Giro stage 1 ITT

Sorry for the delay. Took me a while to find the intermediate split times.

http://anonymous.coward.free.fr/rbr/giro10-1.png

Fred on a stick wrote:
Sorry for the delay. Took me a while to find the intermediate split
times.
http://anonymous.coward.free.fr/rbr/giro10-1.png

BTW, there's some evidence that conditions changed during the afternoon.
Here's a comparison of how the starting order affected the speeds during the
first and second "halves" of the ITT:

http://anonymous.coward.free.fr/rbr/giro10-1-2.png

Fred on a stick wrote:
BTW, there's some evidence that conditions changed during the afternoon.
Here's a comparison of how the starting order affected the speeds during the
first and second "halves" of the ITT:

http://anonymous.coward.free.fr/rbr/giro10-1-2.png

So what you are saying is that the increase in speed is more than what
was to be expected from putting favourites later in the line-up?

A. Dumas Fred wrote:
http://anonymous.coward.free.fr/rbr/giro10-1-2.png

So what you are saying is

Catchy phrase, eh?

Fred on a stick wrote:
A. Dumas Fred wrote:
http://anonymous.coward.free.fr/rbr/giro10-1-2.png
So what you are saying is

Catchy phrase, eh?

Definitely. So, any merit to the favourites-starting-later proposition?

A. Dumas Fred wrote:
http://anonymous.coward.free.fr/rbr/giro10-1-2.png

So, any merit to the favourites-starting-later
proposition?

What was interesting was that the slopes for the two parts (outbound and
return) were so different. Sure, the favorites started later but you'd
expect the fast guys to be fast on both on the first 4.1km and the last
4.3km. What we see is that the relationship between the outbound and inbound
speeds changed over the afternoon, and it changed in a pretty consistent
way.

Here's another view of the same data:
http://anonymous.coward.free.fr/rbr/giro10-1-3.png

Fred on a stick wrote:
A. Dumas Fred wrote:
http://anonymous.coward.free.fr/rbr/giro10-1-2.png
So, any merit to the favourites-starting-later
proposition?

What was interesting was that the slopes for the two parts (outbound and
return) were so different. Sure, the favorites started later but you'd
expect the fast guys to be fast on both on the first 4.1km and the last
4.3km. What we see is that the relationship between the outbound and inbound
speeds changed over the afternoon, and it changed in a pretty consistent
way.

Here's another view of the same data:
http://anonymous.coward.free.fr/rbr/giro10-1-3.png

I don't see how that could be explained by changing conditions. I think
that the first part of the course was less technical and the stronger
riders went "more faster" on part 1 than on part 2.

A. Dumas Fred wrote:

I don't see how that could be explained by changing conditions. I
think that the first part of the course was less technical and the
stronger riders went "more faster" on part 1 than on part 2.

That's one possible explanation -- but another is that the roads dried out
and it differentially affected the speed on the two parts. Besides, that
third plot shows a "local regression" curve and it stabilized and then fell
for the last 25 riders or so.