Beluga killed 3rd reason to bet in his 3rd edition of Easy game book. He wrote this:

"However, Reason #3 always seemed difficult to put into
words for me. I’ve always felt uncomfortable trying to explain Reason #3, and eventually I boiled it down into a simple example. Here is the situation:

 I raise AQ, villain reraises, I 4bet all-in, and while he’s thinking he accidentally shows
me that he’s holding 88. I want him to fold—this is a clear example of Reason #2. I
want him to fold a hand with better pot equity than me.

Now, let’s look at the counterpoint:
 I raise 88, villain reraises, I 4bet all-in, and while he’s thinking he accidentally shows me
that he’s holding AQ. What do I want now?
I still want him to fold. Hm.

So, sometimes I want my opponent to fold the worst hand. This caused me to redefine reason #2 for betting: Reason #2: Bluffing means betting to make your opponent fold a hand incorrectly.

Incorrectly means that if he could see your cards, he wouldn’t fold. Sometimes folding incorrectly adheres to the classic version of Reason #2 (we have J9 and he has QJ and he folds preflop to our 3bet), but other times it simply means he folded a hand he had odds to call with (our opponent folds 6♠7♠ on J♠T♠2♣3♦ to our 2nd barrel with AK). Lastly, it could mean our opponent folds a hand that they could have re-bluffed us with (we 3bet J9o and he folds 76s, but if he had 4-bet we would have folded). This is clearly a much broader vision of the concept of bluffing. Not only does this help us avoid making bets to “capitalize on dead money” which end up being incorrect as either a value-bet or a bluff, but it gives us much greater license to consider bluffing in spots that we might previously have avoided. "

But I don't really get why it was needed to put it under 2nd reason. Does this really change something for us in practice? He says it prevents us from making bets for "collecting dead money". But does this mean it is bad? Are there some situation where having 3rd reason to bet under 2nd reason to bet prevents us from making a mistake? Could someone give me an example?

Another thing, on PSGY there is an example of betting for collectin dead money in the pot:

If you are betting to collect dead money, you just bet to win the pot. You also want your opponent to fold his outs.

Dead money is another term for the money that is already in the pot. At the start of every hand, the blinds are dead money. As the hand goes on, more chips will join the blinds as dead money when players limp, raise, call or 3-bet.



Look at the ace high flop in the graphic above. You raised pre-flop. In addition to the blinds, the pot now also contains the money of your raise and your opponent's call. This is called dead money.

You hold 55. You can't really place a value bet and a bluff bet does not seem wise either. Theoretically, however, both could be possible. Your opponent could continue possible draws that are beat by your 55 and he could also fold better hands like 66 or 77 to a bet.

But none of these outcomes is a primary goal. If you bet and your opponent folds, you won the pot and thus the dead money. Your opponent may have folded a weaker hand with his 76, but he also had 6 outs against you. By betting on the flop, you do not allow him to see a free turn and hit one of his outs. Therefore you force him to give up his outs.

Now, I don't really get this. It feels to me like we are "betting for collecting dead money in the pot" as explained here only if we are against unknown range. But if I DO KNOW he is holding 76s in this exact spot I actually want him to call, not fold. Is that right?

I'm not sure what your actual question is. There's only one reason to bet: Because it has a higher EV than doing anything else.
That EV can come from:
* Getting called by worse (i.e. it's a value-bet),
* Folding out better (i.e. it's a bluff), or
* Protecting your equity by folding out hands that might be losing now, but could suck out, or that might bluff you off the best hand later on (i.e. it's a bet for "protection", or "dead money collection" as Beluga Whale called it).

In many cases, your EV will come from a mixture of value, bluff and protection. e.g. If you raise AK pre-flop, you can get called by KQ (value), fold out 22 (bluff), and fold out J8 and many similar hands (protect your equity from something that has a 35% chance of winning at showdown).

Well I feel I have two problems (questions). I will try to articulate them.

Lets use an example u provided. I raise AK pf and I fold out J8. But wouldn't it be better to be called by this hand instead (so we bet for value)? Does this depends also (if I prefer villain to fold or to call with this kind of a hand against my AK) on sizing of my raise? Where is the borderline between us wanting villain to fold his worse hand and us wanting villain to call with his same worse hand? Belugas example with AQ vs 88 pf is pretty obv to me - thanks to easy arytmetic we can proof we want villain to fold his AQ against our 88 (given particular pf action Beluga mentioned). Or is this just extremely situation dependent and we cannot generalize this in pire vacuum?

Second, what does putting 3rd reason to bet under 2nd reason to bet really accomplish? Does it even matter he killed reason 3?

Make it simple and say you have AA, do you want to get called by 72o? Yeah you do but you raise because you don't know what your opponent has and you want to build a big pot with the nuts. If you close the action and the villain limped in so you're HU and you happen to know his cards and you know he will fold to any raise, OK flat. But you don't know his cards, so you raise for value.

