Nobody wants to see 50% running clocks in the playoffs, and that's good evidence that 16 teams per region are too many.
At the same time, about 1/4 of the games by my count of 46 last night were upsets with a 9-16 seed beating a 1-8 seed. That tells me the Harbin system is far from perfect at picking the top 8, especially among the #5-#12 seeds. I'd bet that the Harbin system is better at getting the top 4 than it is the top 8.
To which some will say "so what?" I get it. How many #9 - #16 seeds according to Harbins were going to win a Region anyway. Probably not many if any. But, when you let 16 in, there's virtually 0% chance of missing the eventual state champion which leads me to be OK with 16 unless we're willing to go 12 and give the first 4 a bye.
I attended the #7 Buckeye - #10 Padua game last night, and Padua was a holding call away from putting a running clock on Buckeye Had Padua at 5-5 been excluded, it wouldn't be the end of the world. Drew Pasteur, however, ranks Padua #4 in this Region which says his math gives them a very viable shot of winning the Region. According to Pasteur, Padua had the 8th most difficult schedule in the entire state in D3. My point is that his logic appears to do a better job of predicting winners due to SoS than Harbin Points.
If we drop to 8, I'd expect Padua to drop D2 NoRo, D2 Maple Hts., and/or D3 stud Ursuline to fill in with an easier opponent. Retaining 16 teams reduces the incentives for ADs to avoid tough schedules. That's the appealing part of 16 team playoffs (except for D1 where 16 of 17 is ludicrous).
So, my first suggestion would be to drop the Harbin system, use Pasteur's logic, and go to 8. If we have to stick with Harbins, then I'd suggest just the Top 4 or Top 16 except in smaller D1 because it doesn't do a good job separating those in the middle.