I think you guys might be right that this is a punt but I'm still not 100% convinced yet. I think as a default I will jam vs 1/3 and then fold vs bigger river sizing's until I understand this spot better.

Anyways here are results.

Sizing, yes, but even just betting. But within that sizing range, it will mostly be the top of his range and some bluffs. So straights, flushes, and bluffs. You're basically hoping he's bluffing, so you can out bluff him. Not something I'd recommend at these stakes. If you were villain, great... that's about where you want to keep it to be consistently profitable at these stakes imho.

I think it is because flush draws that run into top pair on the turn should often check, whereas flush draws that don't should barrel frequently.

I would guess that it's because JxTh unblocks bluffs with both cards while JhTx only unblocks bluffs with the Jh.

Also in theory river raise bluffs almost always come hands that can also call. (I think check down lines/lines where you just have a lot of air and are in position may be an exception to this). So if you don't think you can call because villain is under bluffing then you shouldn't be able to raise profitably either unless villain folds too many value hands. Specifically they have to fold an extra value hand for every bluff they are lacking. Some players might fit into this category, but it is difficult to know if they are doing this without a ton of hands on them or some very good MDA.

And you can always hedge your read/intuition about a spot by striving to over/under do a given action rather than never/always doing that action. It also helps keep you sane when you can just blame it on the rng lol.

This is true in most spots, but it does this on the river because the properties that make a hand a profitable bluff raise are hard to find in hands that don't beat bluffs. Since the solver will always be balanced, a hand that beats bluffs and doesn't have bad blockers will usually hover around 0 EV.

However, if your opponent doesn't know how to select bluffs and bluffs with some hands that are too strong to bluff (but remains balanced), then you would likely start to see some of your bluff raises become -EV calls but even more +EV bluff raises. This shows that there's nothing inherent about a hand's EV to call that affects its EV to raise.

Yeah this concept is my main takeaway from the thread. I need to think about it some more and I'm sure there are exceptions wrt to MDA but in theory if you don't think you can call because villain is under bluffing then you shouldn't be able to raise profitably either. So then we draw our river raise bluffs from a range that should be able to call at some frequency vs villain's bet.

Main exceptions are if he bet's too thin for value in his bet small range and doesn't put enough strong hands then we can exploit that by raising hands that would theoretically be 0 EV calls and just never call them because they would be losing calls.

fodl
if you're hoping he bet/folds sets for 2/3, meh
whereas sets and flushes go that size for sure