This is not a defense of the mentioned trainers. It's a statement about the analysis one should do related to these kinds of stats.

One of the things that allows Pletcher and Frankel to generate such high win percentages is that they spot their horses where they can win by shipping around the country and avoiding tough spots. That's a skill that gets totally neutralized for races like the Derby and BC. Those races are automatically MUCH TOUGHER spots than the races that those guys tend to avoid during the year unless they have "the horse" they believe will win. So it's virtually a 100% certainty that their win percentage will fall sharply for the Derby and BC relative to the rest of the season. Once your win percentage falls enough, bad streaks of various types are more likely and could mean nothing.

Again, I'm not defending them. I'm just pointing out a reality about the evidence.