Or.. does the better player just always win on the day? I think given the punishing effects of slow reactions in certain situations and engagements in SC2 makes it worth thinking about?
I think it would be interesting to see the data on this, especially with some kind of performance / aligulac ranking weighted adjustments (and large datasets). The online data would likely drown out the relatively small number of offline tournaments but perhaps if non-cross server online data points + offline were compared vs. the rest some stats genius could have a crack at it.
There's also many potential confounders; such as tournament structure (GSL style vs. weekender, round robin vs. BoX single vs. double elimination etc.); relative player performance in the booth/on stage vs. at home; and relative latency for each player given their location to the server, internet speed; the effect of longer/late game matches vs. short games; etc.
My basic assumptions in thinking about this would be that:
1) Micro potential / mechanical requirements are not equal between races
2) Mechanical reactions and micro are adversely affected by increasing latency
3) The relative effect(s) of a more or less equivalent ping increase for two players on different servers, are not always the same across every asymmetric match-up based on 1) and 2)
I'm sure most often player skill + performance on any given day >> than any of this but I for one would be curious to see if this could potentially make a difference as a kind of relative effect as a purely academic / statistical exercise.
Bias and intent disclosure: I'm a (shitty) terran and even shittier protoss player and also huge Maru fan. My hypothesis, which I acknowledge is likely a result of my personal bias is that the differences considered above would be greater for ZvT than PvT. For the record, I'm not pushing balance or the mechanical demands for one race are different from another, just that the effect of lag on certain mechanics (splitting/micro stuff mostly) disproportionately affect the various non-mirror matches.
TLDR: Other things being even; does asymmetric race design become more asymmetric when both players face a similar amount of cross server lag?