Data based analysis and commentary on nordic skiing
Posted by Joran on Thursday, July 5, 2012 · 3 Comments
This time we’ll look at the Tour results so far by team (click for larger version):
Filed under Cycling, Fun Stuff · Tagged with bump chart, cycling, tour de france
first of all, thanks for your excellent charts.
I came across them on junkchart and have allowed myself to post there a few comments that you might (or might not) find useful :
- The Y axis : it is shown with a lower time being the better and thus could be deceiving as one would naturally think that the higher is the better.
- Maybe some sorting by performances rather than alphabetical order would make it a bit easier to read ? (either by fastest runner or by fastest group)
- A possible nice to have would be a yellow line for the “maillot jaune” ?
- This seems to have been design for a paper format (A4) which is quite different from the format of a computer’s screen. Maybe showing 6×4 charts rather than 5×5 would allow an easier “glimpse” and general view on a screen.
- Someone else put a comment as well about the x-axis and the prologue, I’ll let the poster talk for himself if he wants to
I’d be happy if you could share your thoughts,
Much of what you suggest I have done in the past, I just don’t have much time this summer for such niceties. For example, sorting the teams non-alphabetically and adding the yellow jersey. (The reason why I haven’t been sorting the teams is that the site I scrape the results from can be quite inconsistent when spelling the team names from stage to stage; omitting or adding spaces, hyphens or capitalization. I spend enough time as it is just adjusting for that.)
I respectfully disagree about the y axis. Everything in (timed) sports is quite consistent: smaller is better. For me, it would be extremely confusing to have “up” mean “better”.
It is not the case that “everyone” is at the median at the prologue. That is simply an illusion due to the vast scale differences as you progress to subsequent stages. Prologues are very short, and so nearly everyone will be within a handful of minutes of the median (or less than a minute, even). But after only a few stages, the values will easily range enough that you can’t really discern differences in the first stage.
That said, I’m aware that I probably have some Stage 1/Prologue naming inconsistencies for some of the past Tours. (i.e. what was really a Prologue is called Stage 1, or vice versa) Again, I just haven’t had time to go back and relabel those. I haven’t been overly concerned about it because “Stage 10″ isn’t really comparable across Tours anyway, since they may be very different lengths/profiles, etc.
Once the Tour is over, I might have time to go back and do some different graphs that are a bit more detailed. At the moment, I’m just concentrating on updating these basic ones on a regular basis.
thanks for your reply, this is obviously quite a lot of work that you do, and my suggestions were only aiming at trying to find possible “theoretical” improvements.
About the Y-axis I wasn’t aware of that consistency for timed sports, so again, thanks for letting me know.
Mail (will not be published) (required)
© 2013 Statistical Skier · Shades of Blue theme by StudioPress · WordPress · Log in