Then don't get pissy when I do it.

Let's say I throw in all 16 combos of KQ into my turn betting range. How much "more equal" do you think our ranges are, and what do you think is the corresponding increase in EV from the increased river betting frequency and river bet-folding frequency?

From my perspective, you're getting all worked up trying to exaggerate how terrible it is that I'm not bluffing the turn without realizing just how small of a change it really is.

My bluffs: forgot to mention Ad8o and Ad9o that both get bet at near 100% frequency in adddition to A9dd, A8dd, A6dd, A5dd, A4dd, A2dd, K8dd. That makes a total of between 14 and 15 bluffs / 119 to 120 total combos = between 11.76% and 12.5% bluffing frequency. Bet folding KJ and AJ seems like too much (24/120) = 20% of my range, but remember that read from the original post:
this is exactly a reason to value bet fold more often than the model suggests. We're also not 3 bet bluffing AJ even at low frequency, so those missed bluffs become exploitive bet folds, which drives up the value bet folding frequency.

It may seem that because of the above read we shouldn't bluff. However, remember that checking back the river is profitable. Vs the perfect opponent in the big blind, checking back the river with A9 will have the same ev as bluffing; both are positive values; just because this opponent is known to pay off thinly, this doesn't in itself make bluffing unprofitable. Perhaps we could check back A9 and A8, but that's as far as I would go with the adjustments to my bluffing range.

This one time? It's insignificant. Over 100 tries with the possibility of being counter exploited? It's significant enough to make me not stray more than a click or two from the margin as noted above with A9 and A8.

I try to answer all of your questions. You've been dodging my questions for years and years.

Not much "more equal" because you're missing many more profitable semibluffs. It's not about KQ, it's about KQ, AQ, AK, A9, A8, K9s, and all the diamond draws.

Going up from 110 to 120 combos isn't that much. The ranges at this point are really quite small.

No, this is exactly a reason to *bet* more. The *fold* part is irrelevant to the discussion of getting value from players that will pay off thinly.

*Yawn* I have no interest in second order effects that probably have < 0.01 EV order effects. That you're even adding this in as something to be concerned about shows me that you're nitting things up again.

For all the analysis you do right, this is one area that you seem consistently wrong on in this thread. You seem to have no intuition at all as to the absolute quantities involved. Yes, you can take your calculations out to several decimal places. But if the input assumptions are noisy, you're adding noise to the noise.

How significant is significant enough? You've not really demonstrated to me that you have any realistic intuition about the actual sizes of the changes involved.

My basic claim is that you're still well within the noise. Worrying about exactly one or two hands is literally pointless because of the quality of the data entering into the analysis.

I'm tired of you throwing in piecemeal information. It's quite annoying and it gives the impression that you're making it up as you go. They also aren't profitable if better hands are consistently calling. You're trying to have it both ways. "Pays off thinly so I should value bet more, but these are awesome semi-bluffs so I've got lots of value here."

I asked you what your range was going into the river. Why didn't you just give me your range going into the river?

Shall I accuse you of dodging?

No, you're wrong here. Folding to the raise is crucial if the read is correct. Vs many players calling is a 1BB mistake per hand. That's 100BB/100 hands.

-0.01BB/hand is insignificant over one trial. -1BB/100 hands can wipeout an expert winrate.

Because the checking range isn't important; it's profitable all on its own. The important part of the river discussion is the value:bluff ratio and the bet fold %.

You keep mentioning KQ like it's the crux of the argument. It's not. You could probably check it back 100% on the turn and not see much difference in winrate. It's 16 combos, right? AK, AQ, K9s, K8s, Ad9o, Ad8o, A9dd, A8dd, A6dd, A5dd, A4dd, A2dd, adds up to 54 combos without KQ. These combos can and will make a significant difference in the equity distribution on the turn and going to the river.

LOL -- OMG YOU'RE GIVING AWAY 100BB/100!!!

Just no.

