GeeTeeOh's avatar


93 points

Great video Richard!

Interestingly enough Qing Yang made a video about a somewhat similiar topic a few weeks ago:

Qing Yang - What is Shania

Also you might enjoy this one:

Uri Peleg - SDV Bluffs

Jan. 16, 2020 | 9:37 a.m.

Great catch! kaytor888

Jan. 16, 2020 | 9:34 a.m.

Hi Seth,
I ran the first hand (QQ on KT7ss) with multiple raise sizes. PIO indeed chooses a bigger size (I gave it 38%, 58% and 78%). However the x/r frequency is not dropping massively and also the turn strategy does not change drastically (%% are dropping obv, but no huge shifts). So I think you are being a bit too harsh on yourself and the results of your sim are very relevant.

Jan. 10, 2020 | 10:40 p.m.

26:20 T5s
I thought that we would want to have a small x/r size here, but that proved to be wrong. However PIO also does not like jamming this combo, it only uses our strongest Tx or non suited Tx/9x with a heart:

Dec. 11, 2019 | 1:34 p.m.

Why was my comment deleted?

I was asking about the river strategy in the second hand. Do you know why PIO chooses such a betting strategy with only the overbet jam and the tiny block? Seems like QJs e.g is worth more than just 10%. Also interesting that KJ is checked at such a high frequency.
Did you run a hotness comparison for the river? The 3 is a very bad card for us, as IP now makes a set with almost 3 combos and this being ~10% of his whole range at that point. So is the reason that such a strat is chosen. How does that change if the river is a 2?

Great video, plz more :)

Dec. 9, 2019 | 1:02 a.m.

Comment | GeeTeeOh commented on WCOOP Highlight Reel

KJs my wording was a bit misleading. I meant flat ir 3b/jam.

The 82s just striked me in connection w the T2s (23.40), where you also want to go for it with this type of hand. Also PIO chooses slightly different variations of these hands, but the EV differences are obv pretty small.

You say that hir River jam is prolly the best play. Don't u think it is too thin vs EP? The o ly hand he is getting value from 100% of the time is QTs, maybe Q9s, so 4 combos at max whereas he is beaten by 31 better combos. I really doubt you are calling too many PPs here, A7s is also only 3 combos. Whereas you have missed spades and A4s/A3s that flopped BDFD. So isnt x/calling more reasonable, esp on a River that is better for your range?

Dec. 6, 2019 | 1:47 p.m.

Comment | GeeTeeOh commented on WCOOP Highlight Reel

6.35 KJs
You talk about 3b/f this one. This seems like a pretty huge mistake. Seems like a 3Bjam or call unless you have very specific reads.

9.15 82s
I am quite surprised in you advocating a pure x/r here. x/r here with all of our weak pairs does not seem like a very sensible strategy. Sure we fold out some hand with pretty good EQ and prevent him from barreling us, but we also bloat the pot with a marginal hand where we will have a hard time showing down a winner if he continues. And if we decide to turn it into a bluff on later streets we will be massively overbluffing.

11.35 AJo
Could be a jam aswell. Do you leave these out, because you think you have a massive edge and want to play an approach of reducing the chance to bust as much as possible?

Nov. 30, 2019 | 6:12 p.m.

I think your question is geared towards something similar that I have asked about in this thread:
There is also already a discussion about the topic.

Nov. 26, 2019 | 8:30 p.m.

With the given information there is too many advice that can be given. Seemingly the structure is like a super hyper turbo, so you should brush up your push/fold skills and also know what to call off. Other than that your strategy in the late stages will be heavily dependant of how many place get paid and what amounts^^

Nov. 21, 2019 | 1:58 a.m.

In case you didnt see, 66-44 are jamming here. PIO just knows how much EQ they have and how they will realize at every node, so I suppose it chooses to jam these PPs because we are a slight favourite, but our EQ realization on later streets is pretty bad.
I wonder whether it chooses to call them at some freq in order to have some hands to call down on favourable runouts for us or to not overdo the x/jamming so that villain cannot start to exploit us by adapting his c-betting range.

