2:38 bottom left you fold KJ73s to a button minraise

The AKQQ fold is normal for me.  The KJ72s vs. minraise isn't, and probably was due to me being distracted/talking.

I don't consider myself a preflop expert though, so I could be convinced to change my mind on several preflop spots rather easily.

I think you really need to raise AKQQ

JT98d probably shows more of a profit in that spot than most AA combos.

1) That's probably not true

2) That's probably not relevant to whether open folding this hand is standard or not

3) Also, our hand is QQ, not AA

Phil, I'm a bit surprised about this fold as well.  We have blockers to KK/AA and we can flop quite well, although it hurts not to have a suit.  I know you might say preflop is not all that important, but would you mind elaborating a bit on why you think this is a normal fold?

1) In some spots, AA is more profitable than JT98.  In other spots, the other way around.  Thus, it's rather a blanket statement to say "JT98 is more profitable than most AA combos" because it depends on so many things.  For one thing, those most AA combos can make you lots of money if you can get a lot of money in preflop when stacks aren't too deep.  You can do that too with JT98, but the profit is gonna be less.

2) Even if it were true that JT98 is more profitable than most AA combos and that hand ranking is not of uttermost importance when we're deep and OOP (which I agree), it doesn't tell us anything about what we should do with AA.  For one thing, it is possible that raising AA (or AKQQr here, for the discussion's sake) is indeed less profitable than raising JT98 in this spot, but still more profitable than open folding.  

3) Also, I do believe that most hands ranked within the top 10% by PPT is playable/raisable UTG - even though hand ranking isn't 100% relevant, it still means something.  I might be wrong here, but it would take some explanation (hence the question to Phil).

In this spot, there's not really any hand that he'll be folding on this turn.  If he opted to play bottom set this way on the flop, or top two, he'd likely just shove into me on the non-spade turn, especially if he's picked up a straight draw.

If he shoved for pot, it would be a very tough spot.  It would be player dependent.  My range looks like it's somewhat low on straight draws.  It's mostly comprised of top set, top two with backup, combo draw with NFD.  I don't play a bare wrap this way.  Because of that, I fear that I'll have to fold the Jo turn too often if I'm not stacking off with top set.

All that said, it's not as if he's going to structure his flop range to prepare for an offsuit J turn so that he can bluffshove with poor equity.  I guess my point is that he has good incentive to shove this turn with his whole range (at least with stacks the way they were in the actual hand).  We'd just have to see how my equity is against that range (which requires a lot of guesswork).

With over 1psb left (like 1.4), things change a lot because he's no longer going to shove into us with a wide range (presumably).  Almost surely not with the hands we crush like KQ and QQ.  I think at this stack depth we can fold against most opponents (even with our bit of implied odds on the river).

When we get to 2+ PSB, we can be very confident they turned a straight, and we have to decide if rivers play out well enough for us to call again.  It looks a lot like we have a set here, so I don't think our implied odds are that great, and I fear that he will correctly pick us off on spade rivers (or have spades himself) at a reasonable frequency.

Phil,

This might sound like a rather crude idea, but imo a good starting point would be counting the flops that are favourable for us.  As far as I know, Buzz from 2p2 O8 forum had used this method in the past.  I will write them down with the risk of stating the obvious. 

For example, with AKQQr we would to flop a set (Qxx board, although AQx and KQx will do it too), straight (AJT, KJT, QJT, JT9 that are not monotone), wrap (JTx), top trips/boats, top two, and some flops where we can use QQ or As blocker, etc.  Add all these flops see how often they come by.

Then, I'd try to estimate how often we will get 3b by anyone.  Very roughly speaking, we can ask PPT how often we will run intk at least one "$3b8i" range(s).  With hands that you plan on raise/folding, this is actually quite important.  With hands that you plan on raise/calling or raise/reraising, this is of lesser importance.  Here, I'd think having both A and K blocks a lot of their 3betting range, and we have lower risk of getting 3bet.  We also open from UTG, if anyone at 25/50 would find it a reason to 3b against less often.  

