It's not that simple unfortunately.
EVF resolution is one thing, and what the sensor is able to feed into it full-time (and with as little lag as possible) is another, there is a reason why the TOL A9II has kept the lesser resolution EVF, and for that matter the SL2's live feed isn't as crisp either, so it is a fine line between them.
I don't think you can argue, necessarily, that Sony kept the A9's lower resolution EVF for the A9II due to performance concerns. Isn't it just as likely that they wanted to update only what was most necessary, while watching their bottom line?
What lag is there in the SL2? Is it between the subject and what's seen in the EVF? Is it worse than any lag observed in the A9? Just curious, if you have a source. And if this kind of lag does exist, you're right, it could be attributed to a system that can't drive the EVF fast enough to keep pace with the subject being captured. And, by all means, if the same folks who calculate sensor throughput want to step in and offer some guesses at the processing speed a higher rez EVF would necessitate, go ahead, but my understanding of this is too amateur to speculate especially about the relationship between subject/EVF lag and processing/sensor speed.
It bears mentioning that both the 3.6 million dot EVF in the a9 and the 5.76 million dot EVF in the S1R can each display at 120 fps, according to Imaging Resource. It would be a shame if the higher resolution and silky-smooth refresh came at the cost of subject lag, but even if this is the case in the S1R, I wouldn't assume the same in a new Canon system with different (and potentially must faster) architecture.