https://www.pokersnowie.com/blog/201...ff-value-ratio

https://www.pokersnowie.com/blog/201...ff-value-ratio

"Let’s compare PokerSnowie’s bluffing frequencies with the theory

If hero was in a pure polarized range, with hands having 0% or 100%, we should having the following bluff frequency F:

On the river X=680 and Y=720, Therefore, in theory hero should be bluffing F=680/(2*680+720)=680/2000=32%

On the turn since he bet ½ pot, he should have 75%x68% = 51% of value hands

On the flop he should have 75%x51%=38% of value hands"

I'm not sure how the writer is getting these numbers on the flop and the turn. I get why the river is 32%

But why are you supposed to have 51% values and 38% values on the flop and the turn? What's wrong with keeping 68% values on all streets or 1% values on flop and turn only to give up most of the bluffs on the river to have the optimal 32% bluffing frequency?

These are older articles and older ideas of what "theory" dictates (specifically the first Matthew Janda book). I hope that these articles are used as sort of an introduction into optimal play, using the ideas from Janda's first book to transition into what is now the established theory.

The theory of using the value:bluff ratio on streets other than the river is a very flawed theory and is not the best way to approach poker; I really wouldn't focus on it all. Instead, check out Janda's newest book where he revises his original strategies.

What's the newest book? Is it the orange cover NLH for Advanced Players?

What's the newest book? Is it the orange cover NLH for Advanced Players?

I agree with Jarretman that this is probably dated information and shouldn't be strictly followed. It could probably work somewhat in theory but seems more complex than just having a solver derive it naturally or using other measures to build ranges going forward from the flop instead of backward from the river.

OP the article mentions this in 1 sentence but it's basically saying you can take all of your betting range on one street and treat them as your "value bets" on the previous street. Then, on that previous street you can now add bluffs to your "value range".

You can then compound this affect from river to flop to derive a 3 barreling range that naturally has some bluffs that give up BUT preserves your correct bluffing frequency for the next street.

Example:

Imagine I want to bet pot on each street resulting in an all in on the river.

So my value betting frequency would have to be 2 - 1 on each street to make villain indifferent between calling and folding.

Let's start at the river. Let's say I have a total of 10 combinations I know I will want to bet 3 streets for value. That means on the river I will need 5 bluffs.

So on the turn, I will now have 15 "value bets" (my 10 3 street value bets + 5 bluffs).

Again I need to construct a range of 2 "value bets" to bluffs. That means now I will bet the 15 hands plus an additional 7.5 hands of bluffs. Now we'rr at 22.5 hands total.

Now on the flop I take those 22.5 hands and add an additional 11.25 hands. So my total flop betting range is 33.75 hands.

The same could be worked out in reverse. Which I think the article was doing.

Say this time I decide I need to bet 1/2 pot on the flop. That means I need 3 value bets to 1 bluff to make villain indifferent or 25% of my range needs to be give ups.

That means on the next street I have 100% - 25% = 75% of my flop hands that I want to bet, some of which are bluffs.

Now I choose to bet pot on the turn. That means I need 2 value bets to 1 bluff or 33% of my hands need to be bluffs. So that 75% range can contain 75%*33% = 25% bluffs that will just give up on the river.

So now on the river I decide to bet 2 pot. I have 75%-25% = 50% of my flop betting hands to bet on the river. Because I chose a sizing of 2 pot I need 50%*40% = 20% of my flop betting range to be 3 street barrel bluffs and the other 30% to be value bets.

You can see why this fell out of favor/is not really that great.

In the last example (though it is extreme) fully 70% of your flop range are bluffs and 50% of your flop range will be giving up before the river and folding. It just consumes too many good hands and leads to extreme over aggression which leads to so much weakness in your checking ranges and overfolding to raises when you bet.

The math is based on perfectly polarized ranges and it works perfectly. We’re not overfolding to raises on flop just because we have tons of bluffs because we can still bluff 3-bet. The opponent loses EV by raising at any point in this model. Our checking range does not need to be protected in this model because it only consists of 0-equity hands.

The model is not wrong, it just doesn’t apply. In a perfectly polarized range every combo either has 100% or 0% equity. Perfectly polarized ranges practically don’t exist in holdem prior to the river. You might find them in some kind of draw game after both players pat, but I’m not sure.