I think you would agree that AKQQss is a very easy open in this spot.  Well, the major difference is that we can flop a nutty or near nutty FD with it, which adds a lot of playability.  However, we flop a FD only ~11% of the time, and it's not exactly easy to get ton of action from worse FD when we have both AKss or AQss anyway.  And when we have KQss and the action gets heavy, somebody is likely to have the NFD or NF blocker.  So all in all, I don't think it makes a world of a difference, although I would prefer AKQQss all day. 

I am writing on my phone and would need to check some numbers with PPT, which I might do later tonight.  Just wanted to throw in some quick ideas for now.  Hope I am not saying something too obvious..

- midori

I think this is a great way to go about it.

So if my blockers change each player's 3-bet frequency to 5% (very rough), that gives us 22.6% of the time that we get 3-bet (.95^5).  This doesn't account for people lowering 3-bet frequency once there's a flat-call.

I also don't know what the 3-bet range represented by that macro is, but when we raise from UTG I would expect far fewer KKxx and AKQ7 type 3-bets and more KJT7ds T876ds etc.  We may not block this range as much, but since you're using an IP 3-bet macro I'm guessing it's not too far off.

I'm thinking that 20% is a reasonable estimate because the blinds won't be 3-betting nearly an 8% range I think.  We drop to 18.5 if we give them 3b %s of 3 and 2 (with our blockers) and keep everyone else at 5.

One small issue is that I'm ~400bb deep with BERRI SWEET (the most aggressive 3-bettor at the table), and much of his 3b range will be unblocked by my hand.  I could see him 3-betting over 15% in this spot (ignoring blockers), and maybe then down to 9% or so with my blockers (more rough math).

Anyways, moral of the story is that I'll be 3-bet somewhere between 15-28% of the time.  If the table was full of 300+ BB stacks IP against me I'd feel differently, but I'm confident enough in you guys that this is an open.

I overestimated the frequency at which I get 3-bet, which I do think makes a very large difference.

Just one thing to point out - Flopping TP with this hand is usually not very exciting (though we will take down the pot uncontested a reasonable amount of the time).  Flopping Trip Aces or Kings is also not as wonderful as it may seem.

If we go with 25% as the frequency (on the higher end of reasonable) at which we get 3-bet and 20% for the frequency at which it folds around...

Total EV = .25*-3+.2*1.5+.55*x

x = EV in single raised pots

I'm getting that x=.82bb for the breakeven point

This means that if we raise and get called by IP, pot = 7.5bb and we've invested 3bb.  This means we'd need to recoup 51% of the pot.  If the BB calls, we'd need to get back 3.83 out of 6.5bb, aka 59% of the pot.

Having a reasonable equity advantage (but playability disadvantage) when called by IP, I think we can recoup more than 51% of the pot (though not by a huge amount imo and I'm not confident in this).  IP against BB I think we can recoup 60% of the pot, so we should be good.  I don't have the time now to look at multiway pots and factor it in, and I'm curious what people think about our ability to recoup 51/59% in HU pots (or, how much we actually recoup).

When you adjust the % that you get 3-bet pre, the numbers change somewhat quickly.  I'm still confident that there are some lineups in which this is a fold, and if you gave everyone 200-400bb here, I think it would be.

I just think that if we have a top 5% equity hand, playability shouldn't even be an issue.  If we had a top 15% equity hand and were aiming for an 18-22% RFI UTG then I can see having the discussion.  

I also don't mind AKQQ as part of our UTG range, our A/K do some good blocking so that our overpairs can get really aggro.  And our range is so FD heavy that I'm fairly ok using my high blockers for value/protection bets on flushing boards.

All good points, especially the fact that we can more credibly use our flush blockers.

