How’d you calculate this?

We want to calculate P(KK+ | LRR), the probability that someone has KK+ given that they have limp-reraised (the vertical line is standard notation for "given"). By Bayes theorem,

P(KK+ | LRR) = P(LRR | KK+)*P(KK+)/P(LRR)

Since we are assuming that everyone LRRs with KK+ in this spot,

P(LRR | KK+) = 1.

Since the probability of being dealt KK is 1/220, and likewise for AA, we have

P(KK+) = 2/220 ~= 0.0091.

Finally, we calculate the denominator, P(LRR), using the law of total probality:

P(LRR) = P(LRR | KK+)*P(KK+) + P(LRR | WorseThanKK) * P(WorseThanKK).

The first summand is 1*0.0091. The probability of being dealt worse than KK is 99.1%, so P(WorseThanKK) ~= 0.991. Since we assume that 2% of players LRR with worse than KK, and they do so with 50% of holdings, we have P(LRR | WorseThanKK) = 0.02*0.5 = 0.01, from which we conclude

P(LRR) ~=1*0.0091 + 0.01*0.991 ~= 0.019.

Now putting it all back into the original formula gleaned from Bayes theorem,

P(KK+ | LRR) = P(LRR | KK+)*P(KK+)/P(LRR) ~= 1*0.0091/0.019 ~= 0.48.

Exactly this. Without the anomalous table image this is a fold all day every day (since you can't call for set value). No but occasionally me is ever doing this with JJ or AK under normal circumstances.

But these aren't normal circumstances. In this case it would depend on a lot on the limper's looseness, which you probably can't judge based on 4 other hands. It's plausible he just picked up KK+ and is playing them how he always does. It's plausible that he's playing back at the young punk.

Without a better read, I'd shove and hope the table image gets me light calls from hands that would never have L/RR in other circumstances anyway.

His mental state doesn't have much to do with it IMO; people rarely have a subset of their "I'll teach this young punk a lesson with a 3bet..." range that's also the "...but fold to a shove" range.

It's a great example analogous to this classic example from any stats class: https://www.math.hmc.edu/funfacts/ffiles/30002.6.shtml

Cliff's: If you test positive on a 99% accurate test for an extremely rare disease, it's still overwhelmingly likely you don't have the disease. The 1% of healthy people the test gets wrong are way way way more than the 99% (of a much smaller pool) of the people who actually have the disease.

Obviously both are toy examples but I wouldn't use 50% because even the maniacs are probably waiting for a hand good enough to play in the first place. (I'm aware results have been posted and am ignoring them.)

It is much simpler than bringing out the big statistics guns. All we have to do is make a range. I am pretty familiar with this particular toy game bc I play quite a bit of low stakes ignition zone, and this is standard database analysis.

As an example from iggy and relating to your numbers: 90% of players have no 4bet other than KK+, and 10% spew hard. This creates an interesting dynamic where we must make a “general” 4bet range bc of the anon player pool, and bc of the great disparity between one extreme never bluffing, and the other spewing hard, the result is that an anon player on average has a somewhat balanced 4bet range.


For those that hate statistics and want to think about this with a more arithmetic based approach...

Considering we already assume the lrr action;
-lrr 98% of the time is KK+
-2% of the time is 50% of holdings, then:

KK+ = 12 combos
12*98%= 11.76 combos

50% of hands = 1326*.5 = 663
663*2% = 13.26 combos

11.76+13.26 = 25.02 available combos
~47% of total combos being KK+

Key takeaway #1:

47% of a range being KK+ is actually a very strong range. If someone only had AKo and KK+ in their range, they would only have KK+ 42% of the time. I realize the equities of AKo/KK+ vs KK+/13 rando hands would be slightly different, but you’d be surprised playing around with an equity calculator.

Key takeaway #2:

Assuming 2% of the time someone has half of available combos is probably a bit of a stretch, but even if it’s not, they aren’t stacking off with that full range. Our equity QQ vs gii range therefore is going to be drastically different, and so the onion peels back farther than the surface layer of 98% KK+ and 2% spazz. That’s an analysis for another day though I’m afraid.

