Hey Matthew. I have 2 of your books but can't start reading them because i hear some stuff about big players disagreeing with you. I came to this thread to read what OtB_RedBaron said but can't find it. Is it possible for you to address this or point me to somewhere where you have already done that?

OTB posts as LorenzoVMatterhorn. His post is #60 ITT, Janda responds afterwards. This is all 5 years old now so some of this stuff may be outdated.

Thaaaaaaaaaaaaanx

Thankyou Very Much! Do these calculations only work for perfectly polarised ranges? How do the solvers work out flop bluffs?

The solvers are generally free of cost for turn and river and some example flop is free also, and likely there is some Youtube etc. material. Some maybe still simple solver is cheap like 75 dollars or so but I didn't need it further than the free version and if I need I need one that can set the situations as I want.

Personally, I am not interested of the details that are of no practical use for me. E.g. one likely needs nuts and pure bluffs but anything less than nuts and anything more than bluffs might cancel each other out but on the river. Before the river, it though has been calculated in, the outs one has, that maybe should not be calculated in as one is not betting with the nuts all the time, but on the other hand the non nuts also have outs. The middle hands might be played both ways but also make one less nutty and less bluffy. Popcorn, anyone?

I own a solver and use it regularly I just would like to know how it decides on its Value to bluff ratios when I use different sizes on all streets

Have you read the new book?

I don't like the terms "value bet" and "bluff" much anymore, unless we're on the river. Real poker is a lot more complicated.

I have read the new book and enjoyed it, and i agree with the sentiment that they are bad terms, however im curious how a solver for example works out the ratio of high and low equity hands that are need to be bet. For example draws to pairs

A solver doesn't work out any sort of ratio.

I have a superficial understanding of how solvers work, but not to the point where I can explain it well. But basically, at least in the case of PokerSnowie (which is not a solver but AI), it's trial-and-erroring it's way to the best possible strategy.

Thankyou very much appreciate you taking the time to respond!

Hey Matt - I have both your books and they are my "go-to" reads when studying poker. I also loved the GTO video series you did with Ryan Fee.

In your first book you mention that hero should defend a bit wider when facing a flop bet in position because villain has bluffs that can improve on the next street. I've been able to see why this is the case from a math standpoint where I compared the calling frequency when villains bluffs have zero equity to when the bluffs have some equity (say 10% equity). The math says you need to defend more when the bluffs have equity, so cool !

My problem is that I can't explain qualitatively why this is the case - how would you explain this to someone who has no background in math ? I retain things better when I can actually visualize why this is so.

Thanks and keep your brain in this game you are a great resource !!

If you want to prevent your opponent from "bluffing too much," you need to call enough so that when they make bluffs with too weak of hands they lose money (or at least break even).

In some situations, a hand with 20% equity may be too weak to bluff (and should instead check-fold). In others, a hand with 10% equity may be too weak to bluff.

If you want to prevent your opponent from being able to profitably and recklessly bet in situations where almost every hand will have some equity (say on a 9d7d5c board when both players' ranges are very wide) then you'll need to defend more aggressively by some combination or raising and calling. In other words, since there is more value in getting to see the turn card when betting KsTs on a 9d7d5c board than getting to see the turn with 6s5s on a Kc2c2h board, you have to defend more aggressively on the 9d7d5c to prevent betting the KsTs from being +EV.

Thanks that helps me a lot !

Hi Matthew,

I'm not sure if it's been answered already I've searched the thread and not getting answer I wanted.
On page 111, you say that when you bet 57% of the pot you should be betting for value 73% of the time. But should our minimum bluff success rate not be 36% with this bet size, meaning we have 67% value bets, not 73%?
Enjoying the book, thanks.

Matt:

Great book. I feel after much mind questing and pondering I've understood just about everything within it's pages BUT I was hoping to have one concept you talked about in the book cleared up to help tie the whole room of concepts in my mind together. Like Lebowski and his rug I feel Im on the cusp of discovering it, or in this case what it means.

It has to do with the concept of You Don't Want Action. You mention "it's important to understand how rarely you want action because this helps you feel more comfortable betting and raising aggressively with less than stellar hands."

Am I overthinking it thinking I may have missed something or was this statement about illustrating it being okay being aggressive with less than stellar hands in which you want to Deny Equity with essentially

That is indeed how I understand it. Let's say you have a "strong but vulnerable hand". Well, you want to deny equity first and foremost. That means that you will bet more often and on the bigger side, and you are happy if they fold. If they call it's fine since you got more value than if you had bet on the smaller side. But you're better off when they fold.