I guess I just feel that equity runs so close in PLO that playability becomes a much bigger part of the equation.  The best we're doing here (after removing AAxx) is around 60%.  If we were playing Limit Omaha High, that's very exciting news (that would be such a terrible game), but I just worry that whenever we don't flop JT or Qxx we have a weak semibluff/bluffcatcher the majority of the time.  I don't know how that translates into our share of the pot, but it doesn't feel good to me.

Again though, I'm only arguing at this point that it's perhaps closer than a slam-dunk raise.  I'm not disagreeing with you.

Phil,

Thanks for your reply, and glad my approach made some sense. :)  Below are quick thoughts of mine:

1) I think 15-28% is a good estimate of the frequency at which we will get 3-bet.  $3b8i range macro, as you suspected, includes lots of "smooth" hands like T876$ds, which are "only" top 19-20% hands in the default PPT ranking.  Because of this, I would think our blocker effect isn't too off, although it might have been slightly overestimated.  

2) $3b8i was just a rough example that I chose, and if we were to be precise we could use different 3betting ranges for all 5 players, for example $3b4i for HJ, $3b6i for CO, $3b8i for BTN, $3b2o for SB/BB, something like that.  But again, I think ~20% is a good estimate.  

3) At first glance I thought that referring to your DB might be a good way to get a feel for these numbers (Q: How often did we get 3b when opening UTG?) with some stack size filters.  However, I'm not sure how accurate that would be. 

4) As for recouping the pot, once again, I think it might help to look at the HM2/PT4 database (filter for HU pots when you are UTG vs IP and OOP) and get the bb/100.  Of course this tells us nothing about how we fare with this specific hand, but we can expect that the numbers won't be too far off.  How much of the pot we can recoup postflop is rather hard to quantify from the scratch. :(  With our equity advantage (and positional advantage against IP), I think 59% is something we can shoot for against the blinds, if not higher.  Against tough opponents IP, I'm not sure if we can achieve 51%, though.

5) After looking at all these numbers, I'm actually not too surprised that this can be considered a fold.  I don't have any experience playing against players of this caliber, and am not qualified to comment on that.  Let me say, I still think this is an open in most cases but there are spots where the EV of folding seems to be at least equal to that of opening.  This is mostly because we have a rainbow/badugi hand, which takes away a chunk of our equity advantage and forces us to fold to most 3b ranges. 

Again, thanks for taking the time to post this! 

- midori

Good call!

In my DB (small sample of UTG opens because my new computer only has hands from this year), I get 3bet 21% of the time when opening UTG.  Like you said, it'll vary based on the makeup of the game and stacks.

The sample for raising UTG and getting called IP (going HU to flop) was tiny/unusable.

I filtered for all hands which I opened and got called and went HU to flop OOP.  EVbb/100 is 4.5.

Seems very low to me, right?  Could be a sample issue (1250 hands), or I could be opening too wide.  I could also just play bad :), but I do pretty well in these games so I don't think that's it!

Phil,

I looked at my number (not that it's super relevant, but just for the reference) - it's 64bb/100 over 1750 hands including MW pots, and 81bb/100 over 977 hands for HU pots only.  Again the sample size is negligibly small and the fluctuation is expected to be huge, but I think 4.5bb/100 of yours is massively under-represented.

V

If we go with 25% as the frequency (on the higher end of reasonable) at which we get 3-bet and 20% for the frequency at which it folds around...
Total EV = .25*-3+.2*1.5+.55*x
x = EV in single raised pots
I'm getting that x=.82bb for the breakeven point
This means that if we raise and get called by IP, pot = 7.5bb and we've invested 3bb.  This means we'd need to recoup 51% of the pot.  If the BB calls, we'd need to get back 3.83 out of 6.5bb, aka 59% of the pot.

I don't think that I can agree with these assumptions. You assume both high 3-betting frequencies and that you have to fold to all 3-bets.

