Theory- Merging Your Range

[aejones]

I write and talk about this all the time.

I obviously believe that a two-way bet exists.

This bet is basically merging your range. Your range exists as a spectrum from A-Z, for example, and so does his. In any given situation, he'll bet a certain % on the river- for people not on a high level, they'll bet A-E for value, check, F-0 becuase it has moderate showdown value, and bet P-Z as a bluff. These letters are completely randomly assigned and vary by opponent (note that some opponents will check behind a hand with no showdown value giving up obviously, but leave it as these ranges for the example).

So this guy betting strong hands and bluffs is pretty easy to figure out and play against, because you'll be able to call with an exact range (say, A-P, the strongest part of your range) when you think he is bluffing, and fold your medium strength hands and hands with no showdown value (say F-Z in this example).

However, ranges are continuous (obviously, there are points in hand ranges, but not chunks- that is, second pair top kicker, second pair second kicker, second pair third kicker, are ranked a notch above each other, but there are so many of them that they are nearly continuous). When people make bets, we assign what range of hands they would make that bet with, so that at the end of the hand we have an idea where they are and can act accordingly. That much is very obvious. However, when we get to the river, there are many times that we're not sure exactly where we are. Many people just say "we're lost, we don't know where we are." That HAS to happen at times in poker games; people at higher stakes play their hands in deceptive manners so that when you get to the end, you're just not sure if they're slowplaying or calling you down light.

One simple example is this (note this will be oversimplified, but I believe Jason Strasser played a hand like this): You have second pair heads up, let's assume there was a draw on the flop, and it missed on the river. Your opponent checks to you on all three streets, and you bet 2/3-3/4 of the pot on all of them.

When your opponent calls you down with fourth pair, you win.

When your opponent calls you down with top pair weak kicker, you lose.

Some people check here; and checking certainly can't be terrible. However, it makes you much too easy to play against (a lot of this is metagame related, and very hard to explain as a generalization). Therefore, this bet might be neutral in expected value because the frequency that you get called by a worse hand and the frequency that you get called by a better hand somewhat evens itself out.

So, if the middle of your range is of moderate showdown value on the river (F-O, let's say), and you have the right image, your opponent is on the right level, etc., then you should probably go ahead and bet F 90% of the time, G 80%, H 70%, etc. as the strength of your hand decreases. All of these things have many other factors to consider, and placing this in an actual theory post rather than the experience of merging your range is difficult to make someone comprehend. I mean, some people will read this post and never understand it, and some people do it naturally without understanding reasoning behind it.

So then, when you're merging your range, sometimes you will end up "value bluffing." I believe this to be true, whether or not people agree in terminology. Sometimes you are betting for value, and sometimes you are bluffing. You don't exactly know which you're doing- but if you feel like you beat the majority of his range, you bet (in this case, it is primarily for value). However, since we've already established we don't know his exact range, there are going to be times that we bet and he's got a better hand and folds, and there are times that we bet and he's got a better hand and calls, and there are times that we bet and he's got a worse hand and folds, and there are times that we bet that he's got a worse hand and calls.

It's really that simple.

I mean, here's a hand that is a pretty good example:

You have JJ and raise the button, you get 3-bet, and it's HU to the flop.

The flop is Q Q 7. He bets 2/3, you just call.

Turn is a 2. He bet's 2/3, you call.

River is a 7. He shoves.

Now, normally there will be general thoughts about this hand: pairs < 7 got counterfeited, he might think you're weak since you didn't put in a raise before the river, etc. However, if he is merging his range, then you can no longer just say "here I think he's bluffing, I call," or "here I think he's got it, I fold." The reason is that sometimes you will fold, and he'll be like "LOL I have TT I was shoving for value!" and sometimes you'll call and he'll be like "LOL KK ship it bitch."

The point is, if people are on a high level, they aren't always going to know where each other is. If you're playing against some donks at low stakes, you'll know "people always play KK/QQ this way when an A comes on the board" and "people always check behind when this river comes scary and never v-bet the hand I thought they had- therefore they must be bluffing."

But as you move up, that is not the case. The higher level of thought, the more people will merge their range and stumble upon value bluffs.

Now, where there is confusion is in the intent and the result. The confusion causes the term value bluff to be termed when good players are merging their range.

[MDMA]

Yaya, some people might find it interesting I'm sure, no news to me. I also still hate the term "valuebluffing", since you actually said yourself that betting in that particular spot, you assume your bet is going to be ev+ in terms of getting called. The fact that ranges overlap each other is still of no consequence.

