre wrote:
> Truman said:
>
> > I think you guys are involved in more of an academic exercise
> > than anything else.
>
> Well, of course we are! What else is ridecamp for? You don't think 
> we're accomplishing anything substantive, do you?
>
> It is 'academically' interesting, though, to discuss the different 
> factors that can influence the data. This sport is SO complex, 
> involving so many variables, that to draw any hard-and-fast 
> conclusions from any of the data we gather is something of a reach.
Actually it's not deriving conclusions from the data, 50's are riding 
faster. What is dicey is providing an interpretation to what that data 
says. That's where you always get into trouble. It's not what the data 
says but why it is that way. I prefer to leave interpretation to the 
philosophers and theologians ;-) .
> Ride times are influenced by a host of factors--weather and terrain, 
> etc as discussed. But there are other things to consider. Was the 
> winning time an "anomaly" in the sense that the winner was WAY ahead 
> of the next horse? Having a 'superstar' winning a number of races with 
> the rest of the field far behind can skew the apparent data, if you 
> look only at the 'winning time'. Perhaps it would be more valuable to 
> look at the median or mean times. Got a couple of 100 hours to do that?
Actually when I was looking at the data for 2002 I did. There was no 
apparent skewing in the data. The distribution is not normal (bell) it 
tends to be a skewed bell curve (mode, median nor mean are the same) 
since very few riders take a prolonged time. Skew normal distributions 
appear in a lot of application. The 50's and LD's appeared to have the 
same shape but I didn't calculate the skew parameter.
> David mentioned pull rates/times with respect to season, and that's a 
> valuable observation. It's possible there are somewhat higher pulls 
> later in the year, as the horses start to tire or develop problems--I 
> think there are certainly fewer STARTS late in the season as people 
> decide "enough is enough" and stay home with horses who are starting 
> to show problems--as well they should! That could slow down winning 
> times, as well.
It also depends on the region. Late in your season is the start of ours. 
The beauty of statistical analysis is by it's very nature it deals well 
with extracting trends from 'noisy.' Such things as hot weather one 
ride, cold another are part of the noise process. They are independent 
of the endurance rides and tend to behave in a well understood manner 
(from a probabilistic sense). It is concerned at analyzing the data at 
the probabilistic level. As long as there are not deterministic biasing 
given enough data the trends will emerge.
> The variables are almost infinite!
They may be large but they are not infinite.

Truman

--

“Ethical axioms are found and tested not very differently from the 
axioms of science. Truth is what stands the test of experience.” Albert 
Einstein


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

Ridecamp is a service of Endurance Net, http://www.endurance.net.
Information, Policy, Disclaimer: http://www.endurance.net/Ridecamp
Subscribe/Unsubscribe http://www.endurance.net/ridecamp/logon.asp

Ride Long and Ride Safe!!

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=