Depending on your RFI ranges , but i think you can make some solid RFI ranges with a defense frequency somewhere around 35% to 40% vs a 3bet from each position slightly tighter oop vs ip. I think the MDF vs a 3bet given normal bet sizes pre is something like 28% before we become exploitable and villains can 3bet any two and auto profit. So as long as your not folding over 72% as a default your going to be ok.

It depends at what stage you found dificult to explot other player tendencies/leaks.

It doesn't take long to build a player pool profile though, in which case we can start building an exploitative strategy against the average player. Playing balanced in a pool where the average player grossly over-folds and rarely bluffs in many situations makes no sense.

I think a lot of people are victims of a confirmation bias when they're assessing player pool tendencies. It's particularly obvious when you see people creating threads asking for advices in order to counter the "massive increase" in agression after moving up one limit.

That is true. But I'm talking about analyzing actual hand data. I guess that doesn't work on sites with no hand history, but if you have hand history, you can look at exactly how much the average player c-bets, folds to c-bets, folds to 3-bets and so on. Generally the leaks are pretty obvious and simple to exploit, though that still requires knowledge of balanced play, otherwise you don't really know what to look for. But just after playing let's say 5000 hands of 6-max, you already have 25 000 hands on your opponents. That's more than enough to find leaks in the most common areas.

But there is a difference between "average" and "typical."

If you played a lot of hands at (say) 10nl and collected a PT or HEM database on the field, you could average VPIP, PFR, and 3B over all the players in your sample, and get averages for the field for a profile of the average player.

But notice that there are a relative few players for whom you have relatively large numbers of hands, i.e. regs, and a large number of players for whom you have small samples. If you construct averages of these stats that are weighted by sample size, the averages will be noticeably different from the first set, because the regs who are grinding daily and multitabling are a different population from the players who show up occasionally for fun.

You can generally tell from the stats from a few dozen hands which group an unknown falls into, and play accordingly, but do remember this: if you are mixing it up with a fun player, a regular can get involved, and we need to be concerned about exploitability.

Suppose, as the OP asks, the general trend among fun players is to have really snug 3betting ranges. Yes, we can exploit this by overfolding. But we can overfold in a balanced way, with balanced ranges for four-bet/shove, four-bet/call, and four-bet/fold. We want our value hands to be balanced by bluffs, but if we are overfolding we are playing fewer of each.

tl;dr: you can exploit the crap out of opponents' overly-tight 3-betting ranges and still do so in a balanced way.