After suspecting that the Res and Flat attributes had negligible effect on TTs I decided to test it. As you can see from the screenshots below they appear to have no effect whatsoever.This could be quite relevant for statmaking as GC contenders with great TT don't need high Flat to help in TTs.
In the test all riders with the same value in TT were set to the same effort from start to finish. The 4 riders with higher TT in each test were set to 1 effort above the 4 riders with lower TT.
I used the 67.7 km ITT from the 1994 Tour variant and the Chronos des Nations.
All riders were set to 70 in all attributes other than Flat, TT and Res. I tested with values in TT of 85, 80, 75 and 70. For each of these TT values I had four riders with the following attributes in Flat and Res: 1 with 85/85, 1 with 85/50, 1 with 50/85, and 1 with 50/50.
Unfortunately I couldn't control race day and wind conditions which appeared to be the most deciding factor in determining the times. However if the Flat and Res attributes had any meaningful impact it should be possible to differentiate between 85/85 and 50/50 riders regardless of these factors.
As such statmakers need not consider the effect of these attributes with regard to TT ability and should be careful not rating good time trialists too highly in these attributes, since it will only affect their performance in other races than TTs.
Funnily those riders with 85/85 Res/Flat did worse than the ones with 50/50:
67.7 km TDF stage with 85 and 80 TT riders:
Spoiler
67.7 km TDF stage with 75 and 70 TT riders
Spoiler
Note Bagdonas' time wasn't the winning time. He was around 3:15 after the winner (Dumoulin).
Chronos des Nations with 85 and 80 TT riders
Spoiler
The 80 TT 50/50 rider is not included as only 7 riders were allowed.
I surmise that especially hill is quite significant for TTs and may be part of the explanation for the rather big differences between equally rated riders, but this test was specifically to illustrate the negligible (at best) effect Flat and Res has on TTs.
I wasn't aware that the TT attribute replaces Flat on TTs. I would have thought Flat was suposed to complement TT, but it explains why Flat has no effect on TTs.
I'd be surprised if Hill/Mountain works the same in TTs as in orther stages, since you rarely go above 84 effort in TTs and as such the Hill attribute would be largely irrelevant.
That could be the case though. In mountan TTs, the TT attribute seem to have little to no effect with Mountain naturally being the prevalent attrbute. On hlly stages I would think it is a combination of Hill, Mountain, Downhill and TT, but possibly Hill is not relevant at all.
I might do some tests on hlly TT courses to determine the importance of these four attributes. Intuitively the order of importance should be TT, Hill, Mountain, Downhill depending on the lengths of the climbs, but this might not be the case.
