re-chapter on pinpointing the optimal river bet size

Assume, as you do, that we have a hand that villain beats 15% of the time and that *on this basis*, he bets exactly 0.83 pot. Then of course villain would be able deduce that hero thinks he beats 85% of villain's range and thus villain would have no reason to call often to make hero's bluffs break even.

In other words GTO strategy on the river would certainly NOT involve betting a different amount for each individual hero hand from a calculation derived only by the % of villain's hands that each hero hand beats...

In this nuts air scenario, we jam river with 20 value/10 bluffs and our opponent is indifferent between call and fold (ie we win the whole pot no matter what he does). So then if our turn betting range were just 20 value/10 bluff, he should fold turn 100% since when he calls turn he loses the whole pot on the river. Therefore, we need more bluffs on the turn, or alternatively the river bluffs can be thought of as "value" hands on the turn.

That's my understanding anyways, I could definitely be wrong.

Hey Matt, you said in your first post you wanted to emphasize high equity hands such as k8s as 3 and 4 bet bluffs. Does this also apply to our opening range too? If so, I wondered what you thought of the Snowie preflop advisor's opening range, which seems to favour high equity hands as opening bluffs?

Hello,
an amazing read.

Hand No 1 in appendix Im having trouble understanding the calling range UTG.

Flop Ad 7h 5d Kh Kc.

UTG:
open range includes AKo-AJo, AKs-ATs,KQo.

check call on flop includes AQ, AJ, ATs, Kd Qd-Td.

the full check call range on turn is shown as KQ(3).

This implies check folding on turn AQ, AJ, ATs I think, all stronger than KQ(3).

Am I missing something here ?

Hi Owen,

If I understand your question correctly, you have to take into account that every time we bet the river with a balanced and perfectly polarized range our opponent has "effectively lost" the hand (calling and folding both have an EV of 0).

Because of this, if we were to bet our entire range on the river our opponent would have no incentive to call the turn with a bluff catcher. To keep our opponent indifferent to calling on the turn, we have to check-fold sometimes on the river. In other words, if we can bet the river with our entire range and remain balanced, we didn't bluff aggressively enough on the turn.

I don't think that's correct, as you mutiplied 34.5 by 0.5 twice (and you should only be multiplying it by 0.5 once).

Assuming the formula in the book is wrong, I think someone mentioned this is a mistake earlier and it needs to be updated in the OP. I'll make it a goal to look over the thread again next week and have a mod update it, I've been trying to wait until there are a few mistakes to update before bothering a mod.

There is a connection. The bigger you bet the worse price your opponent gets, so it's more acceptable to check-fold on a later street.

While of course ranges won't ever be perfectly polarized in actual pokers, in general if you're betting 0.5 PSB (such as in a 3-bet pot) it's going to make sense to check-fold less on the following street than if you're making larger bets (such as in a raised pot where you bet 75%).

I think you're looking at this the wrong way.

Let's say I bet 100% of villains range (with my value bets) on the river, and I want to have an overbetting range of nuts/air. Using your logic the opponent would always fold, but you're not properly taking into account that I'm overbetting the proper amount of bluffs to make my opponent indifferent to calling or folding. In other words, he doesn't know when I'm value betting and when I'm not (whereas you seem to think he does in your example).

Correct.

There aren't really "bluffs" in an opening range as the opening range isn't polarized.

That said there's a pretty big difference between K8s/86s/A9o and there's no way to know which one is better. It really depends on positions (and even then no one knows) so unfortunately I can't really comment much on pokersnowie's ranges.

The range we're dealing with on the turn assumes we bet the flop, so AQ/AJ/ATs aren't in our range.

I actually had the question, "Why are we multiplying what we win when villain folds 50% of the time by .50 again in the formula", typed up in my post but deleted it cause I wasn't sure and figured it was a stupid question.

In other words: We win +17.25 bbs when villain folds 50% of the time. (34.5)(.50)= 17.25

Then we plug that number into the equation and don't multiply it by .50 again? So 17.25 + (.50)(-16.4) = 9.05 So our overall expectation for 5bet shoving would be +9.05bbs given the assumptions is said in my earlier post?

Also since the formula in the book was wrong, does that mean the tables on pages 48-50 are wrong also?

Edit: Went through a couple of situations in the tables and looks like the tables are correct.

I remember checking the tables and they were correct. So unless I'm remembering incorrectly (which I don't think I am, since I remembered there being a problem in the formula) all of those numbers should be correct.

Looks like you typed this when I was typing my edit and didn't refresh the page. Yes looks like the tables are correct.

Ah, ok. I get it now. Thanks.

Also, thanks for the response with the sizing.

Phil Galfond said something to the likes that your book is the only one he's read in years 8 years, something like that. And to give credit (to you) where it's do in this video....

Just ordered the book off Amazon.

Is there any information regarding HUD configuration included? Or can someone guide me to where I can find Matthew's recommendations?

Thanks for that PutmyRobeonRite.

No HUD information since it mostly deals with poker theory.

Honestly I would also ignore the pre-flop ranges and look to post 62 to get a better idea of how 3-betting ranges should probably look. Basically pre-flop doesn't model well and my pre-flop ranges ended up being way too polarized. That said, the math used to figure out if something *must* be wrong pre-flop is still really important, I just don't think my best guess ranges from a few years ago are very good.

So how is one supposed to get the most out of the book in this day and age? Read it, and simultaneously go through this thread for the updated information?

How do you get 66.7% for this:?

"For 0.5 PSB, our opponent must call or raise 66.7 percent of the time to keep us indifferent to bluffing."

Nvmd. It says "2 How we arrive at these call percentages will be discussed later in
the book." So I'll just wait for it.

Probably, but the only real section I have a problem with off the top of my head is the pre-flop section so I think you're good to go for most other chapters in the book. I don't think you need to read through this entire thread but reading the OP is probably a good idea and if something seems confusing or if you do have a question feel free to post it here.

If you tell us what stakes and games you're currently playing I can probably give you some additional advice (and others can feel free to chime in as well).

For Sample Hand 1, the flop was Ad7s5d. You elected to call with 6 combos of 87s and 76s and raise with 6 combos of 99 and 88 with a diamond. My question is what made you flat 87s and 76s instead raise them and flat all combos of 99 for instance. You aren't going to call down with any of them unimproved so why not bluff with the 87s and 76s since they have more equity vs an overpair allowing you to bluff a little more?

I like your thought process and that sounds good to me.

You're welcome :-)