Check it Out!    
RideCamp@endurance.net
[Date Prev] [Date Next] [Thread Prev] [Thread Next]
[Date Index] [Thread Index] [Author Index] [Subject Index]

Re: statistics



At 05:31 PM 6/7/00 -0700, terre wrote:
>From David LeBlanc:
>>I'm not sure that I can answer that directly, but for the rides I was
>comparing, here's some numbers - for LD rides, the median (1/2 were slower,
>1/2 were faster) _winning_ ride speed was 8.47 MPH.  For endurance rides of
>less than 100 miles, the median winning ride speed was 9.42 MPH.  If you
>can't manage 4.6 MPH, then you don't complete, so that's too slow.
>
>This is very interesting stuff--thank you for doing the work!  	The
>question is--what does it MEAN?  That's the problem with statistics; the
>need to interpret the data.

True.

>	For example (and I'm just speculating here) an analysis based on winning
>times could be skewed by one or two exceptionally fast horses winning all
>of the (endurance) rides in the database.  

True, but over hundreds of rides from all regions over more than one year,
that seems unlikely.  Unfortunately, a lot of times the data doesn't
contain ride times for anything other than the winning rider (esp. true
with LD), so the only data we have are winning times.

>We tend to assume that the LDers
>are 1)newbies riding 'entry level'--ie young or not-really-cut-out-for
>endurance horses, or 2)physically compromised riders or 3)pleasure riders
>(or some variation of the above).  

I think this may be a bad assumption. I know personally one rider who is
doing LD this year because he doesn't have time to train, and he's one of
the best in this area - he normally rides endurance.  There are a number of
other scenarios - you could have experienced LD riders with many seasons
under their belt (e.g., my wife prior to this season), you could have a
very experienced rider bringing up a new horse - there's a lot of different
reasons to be doing LD other than the above.  I would also speculate that
you are going to find very few people fitting the above descriptions
placing 1st in LD rides.  It may be that new riders tend to concentrate in
LD, but it is also probably true that top-10 level LD riders may have
different characteristics as a population than the overall population of LD
riders.

>While this is certainly not universally
>true, if it were true in this region, or in the rides analysed, it would
>tend to account for the relative slowness of the LD winners, given that the
>'endurance' winners can be assumed to be 1)experienced riders riding
>mature, conditioned horses selected for the sport 2)healthy or masochistic
>enough to ride that distance, as opposed to shorter and 3)competitive (or
>they wouldn't have won).

No, I took rides from 3 years worth of AERC data, and have enough rides
from all of the regions to be statistically significant.  There are
variations among the regions, but the same trend shows in every region but
one (CT).  It would take further study, using data that isn't available
right now to determine the very interesting question of _why_ most winning
LD rides are run at a slower overall pace than winning 50+ mile rides (note
that 100's were not counted - this population behaves differently than
either LD or rides of 50 up to 100 miles in many ways).

>	Also, as we've been commenting, the perception of speed is
>relative--'racing' or 'too fast' can depend on the individual case.  A
>horse is ridden too fast (in any distance) if it is ridden beyond it's
>capabilities.  I have seen 'overridden' horses come in dead last (not just
>in LD)--they were not fit enough for the work required, and speed had
>nothing to do with it....

This is very true. I witnessed a tie-up happen on an out of shape horse
with an out of shape rider that was well back in the pack of an LD ride.  I
think this is a better concern than whether someone is racing or not at any
given distance.  If I try to set a faster pace over some given distance
than my horse can handle, she's going to get hurt.  This applies to all
riders over all distances, and my personal experience is that each new
level of distance presents a new challenge.  I think it matters a lot less
whether the animal is running top 10 than whether it is running faster than
it ought to be.  As you point out, faster than it ought to be could be in
the middle of the pack.

>	Did you crunch any numbers for "middle of the pack" speeds, or even
>"tail-end" speeds?  Don't know what they would mean either--just wondering!

No, I'd really like to.  Something I want to do is get a copy of the whole
database - that would allow some really interesting queries - such as, what
charcteristics in terms of placing do horses with really high lifetime
milage have?

> Did you look at 100s?

No, not yet - 100's are really different events - for example, completion
rate is a reasonably level function of ride distance, declining from about
86% for LD to around 80% for 75 mile rides, with 50's in-between.  This is
about what we'd expect.  But 100 mile riders are only about 60% likely to
complete overall.  The rider population is also very small - only about 8%
of all ride-rider combinations are people doing 100's.



>Interesting stuff!  Thanks David.
>
>terre
>
David LeBlanc
dleblanc@mindspring.com



    Check it Out!    

Home    Events    Groups    Rider Directory    Market    RideCamp    Stuff

Back to TOC