Just no.

In the first case, it's a 1 BB error per similar situation. Given that you might get river-check-raised once every 2000-3000 hands or something like that, I'm really not worried. (Edit: And have a bet that on a thin value bet where you need to decide between calling and folding.)

In the second case, it's a 0.01 error per similar situation. What percent of turns do you see? Use that and *then* estimate your BB/100 difference. (And even then you're still overestimating.)

Again, you seem to have no actual intuition about the true sizes of the things you're getting uptight about and no sense at all about the level of noise you seem to think you can so expertly parse.

You think you're fine-tuning things that are literally 1 hand type differences that you're saying separate "bad poker" from whatever type of poker it is you think you play.

How about this: Do the actual exercise and show me the results. You keep making these assertions about how big of an impact they are, but your first two attempts at putting numbers on things were just bad.

Go calculate something to try to determine the size of the error.

Not really. I think never bluffing this turn and bet calling your entire river betting range is terrible. It's not "1 hand type." It's your whole strategy. Just because we have similar value betting ranges on the river doesn't tell the whole story; our strategies are very different.

Then let's talk about the turn. What your hand range going into the turn?

33+, A5o+, A2s+, KTo+, K8s+, Q9s+, J9s+, T9s.

You?

I don’t think that there’s any realistic turn strategy that would possibly make my bluffs unprofitable with the exception of A9 and A8. You’re using the argument that we need to make better hands fold in order for a bluff to be the most profitable line. This is the old way of building pre river bluffing ranges. Now we’re aware that there’s also a mix of other ev sources that contribute to the profitability of bluffing ranges to the point that the word bluff itself is incomplete. These ev sources are the possibility of getting calls from worse hands and making weak draws fold. So when I bet the turn with the hands listed that you called piecemeal, making better hands fold isn’t the main objective. This doesn’t determine the profitability of the bet; draw value, showdown value, fold equity, and protection all have a part in creating a situation where ev bet > ev check.

Are you c-betting 100% on the flop? There's nothing in this range here that suggests you're checking behind (at least, not in any balanced sort of way). Not that there's anything wrong with that, because I'm c-betting 100%.

It looks like you're up around 25% 3-betting here. I'm closer to 20%. My ranges are below an analysis of your range.

So with this in place, the 7d comes, and you do what? You say you've got 105 value combos on the river. So that's something like what?

Sets = 12 hands
* JJ = 3 hands
* TT = 3 hands
* 77 = 3 hands
* 33 = 3 hands

Two pair = 2 hands
* JTs = 2 hands

Top pair = 30 hands
* AJo-KJo = 18 hands
* AJs-QJs = 9 hands
* J9s = 3 hands

Second pair = 30 hands
* ATo-KTo = 18 hands
* ATs-QTs = 9 hands
* T9s = 3 hands

---- That's 74 hands. There's still another 31 hands to go.

2.5 pair = 12 hands
99 = 6 hands
88 = 6 hands

Bottom pair = 12 hands
A7s = 3 hands
A7o = 9 hands

Below bottom pair = 12 hands
66 = 6 hands
55 = 6 hands

------ This is 36 additional hands. So you're basically saying that "value hands" as "basically any pair or better"? This seems a little bit absurd given the board texture and the guy that called your 3 bet preflop. What range are you putting him on?

And then you're also got this list:

AK/KQ/AQ = 48 hands
A9/A8 = 32 hands
K9s = 4 hands
All the flush draws (A6/A5/A4/A2/K8) = 5 hands

This is another 89 hands. From context, it seemed you were suggesting that you bet all these hands. Is that an accurate assessment?

And then what's left?

44 = 6 hands
Other offsuit aces (A5o) = 12 hands
Other suited aces (A6/A5/A4/A2) = 12 hands
K8s = 4 hands

I only see 34 hands. This is a manual count, so I'm sure I've missed some. But I'm pretty certain I'm close.

