Pro cycling thread
1 lurker |
83 watchers
Oct 2022
5:02pm, 3 Oct 2022
2,303 posts
|
MudMeanderer
UM: There is an adaptation of normalised power used by a certain site beginning with S that does weight later efforts more than earlier ones, but that's only doing so over 30s periods rather than the whole ride. Fatigue resistance is also something that's also increasingly being considered in high level cycling - not what you can do at the start of a session, but what you can manage several hours in. It can be quite a personal attribute. |
Oct 2022
6:02pm, 3 Oct 2022
20,790 posts
|
eL Bee!
Degree of difficulty based on the data you have. I use metres of elevation ascended per mile so those rides are both in the 10metres/mile bracket. I then look at the pace that I have ridden the ride, and make a rough calculation as to what that would have been if the ride had been flat. It turns out that for every 10m/mile of climbing, corresponds to a 2mph reduction in pace compared to a flat ride. So if you rode those rides at, say 13mph, you'd estimate 15mph on the flat. I haven't actually plotted the acent for the full ride against average speed, but as a rough guide it seems to work quite well and I know that if it calculates a 19-20mph flat ride - It has been a good pace for me (and *us* on the tandem" and if it converts to a 15mph flat ride - it was a slower one for us. It would be really interesting to collate the ascent and average speed data for loads of rides and I predict the middle of the graph (5metres/mile to about 40metres per mile) would be quite straight - with a shallowing of the graph at the far ends of both. |
|
Oct 2022
6:56pm, 3 Oct 2022
2,304 posts
|
MudMeanderer
Fetch: if you do manage to sort a way to correlate any useful metric with elevation profile, I'd definitely be interested in it. I spent some time a short while ago trying to develop a first principle method to use as a correction factor for different courses in our club. I managed to get something that estimates, for a given average course gradient, how much slower a rider could be expected to be compared to a pan flat course at the same average power. However the result changes with how heavy the rider is, how aerodynamic they are, and what flat speed you're comparing it to. That's before you get into different shapes with the same total elevation per unit distance. So reducing to a single comparative value can be tricky (though you may be able to compare means) Increasingly there are Machine Learning algorithms being employed. So Cycling Time Trials have SpinData that ranks riders based on performance without them needing to ride a common course. BestBikeSplit does some numerical variational calculus based on course profile, direction, forecast wind and power profile (I believe) to suggest an optimal pacing strategy. I guess you may have a lot of data that could be used for some interesting ML. I expect a big challenge in your case is accounting for effort - CTT can pretty much assume that all the data they have is almost flat out (for the distance), but training data is likely to be a lot more varied in terms of effort. It's certainly an interesting problem to explore, but I also think that makes it a really quite tricky one with lots of potential approaches. Have fun! |
Oct 2022
7:18pm, 3 Oct 2022
6,574 posts
|
um
I'll ask my friendly in-house pro-cyclist and see what she says, or is taught/trained for about fatigue resistance.
|
Oct 2022
11:34am, 8 Oct 2022
2,060 posts
|
Big_G
Ganna going for the hour record today. Coverage on Eurosport/Discovery and on Ineos YouTube channel I believe. The attempt starts at 7pm U.K. time. eurosport.co.uk |
Oct 2022
9:58pm, 8 Oct 2022
38,456 posts
|
SPR
Good to 'see'* Ganna really but the hour record out there and beat all record today. * I didn't see it unfortunately as wasn't back in time. |
|
Oct 2022
7:43am, 9 Oct 2022
2,061 posts
|
Big_G
I watched it and enjoyed. It was a proper production this time, whereas the coverage of Bigham’s was quite poor. Ganna did quite a big -ve split, as he was behind Bigham‘s time for roughly 20mins. This was apparently in an attempt to keep Ganna’s core temperature down for as long as possible.
|
Oct 2022
8:24am, 9 Oct 2022
45,388 posts
|
.B.
Impressive improvement! Enjoyed Lombardia yesterday too. Valverde top 10 in his last race. Sad for Nieve that he crashed out in his last race. Will miss Nibali, though he’s not been his best for a while. Apparently Gilbert is going too. |
|
Oct 2022
2:44pm, 9 Oct 2022
45,390 posts
|
.B.
Argh they have lost the feed for the gravel worlds just as the chase had started
|
|
Oct 2022
2:51pm, 9 Oct 2022
2,306 posts
|
MudMeanderer
I've switched to Paris Tours which seems to have as much gravel!
|
Useful Links
FE accepts no responsibility for external links. Or anything, really.Related Threads
-
So who won the tour from 1999 to 2005
May 2024
-
Cricket Thread
Jan 2025
-
FOOTBALL
Jan 2025
-
Elite Athletics Thread
Jan 2025
-
DraftEveryone: a US sports wire
Jan 2025
-
Ski/Snow hols
Jan 2025
-
Rugby, the original version
Jan 2025
-
Rugby League Thread
Jan 2025
-
Fetch F1 Fans
Jan 2025
-
Ladies Who Lift...
Jan 2025