Darn, why did I leave this? I usually address this sort of shoddy reasoning with Sisyphean regularity and chipperness when pressed…either way,
@Third Serve gets the blame for this thread resurrection by rousing me w that belated “like”
So, there are actually several “counter arguments”:
One is the thing I’ve already alluded to, which is that small sample sizes are likelier to encounter noise and variance, while larger ones (at least in part) root out such things. Your sample is minuscule relative to the lofty conclusions derived from it. Ergo, your whole thread is a rendered a non-starter.
Secondly, these conclusions are actually undermined by your
own criteria lol. They’re self-refuting and circular. Self-refuting because if “better statistics = higher quality”, then ‘07 Nadal was undoubtedly better than ‘19 Djokovic…his per-point match stats blow Djoko’s out of the water. Yet he lost. Tells us? Perhaps Nadal’s
hypothetical superiority (based on his actual,
statistical superiority) influenced ‘07 Fed’s stats?…
…and circular because the predictable retort invites the age-old chicken-and-egg causality dilemma where we have to squabble over whether Nadal’s statistical superiority is due to his opponent being worse or cuz of a genuine edge (and vice-versa wrt ‘07/‘19 Fed and HIS opponents).
Thirdly, and most directly relevant to the topic: you haven’t actually
shown ‘19 F Fed was even just statistically superior to ‘07 F Fed. ‘19 was the longer match but Federer’s per-point differential was better in ‘07. He also got a higher % of freebies while landing in more first serves. Amusingly you list “aces” as an advantage ‘19 Fed holds even tho ‘07 Fed had a much higher ace% (15.4 -12.3). Again, if you believe this is because ‘07 Nadal was appreciably weaker than ‘19 Djoko (allowing Fed to “stat pad”, so to speak), that’s something we can attempt to tackle…but in the end there’s no way to neatly isolate player quality from match statistics (particularly single-match statistics).
Hope that covers things.