I actually prefer a HJ preflop strategy opening T8s,97s and 76s and not raising QTo.

On this flop I have 258 combos in my range.

For my small cbet approach, without considering check-raising, I want to bet:

Strong value: KK+,88,A8s,T8s,98s,87s,AQ,KQ,QJ,QT,Q9 (63 combos)

Strong draws: All Flushdraws (18 combos)

Marginal draws: Gutshots, 2 card Backdoor FDs, A and K high diamond BDFDs (60 combos)

Weaker value/protection: 55-99 (24 combos), AK no backdoor (9 combos)

This gives me 174 betting combos, or 67.4%

I typically prefer them as checks, but considering my overall strategy I can add JJ/TT to my value range (12 combos)

I can bet the remaining 97s and 76s for air with weak straight backdoors (4 combos)

For the last 17 combos to reach 80% I’m not sure what is best, but I’ll use my AJ/AT/A9/KJ/KT with low diamond (15 combos) to round out my flop range. I think slowplaying QQ is best, so I won’t bet it.

As played, I have 205 combos that reach the turn. I’ll assume villain raises his combo draws, T9 and 3/4 trips combos, so he reaches the turn with 120 combos.

For now I’m ignoring the option to check-raise, but I’ll consider that in later posts.

My hands that can probably bet for value: KK+,88,AQ,KQ,trips,flushes,66 (64 combos)

I can probably get away with bluffing at a 40% frequency or roughly 43 combos.

Most likely candidates:

A and K high 1 card flushdraws (24 combos), JT with a diamond (6 combos), T9s and 97s (6 combos).

This leaves me with 36 bluffs. I don’t love barreling naked gutshots on the 3 flush, but it brings me to 42 combos so I’ll go for it. 

This is a less terrible situation than I originally expected, as I can bet 106 (51.7%) of turns with this strategy. However, this leaves me with a very weak checking range. Facing any reasonable bet I have a hard time calling with more than my remaining top pairs (18 combos) and JJ/TT with a diamond (6 combos). So I will probably fold ~ 76% of the time I check the turn. It is tempting to peel with my AxJ and KxJ but extremely marginal in my opinion, and even then I’m folding 70.3% of hands after checking.

There’s a lot more to look at but a combination of being tired and slightly overwhelmed by how deep I plan to go means I will have to save that for another day. I think the next step is to consider more thoroughly how my value range stacks up against villain’s continue range, and also how many combos I expect villain to fold so that I can adjust my bluffing frequency. Once that is done I’ll do a quick river analysis and then work backwards to adjusting my flop strategy. Who knows where this will lead after that…

I really welcome any and all feedback, thanks.

Very good question Andrew. I don’t think it’s at all necessary to be barreling at a high frequency, but I tend to default to it when I don’t have a better plan. My main motivation for this post (and a lot of the work I’ve been doing lately) was that I took your advice from one of our coaching sessions and recorded myself playing a bunch of times while talking about why I was making bets. After I eventually got around to going through most of my recordings, I often heard myself saying a variation of “I want to keep betting here because I think they fold too much” without any good reasons to back that up. Even when I was more specific, I’d often list 4 or 5 hands that I expected to fold in spots where basic range analysis shows that villains would barely fold anything else.

I’m trying to break down and rebuild my approach to almost every part of the game. Not sure if I’m doing it right, but I’m trying to analyze this hand from my mistaken ‘autopilot’ approach and then design a better strategy for each street by working backwards from the river. One of my overall goals is to strengthen my check-calling ranges and widen my check-raising ranges, but I’m finding it hard to approach those ranges without deconstructing and rebuilding my betting range.

Foucault said
It’s hard for them to fold too much on every street. Typically if a player over-folds the flop, then that implies that his turn range is excessively strong.

For sure, extremely rare player that over-folds every street. My over-generalized perception of reg-ish players at the stakes I’m playing (based on a mediocre grasp of GTO frequencies):

vs FLOP cbet

Roughly 65/25/10 split of underfold / continue roughly correctly / overfold. Many of the better players seem to under-fold against other regs, which is probably a good exploit vs population (although they may just have specific reads). Not as confident in my grasp of optimal raising strategies, but I think there’s about a 75/10/15 split of people 2-betting too rarely / roughly correctly / too often (mostly as check-raises).

vs TURN cbet

Roughly 55/25/20 split of overfold / correct / underfold. That’s slightly less players over-folding than what I suspect, I want to err on the side of pessimism about my ability to recognize patterns and analyze data. I’m much more confident saying that most players don’t 2-bet nearly enough. Probably at least 80% of regs under-bluff and/or miss good value raises, myself definitely included.

vs River cbet

This is much harder to analyze. Predictably, the sample sizes are extremely small after roughly 110k hands this year (lost old database to a virus and/or sucking at file transfer). I’ve identified players that seem to consistently underfold / overfold to river bets, but the largest sample I’ve found for ‘fold to river cbet’ was 8 (with 4 folds), which seems meaningless. Gut feeling is that more players overfold to 3 barrels than underfold, but I won’t hazard a guess as to ratios.

