I am also interested in it. If anyone could sent me a copy of it i ld very much appreciate this

cheers

Hi matt,

I have been doing some b vs bb range constructions (using your method). And i came to the conclusion that the cbet % of the b in these cases is often quite low (around 20-35%). This has to do with the fact that our range contains less 2 and 3 street values hands as opposed to other position where our starting range is so much stronger. Is it ok to cbet at such a low freq?

Or is it better to increase the bluff ratio in these type of spots?

b vs bb = button vs bb?

Button's range is wayyyyyyyy stronger than the BB and has position so the button should be continuation betting at a high frequency.

I don't really look at hands as "value bets" or "bluffs" unless ranges are very polarized anymore or on the river. This is addressed in a lot of detail in the new book.

yes button vs big blind. I looked at the example hands in your book. And tried some sample flops. Constructed a value range, caculated the nr of combos in them and multiplied that with 3 ( 3:1 value to bluff ratio as was advised in your book). But even with such a high bluff ratio we cbet at a really low freq on a lot of flops.

in the section, Examining Complex Ranges -- Defending enough against opens, there is an equation:

(12.5)(X)-(16.5)(1-X)=0

X=0.569

where does the 16.5 come from? I thought was talking about the button 4 betting to 19 big blinds.

That would be how much you'd lose if you 4-bet to 19BB and lost (19-2.5 = 16.5) relative to if you just folded now and lost the 2.5BB open.

Hi Matthew a quick question regarding defending preflop:
Let's say we are on the BB facing a minopen from BTN. If we call we're investing 1BB into a pot that's already 2BB + 1.5BB = 3.5BB (BU's open + both blinds). Our pot odds are 3.5:1. We would break even here by winning this pot on the flop once for every 3.5 times we lose it... but this doesn't account the money won/lost postflop. So if we are defending as wide as we should but x/f way too much on later streets we could be hurting our winrate by a lot.
What I'm missing?

Doesn't sound like you're missing anything.

Being OOP with a crappy range sucks and you need to be careful not to bleed too many chips post-flop. That's why it's ok to have such a high check-fold frequency post-flop OOP (you're OOP with a garbage condensed range against a tighter linear range). But 3.5:1 is a helluvathing, hence you should still defend quite wide in the BB vs a button open. So it's a balance between a few concepts and you need to find the sweet spot for your pre-flop calling range.

^ ty a lot

Hi Matt,

I was curious. Does it make sense that if hero has a range advantage over villain that:

1. as defender:we should/can defend even wider then 1-a?

2. as the raiser we can increase our bluff ratio?


Same goes when we have a range disadvantage.

1. as defender: We should now defend less then 1-a.

2. as raiser: we should lower our bluff ratio.


Curious what your thoughts are on this subject.

Hi Matt,

i've start reading your book and its great. Really like your reasoning and how you approach different problems.

BTW, im confused about a concept that you illustrate in the first part of postflop play, i.e when we defend in position given oppo is betting.

The conclusion that you arrived is: the more likely our bluffs are to become the best hands (i.e they some equity) the wider the defending freq must be.

My problems are:


1) If im reason in the following way it makes sense to me that we have to def more: since oppo bluffing range has some equity he his actually risking less for winning the pot (since he has some fraction of the pot that is the equity of his bluffs), therefore he has to obtain less fold to make an instant profit and therefore we have to defend more.

2) BUT if im reasoning in this other way i get confused: hero out of position OTR and check, oppo bet and the question is how mucj hero must defend? IF oppo's bluff as some equity @ showdown i have to defend less than the classic formula: 1-alpha (where alpha=s/1+s and s is the bet in term of the size of the pot) since:

bluff has no equity:

B: villain bet
P: pot
F: hero folding freq
S: stack size

For making villain indifferent of bluffing(caller POV):
EV(check)=EV(bet)
EV(check) = EV(bet)
S = (S+P)(F)+(S-B)(1-F)
F = B/(B+P)
1-F = calling freq for hero


bluff has some equity E:

EV(check) = EV(bet)
S + E*P = (S+P)(F)+(S-B)(1-F)
F = (EP+P)/(P+B)

Clearly by comparing the two final equation for F we see that in the first case we fold less (so we defend more) while in the second case we fold more (so we defend less).

