There are an increase in threads in the theory forum regarding bet/bluff sizing and bet/bluff frequency lately. Almost always the question boils down to the complication of different ranges for each player given the action and the flop, turn, and river.

The purpose of this thread is to open the subject up entirely to seek the most relevant reading material in on the subject, in hopes of casting the net as wide as possible.

I do not request any summary of the material, although you can do so if you like. I just won’t be debating any of the specific material in this particular thread.

The topic of asymmetric ranges itself might need some discussion also, which would be helpful, of course, by all means.


An asymmetric range is any range that is not continuous linear and uniform, it may be polar or may not be polar. Basically, betting and calling bets in poker creates these types of ranges, and virtually every heads-up flop/turn/river in holdem will have asymmetric ranges.


Thanks,

-Rob

Maybe others are using a different definition, but normally when I'm talking about range asymmetries, I'm referring to how player X has a different range to player Y.
i.e. It's not a toy game where both players have the the same amount of nuts and air in their ranges. In real poker, one player's range (usually the one driving the action) is more polarized than the other (the caller/bluffcatcher), who is more "capped". The asymmetry arises because one player has a betting range mostly consisting of nuts and air, and the other has a calling range of (mostly) mid-strength bluffcatchers.

Quick and basic example for clarity: BTN raises, BB flats, flop comes AKx. The ranges are asymmetrical because only the BTN can have top or middle set or top two pairs. He then uses that knowledge to build his strategy.

Thank you ArtyMcFly, you are very informative as usual. I know you have read more of the latest books, do you have a title/chapter recommendation?

Thanks,

-Rob

I don't think I've ever seen the term in a book. Bob did a thread about it in this sub forum though.

Yes, I have reviewed some of Bob’s threads, and I think we both agreed that asymmetric theory is where the work needs to progress.

I've now spent over three hours trying to make a concise yet comprehensive statement here. Every attempt began by asking myself this question:

How do I best reduce the profitability of my opponent's range?

I think if you can answer that question, then you'll figure out betsizing.

Could ask how do I claim the largest fraction of the pot with my whole range?

I think the answers will be the same.

I believe that B* will perfectly intersect the weighted mean of the potential callers range.

So, if they have more value combos, you size your bet larger to bring the decision into that region.

Even just for the river, and not more complicated early streets, there needs to be a method for assigning ranks organizing all the combos in a range, basically the geometry of the deck.

Just have not read this anywhere.....

[IMG][/IMG]

A rough model of the spectrum of pot share distribution. This isn't equity as Equilab would find, but instead its illustrative of the fact that if both players are playing correctly, then (avg pot share) will always be a positive value for both players; you don't get the whole pot in the long run unless your opponent makes a mistake somewhere or if your range collapses all the way to (nuts on the river) and your opponent's range cannot hold a nut chop. That's the only exception I can think of.

I would think that 0% would not have a betsize and that 100% would have a large betsize.

However it's not as simple as looking at range vs range equities because of implied odds and reverse implied odds as well as the value of protection.

Protection deserves a paragraph here. Ace high board? betting weak pairs for protection seems incorrect to me in all situations. King high board? Maybe some spots for protection bets but not many. Queen high board? More overcards could come = more protection value to be had. See where this is going?

What about boards and range asymmetry that calls for 2 or 3 different betsizes? I would think that certain conditions need to be met for this to happen but I can't put my finger on these conditions.

I think it has to do with the raw number of combos in the ranges; the less combos in the ranges involved, the more we should use one betsize. The more combos in the ranges involved, the less information we give away with the betsize, the more betsizes we should use.

So now we see a difference between these types of betsizes:

a) a betting range that uses one small betsize
b) a betting range that uses one medium betsize
c) a betting range that uses one large betsize
d) a betting range that uses one small betsize and one medium betsize
e) a betting range that uses one small betsize and one large betsize
f) a betting range that uses one medium betsize and one large betsize
g) a betting range that uses three betsizes
etc.

From here, we would need some preflop action to put the wheels in motion to determine betsize.