What exactly is your strategy here? This all seems extremely odd to me.

------------------------- What I'm doing here -----------

This is my range because (as I said) I'm pretty much c-betting everything: 44+, A4s+, A7o+, K9s+, KTo+, QTs+, JTs, QJo

On the turn, let's say I'm value betting second pair or better. And upon reflection, I probably do bet the A-high flush draws, which picks up a few more hands. I was focused on the straight draws/gutshots in our discussion and the backdoor flush slipped my attention.

68 non-semibluff hands

Sets = 12 hands
Two pair = 2 hands
Top pair = 27 hands
Second pair = 27 hands

Semibluffs (all flush draws) = 10 hands
Two way draws = 4 hands (AK/AQ/KQ/K9)
Flush draw only = 6 hands (A9/A8/A6/A5/A4/A2)

And then I'm checking behind the rest = 129 hands
Lesser pairs = 42 hands
Offsuit Aces (AK/AQ/A9/A8) = 48 hands
Suited Aces (AK/AQ/A9/A8/A6/A5/A4) = 21 hands
Offsuit Kings (KQ) = 12 hands
Suited Kings (KQ) = 3 hands
K9s = 3 hands

Edit: I know my manual count here is a little off. I took some of the numbers from above because I got lazy, and I probably am slightly overcounting some of these categories because I'm too lazy to re-count everything and remove the hands that I don't have but you do.

Here comes Bob and Aaron again.

You cant be sure 90%. It's either i'm sure or i dont know.
So you claim that in case of "i dont know" call is 100% more profitable than fold?

The terms value and bluff are pretty meaningless regarding pre river poker at this point. Hands receive ev in a number of ways and putting them into neat value and bluff categories isn’t going to help.

I’m still struggling with this myself and I find myself thinking about pre river value:bluff ratios while I play. Old habits die hard.

I bet KT on the turn and anything better in addition to the many draws that I bet at 100% frequency: AK AQ KQ Q9s, all diamond draws. The hands I mix with on the turn that are bet at high frequency: Ad9x, Ad8x, K9s, K8s.

I realize that my river analysis from earlier is flawed; I had K9s and K8s in the combo count as 100% turn bets. Either way I don’t think it’s going to be much of a problem.

I appreciate your analysis but I think you should be looking at (total turn combo count) > (total river combo count) because this is the one true measure of street to street range consistency. Value:bluff combos for pre river poker are out and value maximization is in.

The last bit I wrote is off. I’d like to change it:

...by betting, we’re creating the effect of devaluing our opponents range by reducing the profitability of the opponents individual hand combinations. This is particularly true of bluffcatchers and draws which are only receiving a small profit from the pot when facing a bet. The hands that counteract this effect are the monsters which receive large fractions of the pot due to the combination of raw showdown value and draw value.

Also I have some mumbo jumbo in my head about preparing a solid river range and strategy for the endgame , but it’s still fuzzy.

Also also, traditional value:bluff ratios tend to make us think that we want action with our value hands and that we want folds with our bluffs. The reality is that we only want action with the strongest parts of our range. This is easier for me to visualize when I think about my opponent calling with profitable draws and bluffcatchers. If they’re earning profit from the pot then that means they’re taking chips that would be mine had they folded.

Thinking about it this way I see that correct poker means never making a -ev decision. Of course I don’t mean we win every hand, but rather that it’s a series of profitable decisions.

Yes, but not on every flop.

----

ArtyMcfly on pre river bluffing frequency:

https://forumserver.twoplustwo.com/1...+matthew+janda

He was talking about no limit, but I think this applies to limit holdem as well.