It's a bit contradictory to how I learned poker in the first place, because we wanted to get "value" from draws back then. But Matthew's arguments are pretty strong. The pot, whatever its size is, is worth winning anyway. Winning a pot of size X 100% of the time is better than profitting 150% of that pot 60% of the time (not counting scared cards, opportunities to lose to a bluff later, etc). His example on the preflop 3bet (I think it's 8bb to win 4bb) is on point.

This doesn't apply to super nuts like when you lock the board with nut full or hands with redraws and stuff like that though, where we dominate so much that we likely want to slowplay some of the time or bet on the smaller side. But those are the rare cases.

Edit : also it's in NLHEFAP, I don't have Applications so I don't know if it's in there as well.

I'm confused.

How often your bluff needs to work and how often you should be bluffing aren't the same thing. Example: If you bet $50 into a $100 pot on the river, the bluff needs to work 33.3% of the time. But you should only be bluffing 25% of the time. Does that help answer your question?

I don't think you're overthinking it. Sounds like you get it to me and I think OMGLillianLee's explanation is good as well.

Sorry I think I phrased it badly. Point 1 on page 111 says "the size of the bet relative to the size of the pot determines how often you bluff". Then at the bottom when we bet 57% of pot we are betting for value 73% of the time. Using the first equation I figured we should be betting for value 67% of the time. Does this 73% include our river bluffs too because these are considered wins, which brings the % up?

You're confused somewhere, so let's start here:

If you bet 57% of the pot, what odds is your opponent getting? When he calls, how often does he need to win for the call to be +EV?

I get that's kind of your question so I'll just answer it here:

Imagine pot is $100 and he bets $57. So when villain calls, he's risking $57 to win $157.

$157(X) + (-57)(1-X) = 0
157X - 57 + 57X = 0
214X = 57
X = 0.266, or 26.6%

So villain needs to win 26.6% of the time on the river when he calls. If we want that to happen, we need to be value betting 100% - 26.6% = 73.4%.

I'm not positive, but it sounds like you may be getting two concepts confused: How often our river bluff needs to work, and what % of our river bets can be bluffs. These aren't the same thing.

Ohhh yes, it's seem very obvious now, don't know how I was getting so mixed up before. The odds we offer him we should be value betting the inverse. For some reason I thought the odds he was getting was 2:1. Think the half pot bet threw me off. Thanks for the reply much appreciated.

Hi all,

first post yay

I think it has been asked before, but I got stuck at deriving the formula on page 337 EV_bet - EV_chk.

I have no idea where I'm wrong but my thinking was like this:

EV_chk = P(win) * Pot

EV_bet = Pot * (1-P(call)-P(raise)) + P(call)*[P(win|call)*(pot+bet)+(P(lose|call)*(-bet)] + P(raise)*[-Pot-Bet]

Any ideas?

Thx

Konstantin

Found it... So for anyone interested:

MDF: Min Defense Freq = 66.70%
Pc*P(w|c): %Range villain calls and wins = 10%
Pr*V% = %Range villain raises for value(=win) = 5%
Pr*B% = %Range villain raises as a bluff = 2.1%
Pr: %Range villain raises = 5%+2.1% = 7.1%
Pc*P(l|c): %Range villain calls and loses = MDF-Pr-Pc*P(w|c)

EVs from hero's perspective:

EV_check = Pot * (1 -Pc*P(w|c) -Pr*V%)
EV_bet = EV_fold + EV_call + EV_raise

EV_bet = Pot*(1-Pc*P(w|c)-Pc*P(l|c)-Pr) + Pc*[P(l|c)*(Pot+bet)+P(w|c)*(-bet)] + Pr*(-bet)

EV_bet - EV_check equals the formula from p.337:

(bet)*(MDF-Pc*P(w|c)-Pr*V%-Pr*B%) - (bet)*(Pc*P(w|c)+Pr*V%) - (bet+pot)*(Pr*B%)

where B% = (1-V%)

It is strange though as this indicates that we lose only our bet when we bet-fold instead of losing pot+bet....

Honestly, I'm pretty lost, but I think your confusion is "relative to checking."

So if you would have won with the best hand by checking, but instead bet-folded after your opponent check-raised, you effectively lost not only your bet but also the pot (since you would have won by checking).

Are we on the same page for this?