| Changes in stage 17 compared to stage 16
| Favourites by time :
| +0:49 Carlos Sastre
| -1:14 Samuel Sanchez
| -1:14 Andy Schleck
| -1:24 Frank Schleck
| -1:24 Alejandro Valverde
| -1:26 Vladimir Efimkin
| -1:26 Denis Menchov
| -1:26 Christian Vande Velde
| -1:26 Cadel Evans
| -4:43 Damiano Cunego
| -5:12 Kim Kirchen
| -23:29 George Hincapie
| -27:13 Jens Voigt
Just straighten Old Davey out will you Bob?
How do you compute "Favourites Gain by time" for the GC leader when
that individual was not GC leader the previous stage?
> Biggest gainers by time :
> +0:49 Carlos Sastre
You might consider changing the code in your bot a little. As you
know, Sastre gained more than 0:49. Your code is comparing him to
himself today (yellow) vs. himself yesterday instead of himself today
(yellow) and Schleck yesterday (yellow). MIsleading.
Quite simple : this morning he was 4th at 49 secs, tonight he is 1st at 0.00
therefore he has gained 49 seconds.
See my answer to Davey.
I think there might be one other algorithm. How about showing this time
gain as nth-to-1st PLUS the gap to 2nd?
I think that shows, as best as possible, the true time advantage as best
as possible. In your formula, it doesn't seem to matter if Sastre
finished 1 second or 1 hour ahead of the second place rider. In my
formula, that's accounted for.
Showing the time gain relative to the former leader wouldn't account
for, for example, what happens when a leader drops out of contention,
because it would show large but meaningless time gains.
My method also hews closer to what you do on days when the leader keeps
the jersey and gains on all rivals: time advantage is listed as however
much more time he has over 2nd, even if a different rider is in 2nd
It seems strange to benchmark the yellow jersey against second place,
but much like bidding on eBay, he's fundamentally the guy who determines
the effort (or interestingness) required to win the Tour.