That's not to say that pre river betting ranges are merged, but rather that when viewed as a whole, pre river betting ranges will include hands that earn profit in a combination of different ways:

Let's look at combinations from the top, middle, and bottom of my turn betting range:

top: JJ, the monster that is a set of Jacks is a very profitable bet which earns a large fraction of the pot almost entirely due to unimproved showdown value; however, there is some draw value here as this hand might improve to beat 98s which turned a straight; protection value is essentially non existent here as any hand that could improve to beat three Jacks on the river is surely calling the turn.

middle: J9s, top pair and a gutshot is a somewhat profitable bet which earns a fraction of the pot not just because of unimproved showdown value, but also because of draw value and protection. Sometimes this hand will improve to a winner on the river and sometimes the opponent will fold a hand that could have improved to a winner on the river.

bottom: A2dd, flushdraw and a single overcard is (I hope) a hand combination that is likely slightly more profitable as a bet than as a check. This hand receives ev from the pot in every way possible; draw value, protection, and unimproved showdown value are all contributing factors that will determine the profitability of betting this combination.

bottom: Ad8o, gutshot and a single overcard is (I hope) a hand combination that receives ev from the pot in every way possible; draw value, protection, and unimproved showdown value are all contributing factors that will determine the profitability of betting this combination.

----

When viewed as a whole, the turn betting range may seem to be polarized with value hands at the top and bluffs at the bottom, but this is a bit of a misrepresentation of why we bet; this gives us the incorrect impression that a bluff by definition must make better hands fold in order for that bet to be profitable; it also gives us a false sense of pre river indifference, which isn't about making the opponent's bluffcatchers zero ev. Instead, pre river indifference means something else that I am still unable to put into a concise and coherent statement.

me on premodern pre river indifference:

a feeble attempt at nutshelling modern indifference theory by me:

https://forumserver.twoplustwo.com/1...08/?highlight=

We're down to the point where you're either going to show me something computationally or we're going to part ways in the conversation. I'm less interested in your thoughts about game theory in the abstract and more interested in the details of this particular hand and this particular situation. For as much as you like to theorize, I am really looking for you to analyze.

I've laid out my turn range and turn strategy. You've laid out your turn range and... well... sort of a strategy. I'm betting around 35% of the time and you're betting more than 50% of the time. I'm betting value hands and a few draws (roughly 15%) and you're betting KT+ and a bunch of draws (roughly 45% of what I'll call "made hands" to avoid using the word "value").

You have enough information to do something with it. Not some abstract theorizing about what you think is happening in this hand, but do a true analysis of this situation. Show me some assumptions and some calculations.

If you stick to your never bluff the river plan and bet call KK+? I don't need to analyze to know that you're vulnerable to a check raise range that consists of 100% value. I know you're calling my check raise, so I wouldn't bluff.

If you reconsider this plan and decide to bluff the river with those missed draws at the bottom of your range, then you're still vulnerable to a check raise range that consists of 100% value even if you bluff at the frequency which makes the big blind indifferent to calling or folding hands that can only beat a bluff.

This is because in both cases you're bet calling at a frequency that is too high; check raise bluffing will be unprofitable for the big blind.

One of the conditions of equilibrium is that you could tell your opponent your entire strategy and there would be no ev for that opponent to gain through adjustment. Vs your strategy, I could check and fold hands as strong as AQ; this would cause an ev shift in my favor through saved bets. The ev of my checking range will also increase due to the pots I win on occasion when you check back hands that should have value bet or bluffed the river.

Since I'm gaining ev with these adjustments, you must be losing ev.

----

If instead you choose to bet fold at frequency that makes the big blind indifferent to check raise bluffing or check folding, and if your value range includes all hands that have ev bet > ev check, then you will not be vulnerable to the adjustments above.

Ok. Bye.

At no point in time did I say I'm playing an equilibrium strategy. Nor did I ever say that I'm not vulnerable.

Your failure to actually analyze anything at any meaningful level and simply remain theorizing tells me enough about your willingness and perhaps capacity to actually demonstrate something specific about this situation, despite having provided fairly detailed information about my approach to the situation AND having provided you an explicit outline of what you can do to demonstrate something meaningful.

I will now accuse you of dodging. "Peace."