In the recordings, I was exclusively using the generic ‘they fold too much here’ on the turn, but it was kind of shocking to realize how weak / auto-pilot my thought process can get. It’s all well and good to do complex analysis, but I don’t think I’ll reach a point of effectively implementing my study work until I deconstruct my over-analysis and derive much more simple conclusions. Probably not the fastest way to improve my game but I’m a huge nerd and I over-analyze everything, so it might be the best way for me to learn.

Been doing some oldschool pencil and paper work for the next step, will probably get it posted Monday.

theginger45 said
I’m surprised to hear you suggest that regs are under-folding to flop c-bets. I would have thought it would be the opposite, unless games have changed in the ~1yr since I was playing outside the US. Are you consistently noticing regs with <40% fold to c-bet stats over large samples?  

In retrospect, my ratios are way off, but I play with a lot of regs folding 30-35% over >1k hand samples (still pretty small). That being said, there are more regs folding 55%+ over similar sample, so I was wayyyy over focused on a tendency that doesn’t apply nearly as much as I wanted to believe.

I keep doing this throughout my learning curve. I notice a tendency from certain opponents, or realize I can 3bet/check-raise/whatever wider in certain spots, and end up over-valuing my ‘new discovery.’ I guess it’s better to be constantly looking for new adaptations rather than playing paint-by-numbers poker, but I should find a way to remain more consistently humble/skeptical.

BugFuzz said
Personally i think your problem is pre flop range.

I appreciate the advice, but this preflop range is a bit tight if anything. I’m pretty comfortable navigating post-flop (however poorly I may be doing it), so I don’t find myself thinking about hands in the way you describe. I agree that we want to open tighter if we expect button to flat very frequently, but 7 hands at deeper stacks is not a large enough sample size for me to start adjusting my open range without more info (villain was not overly loose in any other way that I had noticed).

Foucault said
This is also not the sort of situation where I’d try to apply the “bet 80%” heuristic from Poker’s 1%. 

I’ve never read Poker’s 1%, but I agree there’s no way to cbet 80% OOP against remotely decent opposition. Again, this whole analysis is starting from an approach that I know to be flawed so that I can figure out exactly where and why it breaks down, and then redesign it. When I finally get the next steps posted I think that’ll become more clear.

I messed up counting combos earlier, the cold-call range I gave villain has 152 combos after the flop. Based on the flop strategy I assumed, I expect 116 of those to call and 7 to raise. As played, after 6 turn, there are 112 combos in villain’s range.

Hero Bets 6.17bb, BTN calls 6.17bb

After I bet just over 3/5 pot on the turn, MDF suggests that villain should defend 62.2% of his hands, or 70 combos. I’m having a tough time finding that many combos to continue with. I think there are 55 combos that easily call or raise: 27 top pairs, 8 flushes, 9 nut FDs, TdTx + 9d9x (6 combos), trips/boats/quads (5 combos).

King FDs (3 combos) are probably the next best hands to continue with, but there don’t seem to be great options for the last 12 combos.

I think villain’s strongest remaining hands are: TT,99,As6s,AhJd,AsJd,AcJd,AhTd,AsTd,AcTd,Ah9d,As9d,Ac9d,JsTs,Js9s

I really don’t like calling with any of those hands except for maybe TT/99. On deeper stacks I think raising AxJd and AxTd might be reasonable but I doubt villain wants to shove those hands, and any smaller raise seems silly when I only have 16.7bb behind. I wasn’t planning to do this, but I think we should change V’s preflop range.

If V folds A9 / A6s, 3bets A5s / K9s, and flats 1/2 AA combos they have 149 combos in their coldcall range (131 on the flop).

I’ll add 2 combos of AA to V’s Flop raising range, and keep AdAs calling. Maintaining assumptions about villain’s flop play, they now have 3 less FDs, 2 less  BDFDs, and 12 less A-high hands that call the flop. Villain calls flop with 100 combos and raises 9, arriving at this turn with 97 combos.

Facing a ~3/5 pot bet, villain should defend roughly 60 combos. They now have 1 overpair, 2 less flushes, and 3 less nut FDs, so 51 combos that easily continue. Adding KdJx to that range makes 54 combos that can continue on the turn, which is probably close enough to MDF to be reasonable. I think it’s extremely marginal, but if villain wants to hit 60 they can call the remaining combos of TT/99.

More soon.