First off, the first player is a half stack, so he can probably only 3-bet a tight range for value here. Around 7% 3-bet from him and you can 4-bet get it in profitably:

13638/194580 = 7.0%

Against a range with less 4%!2% hands and with the same effective frequency our equity goes up even more:

Second, against BERRI SWEET, you said that you could see him 3-betting around 15%, and 9% after your blockers. No hand has this strong of a card removal effect. Even AAKQr only lowers the following 14.2% 3-bet frequency to 12.7%. AKQQ lowers it to 12.6%. This is the range:

As you can see, you will have over 53% equity against such a range. This should be enough to warrant a call, especially with your nice blockers. Given that you will have many suited hands in your range, you should be able to rep nut hearts very well.

And lastly, versus the blinds you can call deep in position. This should be profitable with your hand.

Concluding, I don't think that you should be folding to most 3-bets. And therefore your EV when getting 3-bet will be higher than in your calculation.

Hey Phil, 1250 sample is way too small.  Looked at my NLHE winrate with these filters ev43bb/100 over 800 hands standard dev. 165bb and with 70% confidence we'll run + or - 58.34bb/100.  95% confidence + or - 116bb/100.  And it's even more extreme in PLO. 

So your true winrate can easily be 50b/100 just have been running bad in that spot. 

@seven7s:

1250 hands make a pretty small sample, but it's not really easy to gather lots of hands in these spots.

Let's say you played 1M hands.  Of these, you got dealt ~167k hands from UTG.  If you are opening about 20% of these, that's 33k hands.  Now, for you to play HU, you shouldn't get 3b, shouldn't get 5 folds and take it down preflop, and when you get called, you should get only 1 caller.  This just doesn't happen very often.  Say it happens ~50% of the time (which I think is an overestimation), and we're already talking about 16k hands at most, out of 1M hands.

Not to mention that I would doubt anyone has played 1M hands at 25/50 so far..

Yes I know, that was my point.  Filtering for specific winrate spots isn't going to yield any useful data really unless you're either winning absurdly or losing absurdly beyond what is probable. 

Yeah, that was kind of my point too.  Phil's 4.5bb/100 is very likely an underestimated winrate.

GT - To be clear, I don't disagree with the open.  I've always said that I think preflop is a bit overrated (though as games get tougher, it will matter more), and I have never put much focus on it.

I feel pretty strongly that we'd be losing by calling BERRI SWEET's 3-bet OOP here, however I don't feel certain about it.  I don't think that 53% equity means all that much in terms of what our actual  EV is.

There are a lot of very strong players that feel differently than I do about how much equity you can realize with holdings that have 'poor playability' so I'm sure you're not alone in disagreeing with me.

There are a lot of very strong players that feel differently than I do about how much equity you can realize with holdings that have 'poor playability' so I'm sure you're not alone in disagreeing with me.

Against a tough player like BERRI SWEET you likely need to call with a range that has good board coverage and card removal effects. AKQQ fits very well in that range, for instance you block AA/KK and hold a strong bluffcatcher at the same time on boards that are good for AA/KK like Axx/Kxx/AKx/x22-x55.

And of course how you play AKQQr UTG is not going to affect your winrate by much, since it only happens 1/6*12/270725 times!

Thanks man! Hope you'll enjoy the rest of the series too!

Yeah, this was an error.  Perhaps I make more mistakes than I realize while recording a live video, but I think it was likely the combination of recording live & being underrested.

Interesting comparison. AA92r has a bigger equity advantage than AKQQr, but it improves on less flops. But when it does, a set of aces is always better than a set of queens. However, the PPT rankings are very similar:

I'd always open AA92r because you don't need to fold to a 3-bet.  Whether you can 4-bet is dependent on stacks and ranges, but you don't run into the issue of having low equity/visibility against a 3-bet range.

As you could see in the rough calcs above, folding to 3-bets is the main thing working against the AKQQr.

GT: Good points, and I edited my original post accordingly.  

