Betting midpair + pocket pair absolutely makes sense to me, we have bad potential to improve to 2pair so we rather turn our pair blocker into a bluff. All I can see on those pics is, that solver is well mixed doing both betting and checking like it does in many situations and often you cannot even see a pattern. So it does not say we should check it in particular. If you start checking alls mid-pairs that´s probably too much of a leak... You need some randomizers and an ability to spot those hands ingame like I do with bdfd´s and/or straight draw blockers to approximate a frequency. Betting straight draw blockers makes sense because of the mentioned blocking effect and card removal, just you having it does not giving you more realizable equity. Your range vs range Equity is like pretty much the same anyways on similar boards with alls those combos.
I am not betting 100% mid-pairs lol. Idk what kind of semi-bluffs you are talking about on an AK2rb flop when we are up against unknown. He is not supposed to have lots of wheels. The average players does not check-raise his AQTx top-pair+gutshot and first it´s probably not a good idea either from his perpective. I mean I havent seen that move once over like 100k hands. What your plan is here when you check all mid-pairs? I mean what I could imagine is calling a bluff on a river blank with mid2-pair after turn went check-check aswell but that´s something not many player do with the Ace on board? But if they do that might be a note. I´d say solver is even mixing all twopairs or better on dry flops and not betting or checking them in a recognizable pattern either. Ofc with some tendencies like betting top-set more often. And you cannot imitate this spot like solver since idk how much combos that would be to memorize.
PS: Against nits and most tags it´s a cbet anways even if they check-raise all their gutters, I mean that´s not only unbalanced but also kind of stupid in a tight environment and nobody does something like that not even someone who has 80/50 with 35% 3bet and AF 6. That is the power of the Ace and the common knowledge that it is well presented in all ranges. So stab and take down with that improved scare-effect and know where you are. Because he will fold so much (everything weaker than top-pair which is a lot alrdy). No need to balance stuff here. We are miles away from GTo anyways against 99% of population on this flop. In general regulars check-fold without Ace+backup (usually gs), set or wrap which is reasonable against all those tight cbettors like you but if we want to make money we definitely want to bet more than 45%. I think sth like 60% is fine to start out until villain starts check-raising solver-like frequency or widening check-calls with its top-pair range.
PPS: Against nits or players with VPIP lower than 20 I would even cbet around 80%. Yesterday I had that nit 14/9 openraising from UTG and after him check-calling twice I knew he had set at least but the river brought a broadways straight and I tried to bluff him off his top-set on this scare-card with a pot-sized bet with a random blocker from a like 10%+ capped cold-calling range preflop. Did not work lol but he tanked around like a half minute pretty fun in hindsight since my turn barreling range was likely more hands already than his opening range xD.
PPPS: He was not unknown we had a decent history of thousands of hands and I already figured him out pretty well. I am not bluff triple-barreling nits as a manner of habit lol.
Mid-pair is a good randomizer for bluffs since it blocks continuing ranges decently. Or what kind of hands do you bluff here beside QJ, QT and JT gutshots, they do not make up the mentioned 44% of a 30% CO opening range .. (also some have to be checked back for balance purposes) ..