Same with AK except that hand is more complex, it occupies this weird space where it's a monster but also the nut bluff hand.

Well there are two things we don't know. We don't know what villains cards are and we also don't know what is he going to do with them. So one thing is villains range and second thing is what is he going to do with particular parts of it.

Anyway ... Yeah ok, so if he has 72o and he is gonna fold with it, but he will be calling me down postflop if he catches a pair I would ofc just flat with my aces - flat has more EV than raising in that particular situation. So what? We are talking about betting/raising and reasons for our bets/raises. I thing u r wrong - the fact that we don't know his cards by it self doesn't make our bet a bet for value. Or does it?

Yes, in a vacuum, you want villain to call with J8 when you have AK, but then when the flop comes J8K, you wish he'd folded pre. :/
In the long run, very few hands make more than 1.5bb per hand, so pre-flop at least, you'd almost always prefer to have everyone to fold to your open. (The exceptions are something like TT+, AK, AQs, depending on position and skill edge.) If you have the nuts (aces), you definitely prefer to get some action, because the EV of aces can be over 10bb per hand, and you clearly don't win that much if you just pick up the 1.5bb of blinds. This concept carries over to post-flop and it's explained very well in Matt Janda's latest book, in a chapter called something like "Why you don't want action". Essentially, the math shows that if your long term average EV is greater than the current size of the pot, you want to get action, but if your EV is less than the size of the pot, you'd prefer to take it down immediately and win 100% of the pot now, instead of X% of a larger pot.
e.g. If you have top set on K72 in a pot containing $10, your EV is likely to be greater than $10. You can extract more value by getting called (or by inducing bluffs by low equity hands) than by taking the pot down straight away. You might have 93% equity in a $10 pot, but if villain bluffs for half pot with 98s, and you call and see a turn, you'll "own" at least 80% equity in a $20 pot. (On some turn cards, villain will be drawing dead, so you'll have 100% equity in a $20 pot, which is clearly better than 93% of $10).
With a hand like 66 on K72r, however, you "own" a smaller equity proportion of the pot. In some cases you're already way behind (vs Kx), and in others you might have 70% equity (vs QJs for example). If you allowed villain to realize his 30% equity for free (by checking), you risk villain hitting a 6-outer or going runner-runner, and you'd thus lose a 70% share of the pot. So in some spots it would be correct to bet your 66 and sometimes collect 100% of $10, instead of checking and only winning 70% of that $10 pot. It's kind of impossible to work out the EV of every hand, because ranges are wide and there are thousands of possible runouts, but with experience you can estimate whether betting to protect equity will have a higher EV than checking to keep the pot small.
Generally speaking, however, when you're not at the absolute top of your range you'd prefer villain to fold, because 100% of pot is better than 90% of pot or 60% of pot. This doesn't mean, however, that you should get mindlessly aggressive with weak hands. If villain has many better hands (or has a range advantage), the last thing you want to do is bloat the pot, because then you're increasing his EV. You can bet more often when you have the equity advantage and you expect a lot of folds.

Very.good answer (thx for that), except for the first part: "Yes, in a vacuum, you want villain to call with J8 when you have AK, but then when the flop comes J8K, you wish he'd folded pre. :/" .. well if have AA and got called by 72o and flop comes 77K I also wish he'd folded pre - same logic.

P.S.: u mean this book ? https://forumserver.twoplustwo.com/3...hread-1660716/

I'm not saying that we raise for value because we don't know his cards, we raise for value because our hand is an overwhelming favourite vs his range. When that's the case, we need a good reason not to raise.

Ah ok, we agree on this one, but in that case u haven't really helped me with my question (no offense) :/

Well, you said this: But wouldn't it be better to be called by this hand instead (so we bet for value)? , so what I'm saying is yes it would, but since we don't know their exact hand (or that they necessarily fold if we raise), we raise for value.

IMO preflop starts with the fact that we broadly have to build pots with good hands. We also want to bluff because it's free money to get folds. Put that together with the fact that we need to balance so we can't be exploited and now we end up with ranges that we 'have to' play.

Without solving multiplayer NLHE we can argue endlessly about 'correct' ranges but generally say it's free ring, just about everyone is going to agree that UTG we fold KTo but raise on the BTN if unopened.

I think I see where r u going. Thx.

I don't think this is all that complicated. The key is that the 4 bet puts you all in. Let's look at the math of a 55/45 race.

Your EV advantage is 10% of the pot. So let's say you raised 4 BB, got 3 bet to 12 BB and you shoved. If the villain folds, you win 16 BB. In order to make make the math +EV, the pot would have to end up being over 160 BB minus the shove you made. Unless you make a massive overshove, you aren't going to get the advantage of playing it out vs. getting a fold. In addition, 88 just isn't likely to be a hand that you're going to want to play on the flop with someone's 4 bet calling range without hitting a set.

What Beluga is acknowledging is that just because we made a +EV play, it doesn't mean that we prevented our villain from also making a +EV play.