"Hope I have set up the filter correctly. I run at -39.7 ev bb/100 in these 2151 flat hands (most are flat 2x). The results really shock me as I lost most EV chips on hands like JTo, 87o, Q8o, K6o, 98o,76o, T8s, 97s, K9o, Q9o, 98s. I thought these hands were all perfect hands for flat but I run at -200 ev bb/100 for these hands."-39.7 bb/100 seems like a pretty reasonable number to me, considering you're supposed to play everything up until -100 bb/100. I'm a little bit suspicious of your -200 bb/100 claim with those hands - are you sure you're not just cherry picking hands with bad expectation from low sample size? If you're doing that badly with all those hands then there has to be a group that is +100bb/100 to make it all average out to the -40, which makes me wonder if you're just looking at some  small sample size things. For example, T9o +51bb/100, JTo -226bb/100: It's not like these are actually accurate representations.Even for my sample size, when I did analysis for the articles, I did it in big clumps of similar hands to get a better sense of true historical expectation. I think you have to do the same. "As I only have 20-70 sample hands for each one, to reduce the variance problem, I add one additional filter for Hole cards as K9o-K4o,QJo-Q5o,QJs-Q6s,Jo-J6o,JTs-J5s,T9-T6,98-96,87-86,76os. I still run at -68 EV bb/100 for these 1667 hands."Hah OK you're one step ahead of me. Remember, -68 EV bb/100 is still better than -100 bb/100 from folding! Playing OOP with marginal hands is hard, and you're looking to limit losses. My total historical EV is -20bb/100 flatting 15-25bb deep.I wouldn't go too crazy worrying about it - maybe hold off on flatting T6o, 76o, as you improve. I don't think it's that you're too fit or fold - one of the main points of my "flat wider OOP" article is that my stats are super super boring in terms of check/raising and all that. You may be missing opportunities to bluff turn/river when checked through. Make it a priority to focus on these holdings, feel free to post hands here when you're unsure of your line. 

"BB VPIP: 75 3bet: 24% Flop fold vs cbet: 29 Check/Raise: 18 Fold to 2 barrel: 40 Aggressive vs limps  What types of adjustments are you making against this player both pre and postflop?  How wide are you opening preflop considering you aren't picking up the hand uncontested preflop or on the flop that often?"We really should be able to go to serious valuetown against this guy. I'm definitely openfolding my bottom 30%. Really try to hone in on where he's putting his air hands - if he's rarely folding to c-bets, or double barrels, well there is a certain amount of air hands that have to be in his range, and it's not like he's check/calling twice with them. So his check/raising range may be more suspect, or he may donk with them, etc. Expand your thin value betting range of course, experiment with different sizings both pre and post. Some of your adjustments will depend on how thinking you think he is - a fish with these characteristics I'm often starting to make it bigger with value hands because fish tend to be more price insensitive. But regardless, realize that you're getting a ton of value for the top of your range and that it's going to be tough to do well against those frequencies with the bottom of your range.

"How many hands vs 1 player would you like to have before putting much stock into the following stats: -fold to cbet -3bet -fold to 3bet -fold to 2 barrel -flop check-raise -fold to flop check-raise   Obviously we're also factoring in hands shown down, notes,etc... but it's nice to also have a good idea of how much stock to put into particular hud stats over different samples." These questions are always weird to me because they're looking for binary answers when all good answers are going to be in the continuum form (I know you know this, mostly saying this for others who might read). I think the very first time they fold to a cbet, or 3bet, or fold to a flop check/raise, we're already getting information that we should be making slight adjustments to in good Bayesian thinking. And as our sample size increases we make more and more heavy adjustments.Already if our opponent folds to the first 3 c-bets on dry boards it's very likely he has a low check/raise bluff frequency. Already if he 3bets 2 of his first 4 opportunities it's likely he has an expanded 3betting range. But you don't assume his check/raise bluff frequency is 0% and his 3betting is 50%, you say, OK, here's what I've observed, here's my sample size, here's what most people do, and start operating under the assumption that he "probably has a low check/raise bluff frequency", and "probably will 3bet more than just monsters", ready to adjust as you get more information.I know this isn't really a satisfying answer, I do think once you get a few hundred hands against someone your confidence interval about those type of frequencies starts to narrow considerably. 

"Is 33 the strongest pair you are raise/folding vs a jam 30bb deep?  Do you ever prefer raise/calling or do you just raise/fold til 25bb & then jam? Also, if somebody 2.5x or 3x 3bets you with 30bb when you have 33, are you jamming as a standard or do you want to know they're an aggressive raiser 1st?  I'm guessing jam as standard."Yeah, I think I raise/fold 33 and raise/call 44. I think you can openjam 22-33 up until the 27-28 range, but that's purely intuition, I haven't actually done the math. You can also argue for raise/folding 44 30bb deep but I just can't bring myself to do it.I'm jamming as standard with 33. I'm willing to jam any pair readless over a 3x 3bet readless 50bb deep.lol 30bb superturbo-aments :p

"Against an 80% raiser do you think we should be flatting hands like K3o, Q5o, J6o, T6o, 96o, 86o, 76o, Q3s, J4s, & T4s?  If so, do you think we should ever flat any wider, particularly if our opponent starts raising closer to 100%?FWIW, I would test them with some 3 bets with these hands 1st but many people defend too wide vs 3bets that it becomes -EV."Some of these hands are better to 3bet with than others, against an 80% raiser I think you can make a small 3bet/small c-bet and have better expectation than flatting a lot of the time.I think you have a good sense of the borderline - all of those hands I think are really close between playing and not playing. For you, I'd recommend playing them, but there will obviously still be postflop factors that can put things on one side or the other of that line.