My confusion arise because the two result from my two reasoninga above are in contradictions. My intuition is that im doing a really dumb thing and im comparing two different objects, that are: in the first example we are considering the equity of improving to a best hands while in the second one im using the show down equity that is different. So my conclusion is that before the river, i.e when oppo bluffs can improve to best hands, we can defend more (if we are the caller) and bluff more (if we are the bettor). While if oppo is bluffing hands that have (significant) show down value we should defend less (fold more). Moreover by doing this, namely to fold more, we given an incentive to oppo to bluff more and so reaching the right hands to bluff.


I would like to have your opinion on all above since i would like to riconciliate the two reasonings. Could you please help me?

Really appreciate you time and thanks in advance.

1. Your opponent frequently should not be able to make an immediate profit with his bets and almost (if not literally) never with his raises. This is true even if your range is weaker than his, but it depends on lots of things other than just range strength.

2. When you're raising all that matters is the strength of your raising range. The stronger your "value raises" (the strong part of a polarized range) and "bluff raises" (the weak part of a polarized range) are, the greater the % of your raises can be bluffs.

This is only true if your bluffs shouldn't make an immediate profit, which is very frequently NOT going to be the case.

Example: If I min-raise open the button, big blind calls, and the flop is the 9h7c3h. Here, not only do the terms "value bet" and "bluff bet" pretty much suck (most hands are so far from 100% or 0% equity the terms aren't useful), but the button range is so much stronger than the big blinds and we have position that every bet should be +EV. So just because when we bet here with Kc8c and have a lot of equity vs the big blind calling doesn't mean he should call wide. In fact, if the big blind calls here he's going to get crushed.


Ok, so it sounds like you're confused and doing what I just suggested you not do. Do NOT!! try to prevent players from being able to profitably bet any two cards against you when they have a stronger range than you and position.

Just clear your mind for a second, then think of this:

1) The better villains "value bets" and "bluffs" are, the greater their betting frequency should be. If you're OOP, this means your strong hands will make more as check-raises than bets, because villain will bet at such a high frequency.

2) If villains range is way better than yours, then even if you check all your good hands (like you would in the big blind vs button situation) your range still wont' be strong enough to defend aggressively against villains bets. So if the button bets 1/2 PSB here, do NOT!! think you need to defend at least 66.7% of the time.

3) If you find yourself unable to defend your checks AND you have a betting range, then you likely have a problem. For example, it used to be very common for the CO to bet very aggressively against a button cold-caller, but not defend many checks. Clearly, this doesn't make sense. If your checking range is so weak that the button can profitably bet very aggressively, then you need to check more strong hands. This is why you see optimal type bots/solvers show that you should check at a very high frequency when OOP in CO vs BTN or UTG vs CO type spots.

However, this still doesn't mean if you check CO vs BTN and the BTN bets 50% PSB you'll be able to defend aggressively enough to prevent him from being able to profitably bet ATC. That's especially true since checking (as the button) will be +EV with ATC too.

Hope that helps, if not let me know where you're still confused

hi, Im not a native english speaker and I struggle with understanding the following:

"Stated again, if the preflop raiser 4-bets 27.5 percent of the timeand calls a 5-bet 55 percent of the time, he will end up felting 15.2percent of his opening range.
This is an important frequency to keep in mind when playing
against opponents who refuse to flat 3-bets out of position.
Against these players, we need to defend at least 15.2 percent of
their total opening range or else bluffing with hands like ace-rag
suited and pocket pairs becomes profitable for them."
s. 45-46

What does "he will end up felting 15.2% of his opening range mean" ?
What is felting ? If I translate that to german the word makes absolutely no sense but I still think I just dont understand the word itself used in the context maybe.

Felting means go to showdown. It originated from literally flipping your hands over and placing them on the felt in a casino, like our ancestors did.

So a good example of this would be "If a player bets 100% of the pot on the river and his opponent doesn't think the bettor should be able to profitably bet any two cards, he'll end up felting 50% of his hands." So he'll call 50% of the time and (if playing live) flip over his hand, actually putting it on the felt.

