I just posted a thread hoping to spark some discussion about how to respond to a 70% minraiser 20bb deep, because I think that's a really hard question to answer given how tricky it is to evaluate "call or jam light" decisions from the big blind. I noted that for once, I was posting a table and not at all being authoritarian about the analysis, putting it out there without my piece of "this is what it means" already attached to it.That's not going to be the case for this thread.Most opponents won't open as much as 70% from the small blind. Coming up with an aggregated percentage is tricky - I think most would agree it's somewhere between 50-65% that opponents raise on average (they will limp some as well), and if I had to pick a number, I'd say that it's about 55% at this stack depth when we know nothing about our opponent except that we don't know him (technically, that gives us information. I'd use a slightly more aggressive strategy against a random reg, even if I had never played him before, because regs tend to have higher opening frequencies). So let's use a 55% opening range (specifically, propokertools' top 55%), and a calling range of AA-22,AK-A5,KQ-KT,Ax4x-Ax3x,Kx9x,QxJx-QxTx. That's some light hands, but nothing crazy.I made a table of the expectation from 3bet jamming given these parameters compared to the start of the hand. -1.0bb is the same as folding to the raise. Here it is:In the traditional paradigm of looking at 3bet shoving (is it better than folding?), 35% of hands make the cut, being equal to or better than -1.0bb from the start of the hand. So, people will tell you that jamming a hand like J8s is +EV, because shoving is 0.3bb better than folding.This, however, is a ridiculous standard for deciding whether a play is +EV. We should always be comparing against the next most reasonable alternative - in the case of J8s, the obvious other choice is that we can flat.Lifetime, with Q8s, Q7s, J9s, J8s, J7s, T9s, T8s, T7s, 98s, 97s, and 87s, a cluster of hands similar to J8s to increase sample size, I am 0EV from the start of the hand from flatting. That's a whole 1bb in expectation better than folding, and much better than the expectation shown in the table for shoving - even though every single one of these hands is "+EV" when only held up to the standard of a jam just having to be better than a fold. That's why all of these hands, by a very wide margin, should be called instead of jammed against an unknown opponent for 20bb against a minraise.When you actually narrow it down to the hands that are worth 3betting, it's about 19.5% of hands. Since your opponent will openfold more often when you have bigger hands due to card removal, you'll have an even smaller actual 3betting percentage.Despite their mediocre equity from jamming, KQ, KJ, KTs, and QJs do extremely well in 3-bet pots, and gain a lot of value from making a potsized 3bet and inducing calls from loads of dominated hands, or hands with two undercards, that we play very well against on the flop. I recommend 3bet/calling these hands readless and bet/calling a ton of flops when you get flatted. Along with those hands can go your premium pairs, which also want to get flatted by a wider range.3-bet bluffing with non-allin bets when readless is a different story, and I'm willing to consider arguments that this leads to the best expectation for hands like J4s. However, the math is pretty clear that it's spew to jam with these sorts of hands without evidence that your opponent is very likely opening a wide range.Note also, as per the "Do you call too tight OOP?" article, we're still playing a wide range even 20bb, about 56% of hands readless, with the borders debateable. I do think that if you're only playing 45% of hands at this stack depth readless, you're bleeding away a lot of money.I'll open this up to comments/questions.