""Chadders post about non-all-in 3bet bluff"fA few things to be careful about. First of all, the percentage of folds you get when you're 3-betting value hands is going to be a good deal higher than the percentage of folds you get when you're 3betting junk, due to card removal. Second of all, part of your 3bet bluff success rate may be because of a low 3bet bluff rate in the first place, which would change as you started to do it more. All that said, I have been somewhat suspicious for a while now that a higher 3bet bluff frequency is indeed optimal as standard, but didn't really want to teach it out of fear that coaching something I didn't really do very much that could potentially be very wrong to players who are still developing might not be the greatest idea in the world. But my numbers from 3bet bluffing have been pretty solid, which at the very least is a case that if we're not doing it against opponents who are better than most to do it against, we lose out on a ton of expectation.

"JSH hourly rate post"A lot of this is less generalized math and more personal preferences/multi-tabling skills. I can say what I did - I was always in a ton of lobbies, but unregged pretty quickly when I got action above $500. I'd still play $500s and $200s at the same time, $150/$500 seems somewhat close, again going to depend on personal things.I'm a big fan of getting in volume whereever you can. In my 100k month I played 702 games below the $200 level. I think staying regged in a wide range and then unregging as necessary is better than not regging low when you're regged in 1ks but no action yet.

From the marginal hands OOP article chadders:" One final note: Throughout, I've said that we should be playing these hands - that doesn't necessarily mean we should be simply calling them (although that's going to be our most frequent play). Many of these hands, and the hands just a little bit worse than them, are really good 3bet bluff hands, particularly 84s, J5s, 96o, etc. My expectation from 3betting this range is actually much better than calling, but my sample size is a bit too small to make too much of it."

"Are you more likely to go with the tiny sample size and say "it's likely he's 3b jamming wide, so I will mr/call A4o 24bbs" or do you think "it's more likely he's just gotten a lot of decent cards in the first 5-10 hands, mr/call A4o 24bbs is leveling myself"It's clearly somewhere in between, which is the idea behind Bayesian thinking - it could be one, it could be the other, hedge your bets as you get more information. I will say that I basically never minraise/call A4o for 24bb - your opponent has to be jamming a shit-ton before that becomes correct, because you're only marginally ahead of most of his expanded jamming range, and crushed by the standard jamming range of Ax/pocket pairs. I'd much rather raise/call KTo for 24bb than A4o.Like seriously I haven't done the math in a while but it's something like 40-50% that your opponent needs to be jamming.At that point my best guess is that he has an expanded 3-bet jamming range but nothing crazy. KJo becomes a solid minraise/call for 24bb, etc, but not K9o/A4o. We assume it's somewhere in between what we're seeing and what most people do.

85% certainly seems reasonable. My minraising range was lower but I always kind of berated myself for it as it seemed like I was doing just fine opening the bottom of my range. I do agree 92s is better than 32s 25bb deep :) My rule of thumb did come to be making sure I didn't openfold anything suited first hand. Folding the worst is also reasonable though. Since you're good on the numbers end though it's worth taking a peek at your expectation with the bottom of your range 20-25bb deep and seeing what comes up. 

"Shouldn't a hand like Q2s do a little better flatting a 2x than a hand like Q6o?  I would also think that hands like 86o & 96o would do just as well or better than Q6o."Seems reasonable to me."With the hands K4o,Q6o,Q5o,Q4s,Q3s,Q2s,J7o,J5s,J4s,T7o,T6o,T5s,T4s,97o,96o,95s,87o,86o,84s,76o,74s, & 64s 20-30bb my EV bb/100 flatting a raise is -60.21 over 2,370 hands."http://www.youtube.com/watch?v=b-XhyLqC344&feature=related"I want to further expand my readless/vs 50-60% opener flatting range.  I was thinking something like A2+,K3+,K2s+,Q5+,Q2s+,J6+,J2s+,T6+,T2s+,96+,94s+,86+,84s+,76,74s+,63s+,54s.  It's pretty wide but I think an argument could be made for flatting all of these hands.  Do you think it's naive to believe all of these hands could be +EV flats?  If so, how wide would you go?"I don't think that range is naiive. You may also get some benefits from opponents ragetilting at you calling 94s preflop. I also (again) should mention that 3-betting the bottom of this range is often going to have better expectation, even against a 50-60% minraiser. (ok, just read to the end, you said that)It's hard to come to too much of a conclusion about this type of thing because the best evidence would be somebody saying, "hey, I flat all those hands, and I'm +EV/not +EV", but that's not out there. Absent that, how can I possibly say that even calling 75o OOP is for sure no good? The best evidence that makes me suspect very good players can do well with very loose OOP ranges is our border expectation for our current ranges, and the fact that I've seen h2olga play some very wide hands OOP (and tangentially, Serkules, in different ways). That tells me my suspicions that we can play an even wider range aren't too far into crazytown.

If you're really sure of those assumptions, it's a math problem. Plugging in a reasonable calling range tells me that shoving A2o is fine for 35bb (+0.8bb better than folding) . The problem is adapting this understanding to the real world and making better real world decisions from it. I think shading it to the conservative side is perfectly reasonable given those concerns and that your ranges look OK. Naturally you should be 3bet/calling some of those hands.

Seems close readless. Q7s I would say is surely a jam readless, Q2s I would lean towards folding mostly readless. As for whether flatting is an option - I wouldn't take it off the table completely, make sure you've read my article on shallow BB play which talks a lot about this question and how tough it is to answer well.