Cumulative stats - Federer [3] vs. JMDP [24] - USO 2017 QF

falstaff78

Hall of Fame
Here are cumulative stats for Fedr and JMDP through the first 4 rounds of the USO 2017. Please make it known if you wish to be added or removed from the notification list.

High level message: With the exception of record in sets, Lord Fedr has stacked up slightly better numbers than JMDP, versus similiar opposition. Service games are similiar - difference in performance has come from return.

  • Lord Fedr has won a lower pct of sets (dropped 4 vs. 2), but done better at winning games (60% vs. 56%) and points (55% vs. 53%).
  • On serve they have been about the same
    • they have an identical hold% (88%)
    • they have similiar SPW% (71% vs. 70%)
    • also they have outperformed the tour by the same amount vs. their opponents (11pp vs. 9pp)
    • both have been woefully unclutch at saving break points - Fedr 20 pp below his regular win% and JMDP 9 pp
  • The difference in performance has come from return - Fedr has broken more often (31% vs. 24%)
    • Interestingly, on return points (as opposed to return games) they are quite close - Fed leads 42% to 39%
    • The large difference of 7p.p. in break% comes from the fact that Fedr has been much more clutch on break points
    • His break point winning% is 7p.p. higher than non-BP win%; for JMDP that number is -3 p.p.
  • Fedr leads in all fatigue metrics (except points played) - this is likely to be a factor given JMDP's health situation
  • Also, Fedr has a slightly cleaner game than JMDP - better winners to U/W differential, and slightly better at the net

Enjoy!


JiGByCh.png

VY4qDIb.png




@Meles @Gary Duane @Chanwan @Sysyphus @Red Rick @BeatlesFan @Silence @VaporDude95 @Vrad @mightyjeditribble
@stringertom @Sentinel @73west
@Prashanth HS @Firstservingman @Fedberg20
 
Last edited:

kOaMaster

Hall of Fame
Great stuff as usual, very interesting. I would not have thought Del Potro having so many more meters covered, thats a surprise to me. He's not really the long rally-kind of guy
 
It's interesting that Fed has lost more sets than Delpo, but his %age of winning return games is better (and %age of holding serve the same). Of course, it's a matter of how he won those matches.

I'm not sure that I agree that they played similar opposition, though, despite the similar average ranking. Delpo having to play a top ten player is quite a difference (of course it is only one match out of 4 played, but still that has a significant effect on the data).

I know that you are already weighting by number of points played, but I wonder whether there is a better way to compute the average? E.g., what about using an average of the logarithm of the ranking, or possibly taking an average of the ranking points / ELO points, instead of the ranking, and then seeing where that average number of ranking points would place the player in the current rankings? Have you thought about trying something like this?
 
Top