Phil: That's true, and assuming we at least break even in a 3-bet pots (otherwise we wouldn't have called) with AA92r, it will change things by quite a bit.

I rarely c-bet overpairs on a board like this because I'll end up checking 2/4/5/7/8/9 and being in a very ugly spot (and the board pairs aren't fantastic).  I think I don't have a ton of fold equity, except for against very weak hands that my AAK2 would be doing pretty well against.

I think in hindsight I should've led this turn.  I have enough FE and it's not a total disaster to bet-fold a hand with (usually) at most 8 outs against his shoving range.  With KQJT, for example, I'd hate to bet-fold or bet-call and would likely just x/c.  

You heard me say in real-time that I thought my hand was too "middle of range" for a bet, but it's actually close enough to the bottom of my range & with the right amount of outs for a play like this (enough that we have a nice semibluff but not so many that we hate to bet-fold, especially because we can't really check-call anyways).

That's interesting because that seems to suggest that you x flops v. often in this situation as the flop is T high or lower 32%+ of the time with card removal and presumably you are often checking flops like JT9 or QQ2 quite often.  And while I agree that his folding range doesn't do very well against AAK2 it probably does much better against your particular holding and you don't need to bet many overpairs to allow you to bet hands like your current holding and push him off equity.  

Also while he has hands like T6 and 542 more often than you do I think most players generally don't have them very frequently as a % of total range which limits how often they can push you off equity. 

I guess I am saying I would like to bet overpairs some of the time and be able to bet hands like this rather than play a strategy where I am checking hands like this very often because even though I rarely have the nuts I think the equity gains are worth it compared to the times I lose equity because of the times he has/reps the small but extremely strong part of his range.  I wonder do you feel very strongly about this as I am curious as to whether I should reevaluate this strategy against stronger players.  

A flop of T82 or T42 would play much differently for me.  I think that the IP player gains a lot from his position (and the weakness of our range) when there are so many turns that put up a potential straight.

Even if he doesn't have it at a high frequency, it gives him a lot of leverage and allows him to bet his weak/strong draws and all 2 pair with impunity as it's very hard for me to find x/r hands.

Good question, Zach.  I'm not ultra confident in my fold (or any preflop play, for that matter!).  My reasoning in realtime is looking at my equity vs his open range and realizing that in a best case scenario (BB folds), 3-betting nets me $50-100 maybe?  It's hard to come up with a frequency of the BB VPIPs and what my EV goes to in those cases, but given that this is a somewhat unique situation, I wasn't exactly "prepared" for it.

I tend to shy away from low-reward best case scenario spots, just assuming that it's difficult for folding to be a significant mistake and that it's very possible for 3-betting or calling to be significant mistakes.

While I still don't know how I feel about this specific spot, I think my approach is a good one to take whenever you're entering uncharted territory.

Ty man.  Yeah it's very hard, and it's so gross when the river improves him and you call it off.  Still, I just didn't think many hands made sense for him other than overpairs or maybe flopped gutters with no pair.

Given that he didn't bluff this river, I am second guessing my plan though.  Some people think KK/QQ is too strong to bluff with here on the river and if he's one of those people, all I did was give him a free 2 outer in a very large pot.

This specific spot, once again, I'm not too sure about.  MOP is a very strong player with a tight CO opening range, and we're 400+bb deep.  I probably could still get away with calling but I'm unsure.  I definitely always play at 100bb.

 MOP is a very strong player with a tight CO opening range, and we're 400+bb deep.

Can you explain why you would play this 100bb deep but not 400bb? Doesn't the fact that we are super deep increase the implied odds of this hand? I feel like we always know where we are in the hand.

Well, I guess i'm not surprised this wasn't mentioned since phil actually HAD AKKx so two people have aces was impossible.  But, I guess I'm hoping someone could comment on the spot MOP is in with aces, or the general concept of two people having aces.  It was funny to me when 30 seconds later you got 2 buyins in against two people with aces.

thanks