There is no such thing as valuebluffing, the concept is purely a result of people not having PERFECT information, which in turn leads to overlapping ranges. M'kay?

[aejones]

[ QUOTE ]
Yaya, some people might find it interesting I'm sure, no news to me. I also still hate the term "valuebluffing", since you actually said yourself that betting in that particular spot, you assume your bet is going to be ev+ in terms of getting called. The fact that ranges overlap each other is still of no consequence.

There is no such thing as valuebluffing, the concept is purely a result of people not having PERFECT information, which in turn leads to overlapping ranges. M'kay?

[/ QUOTE ]

I agree it's not news to me, but you also will likely agree that even some of the better players don't have a very good understanding of this concept.

As for value bluffing, I use it at times simply because sometimes I'll bet the river for value, he'll tank and I'll be like "call call call" and then he'll call and flip over the winner, and I'll be like "[censored]." My point is, a 'value bluff' can be when you bet for value but get called by a better hand- you were betting for value, but it turns out that in order to get the best result, it was actually a bluff.

It just all gets confusing, we obviously (and most people in this board and HSNL) understand it, but whether or not you want to use that term I guess is just preference. It's a completely random term that is entirely an oxy moron since the two terms next to each other are counter productive. Honestly, it doesn't seem like it makes any sense to argue over whether or not a made up term exists. I mean if ranges overlap as you say, then there are times you won't know whether you're betting for value or bluffing.

[punter11235]

I think the concept is only valuable vs big donks.
As Strasser and the like thinks otherwise there is some chance I am wrong.

[aejones]

[ QUOTE ]
I think the concept is only valuable vs big donks.
As Strasser and the like thinks otherwise there is some chance I am wrong.

[/ QUOTE ]


How is this only valuable vs big donks? I think it's only valuable against good player? Big donks will call/fold with a more precise frequency, thus making you value bet thinner but a more precise range that doesn't neccessary have to be merged with gray area.

[Ansky]

[ QUOTE ]
I think the concept is only valuable vs big donks.
As Strasser and the like thinks otherwise there is some chance I am wrong.

[/ QUOTE ]

It's precisely the opposite.

[punter11235]

Ok question :
Do you mean that sometimes in given situation (board/stacks) you bet same hand for value and sometimes in the same situation but diffrent reads you bet the same hand a bluff ? Or do you mean that sometimes you bet the hand not knowing if its for value or for bluff ?
If its the first I agree it can be a case.
If its the second I think it doesnt have any merit.

(EDIT : so I am probably one of those people who will never get it [img]/images/graemlins/wink.gif[/img] ).

[jfish]

ae,

i like this thread and i think the concept grows more important once you and villain have had multiple encounters. that being said, i think your JJ example on the QQx flop is irrelevant.

[aejones]

[ QUOTE ]
ae,

i like this thread and i think the concept grows more important once you and villain have had multiple encounters. that being said, i think your JJ example on the QQx flop is irrelevant.

[/ QUOTE ]

Yea, I just made up that example real fast; I think in general though situations where it's very likely that both opponents have a pocket pair say, are somewhat interesting, because they tend to be very thin and the relationship to the board is so important (i.e. situations where the seperateion between top card and second card is a lot.. K 5 3 flops, where there are multiple versions of second pair on different notches).

[LearnedfromTV]

punter he's talking about the second case, and like mdma said it's possible because you have imperfect information.

If your imperfect information were of the form: this guy's range is a-z, and he'll always call with a-p, then with whatever hand you have you would either have a value bet or a bluff.

But if either of the following is true:

- villain is planning to call based on factors other than hand strength (i.e. on board xyz he'll call some percentage of the time between 0-100 with a subset of hands as a game theoretic randomizing play). A subset of this case is if Villain will call or fold occasionally based on holding a key card (e.g. nut flush card or straight blocker(s), even easier to understand in omaha)

- our information about either villains range or his calling range is imperfect. I.e. our read isn't good enough. For simplicity, say villain is one of two types and we don't know which other than probabilistically; 60% of the time he has range x and will call with range y; 40% he has range z and will call with range w. Under these conditions, it's possible for a medium strength hand in our range to be both a profitable bluff and a profitable value bet.

For an extreme example of the second bit, imagine playing limit poker against someone who has one of two strategies, but you don't know which: if you bet the river either he'll fold every hand, or he'll call with every hand. The hands you've played so far are enough to tell you that there's a 50% chance he's the calling station and a 50% chance he's a folder. There is a significant portion of your range that should bet without knowing if the bet is a value bet or a bluff.