A lot of people get stuck at a stake on ignition and wonder why they can’t beat a game where people are 4bet stuffing 59s. It’s because the range that contains 59s is actually very, very strong.

I think you're getting too hung up on the specific 98/2 numbers that I just plucked out of the air to illustrate the point in a simple way. As I said later in the same post, the reality is more of a spectrum, although I do think based on experience that 2% is a reasonable ballpark figure for the portion of the player pool that will get truly spazzy like Villain did in this exact spot. When you combine that with people on earlier parts of the spectrum that show up with like 77+ rather than KK+, you have a clear shove.

Fish who are trying to play well do adjust, they just do it badly. I can't tell you the number of times I've tilted a table raising in position all the time and someone just gets sick of it and shoves, or calls down with a very weak hand. Some fish just won't but some will and it's pretty obvious which ones will and won't. In this example V told you how pissed off hes getting.

You don't raise every hand and then get scared with QQ, wtf. Especially vs a 30s overweight WG whose blood pressure keeps going up because you raise every hand.

It was a different guy.

I shouldn't post before coffee . I stand by the thesis of my post though

I do agree with you that fish do adjust, but they do it badly. My main point is that they often doesent adjust as we 2+2ers would do, and we shoudnt project our own views/skills in poker onto our villains.

I also do agree that its extremely crucial being able to identify wich villains who will not adjust to our aggressive plays aka just wait for a nutted hand before they fight back, and the smaller amount (in my opinion and my experience) that have spazz potenial when they get tired of us betting/raising.

What i dont agree with is that QQ is an autostackoff here, or that we somehow is "scared" if we dont stackoff. Its about knowing the tendencies of our villain right+ his mindset how he approaches poker, and the times i have folded QQ in similar spots pre i do it because i gauge villain to simply having me beat.

In general though from my own experience in the games i have been playing (different homegames around Norway+ Vegas games) is that the vast majority of villains doesent adjust at all, or at least they dont adjust the way we often expect them to.

In the same way you can't take a line that induces bluffs and then fold a big hand, you can't play 100% VPIP and PFR and then fold a premium to someone playing back at you.

We totally agree on this premise too, and QQ may very well be a stackoff here too dont get me wrong, i wasnt at the table with this particular villain. I think our disagreements lies uppon how and _if_ our villains is adjusting in the way you believe they are. Yes, in this instance obviously villain was spazzing out regarding results, but over a bigger samplesize i am not convinced that is the case often enough.

Without reading results, I call and jam all non A or K flops.

You either play in a room full of drooling idiots or you're just terribly unaware of your surroundings. This is 2018 poker, not 2008 poker, people don't adjust like they should but they will adjust, especially vs extremely loose or extremely tight play.

Everyone suggesting fold needs to really grow a pair, or get a proper bankroll. Hero has raised every hand, he has a short stack, slam dunk get it in spot.

Regardless of what the right thing is to do with QQ here, both of these are terrible ways to think about poker decisions. If preserving bankroll is more important than EV you should move down. That has nothing to do with why people are advocating folding. And if proving your virility is an important motivation for you at the poker table, I hope you play in my games.

(I'm still calling off QQ in this situation but probably not TT. It's true not a lot of people adjust well; many people think "I can't wait until I get a hand to teach him a lesson.")

I disagree, the motivation behind folding here is just fear. Fear of losing a hand, seeing monsters under the bed, etc.

"Aw man, this is AA or KK always!"
"Well maybe it's plus EV, but it increases variance."
Etc.

There is a big difference between QQ and 1010. Hero has sat down and won every hand, his stack is relatively small. This can as bad as AJ+, 88+ and I don't think it's impossible that this range is giving v too much credit. This is a super easy get it in spot, if you're up against AA or KK gg.

JJ+, AK+ we have 47% equity. 98% of the v range being KK+ is not a reasonable assumption either, we all know that (I hope).