Matthew - no time for forum post, get on with the book !

Hi matt,

thanks for your replay but now i get confuse by this:

Do NOT!! try to prevent players from being able to profitably bet any two cards against you when they have a stronger range than you and position.

my doubt is: by estimating villain range im adopting an exploitative approach, if im playing a game theoretical sound strategy should i not care about villain range?

I know that Nash equilibria arise by an iteration of 2 rational player that try to max their EV and therefore, after some iteration, they will tend to an equilibrium, i.e the Nash equilibria.

It seems to me that we are using a mixed approach, and this is what confuse me. With mixed approach a mean: we apply game theory concept, i.e min def freq, ecc.. but we still care about villain range and try to estimate it, i.e we switch between "exploitative glasses" and "GTO glasses"....

if this is the case, it now make sense what you are telling me: Do NOT!! try to prevent players from being able to profitably bet any two cards against you when they have a stronger range than you and position.

Am I wrong?
cheers Matt

It's pretty much done on my end and 2+2 is quickly going through all the necessary steps to get it out.

Ignore everything above but the bolded, since this is why you're getting confused.

Optimal poker cares about villain's range and always assumes it's playing against an optimal range.

Think about it -- Of course optimal poker has to care about villains range. If it didn't, it'd assume a button opening range is no different from a big blind 4-betting cold range, which is of course ridiculous. It just always assumes that whether it's playing against an opening, 3-betting, 4-betting etc range that it's always playing against a perfect range for that given position.

So for example, an optimal player min-raising from the button will assume the big blind cold calling range will be a very weak range and adjust accordingly (bet at a high frequency and expect lots of folds). So even if very advanced artificial intelligence software can't know the exact optimal ranges (no software currently does), it's estimating ranges everywhere because if it didn't it'd be absolutely impossible to play poker.

Hi Blunderer:

Matt worked very hard on this book and his work is now done. The book has been sent to our index builder and based on that we should have it by the end of May.

Based wishes,
Mason

Hi Matthew,

I'm really enjoying your book as mathematician and a poker player with 18 years experience. One of my favorite sections is the one on pinpointing optimal river bet size. In 18 years I've never solved for this, I actually paused and came up with 0.83 through trial and error. However I believe this only to be correct in a vacuum, there is added value for your range in betting larger that we can actually quantify. Lets assume we can value bet 50 out of 100 combos here, the 83% bet size means we can be bluffing 31.2% of the time, that would be 22.67 bluff combos and 50 value combos. If we bet pot we can be bluffing 33.3% of the time, meaning 25 bluff combos and 50 value combos, adding 2.33 more bluffs to our range. Since when we bluff with a balanced range we have essentially won whats in the pot this would mean that increasing our size from 83% to 100% of pot gains us 0.0233 pot sized bets. The EV when we bet 0.83 is 0.20455 PSB and the EV of betting pot is 0.20000 PSB. We lose 0.00455 PSB by betting pot but more than make up for it with our bluffs, the equilibrium is somewhere between 1.2x pot and 1.3x pot IIRC from trial and error last night, I'd love to dive deeper here but I'm short on time. Without your book I wouldn't have even got this far, I really like your work but can you update that section taking this into account, I know it'll make the most complex math in the book even more complex, we'd also need to add a variable for how frequently we have a hand of this strength. It's been 20 years since I last studied calculus but I think it's worth looking into and you're one of the few people I consider more qualified than myself in the area of poker math.

Thank you for opening my mind to new ways to apply math to poker. I really enjoyed this one, not for everybody but as a math guy this was right up my alley!

Hello Mr Janda... Im going through your book and as many others would like to have the BvB 3b ranges but unfortunately wasnt able to send you a pm probably cause im not active on forums.. Any chance to get in touch on skype? That would be awesome

Also would like to ask you what changes would you make to the ranges you purpose 4 years after writing the book also considered that most 6max players dont have a flatting range from sb vs co,btn? (At low stakes at least)

He has a follow up book coming out this month.

Yes Husker I noticed... is it kind of an update of the first book u think?

Check out the thread on it

http://forumserver.twoplustwo.com/33...janda-1660716/