@ Ryan
I think your non-AI x/r size is way too big here. You should try giving OOP the possibility to x/r to min or slightly bigger, then we might see some non-AI x/r.

Nov. 20, 2019 | 9:33 a.m.

First hand:
Seidel has 66, 77, JTs, QJ and QTs at some frequency. Also AA and KK.
Davies is beat by trapped Flushes, AK with one diamond, JJ. All of these combos can be heavily discounted. The hand he is most likely loosing to is QQ. Not sure where 89s fits here, i.e how often Seidel checks vs bets this combo OTT.
But if you look at this range there is still a ton of combos he can get value from. Having the Td blocker is extremely important as he blocks several of Seidel's potential traps.

Second hand:
Would be interesting to know which cards have the diamond suit OTF and whether Koon holds the 8d. Without that info it is pretty hard to judge how many flush combos Dvoress can have and also how many very strong bluffcatchers he could have.

Nov. 20, 2019 | 9:22 a.m.

Comment | GeeTeeOh commented on good or bad fold?

Fold pre, bet bigger OTF, call OTR. Folding vs 40% PS is ridiculous, he is valuebetting random Qx for that size.
The whole hand shows that you should do a lot of watching of videos here...

Nov. 20, 2019 | 9:10 a.m.

Thanks vor the vid Sauce.
You have folded small PPs from MP-CO a few times during the vid. Don't you think they perform very well vs passive and stationy opponents and that the gains here outweigh the negative rake effects?

17.55 KTs
I am a bit surprised that you go for 1/2pot here. PIO prefers to mostly bet small (~30%) or overbet. Wouldn't it make more sense to go for the even smaller sizing, as they will play even worse against it? Or do you just want to keep your range together and only have that 1 size in order not to be too exploitative?

24.30 AQs
You say fine hand by him. Don't you think he could eke out a Vbet here?

Nov. 16, 2019 | 1:43 p.m.

Thanks for the responses guys.
I understand that we have to fold a lot, once our checking range looks like this OTR. My question was, why PIO does not put more bluffcatchers in the checking range and instead uses them to Vbet. This also often happens when we have x/r OTF, bet Turn and now check OTR. It keeps barreling with most of the value hands and thus we are left with a super weak checking range that has to fold tons vs even a small sizing.

PrankCallRiver kind of has given an explanation. So basically as I also have already written, we are allowed to get to such a node with a very weak range that will fold a lot, if on an earlier node (or nodes) we already have generated EV by betting?

Nov. 14, 2019 | 1:03 p.m.

Great series Luc!

26.00 KQs
Do you also want to have have a non-AI x/r in that spot? Under non ICM we probably have some shoves but mostly want to x/r small, but here I would opt for either jam or call if we continue.

You say 99-TT are better continues OTT. I imagine they fold and we rather call lower PPs as they unblock his bluffs (actually not sure how many PPs we have to call here, as AK and AT are plenty of combos as a bluffcatcher. He might 3b/jam AK some of the time OTF though...).

Was about to write how bad flatting KQo is with the lagtard in the BB, but European delivers... really well played by him.

48.30 Last hand
I am very curious how you construct your betting range OTT here. Under cEV you would bet your good value hand, your best draws, some of your weaker draws and some very weak hands. Under ICM, what would you do with the 2nd category, i.e KJhh or cc? Seems to me we cannot b/c these hands and if we check them our betting range becomes even more polarized and his x/r jam even worse.

Nov. 14, 2019 | 1:17 a.m.

I have come across these spots a few times now, where we end up overfolding massively vs a smallish betsizing. These are always situations where we had the opportunity to bet with a part of our range, did that and now have the other par of he range that is very weak.
So for example we defend the BB and it checks down all the way to the river. We check, IP bets small (1/2PS) and we fold >75%. We could build our range in a way as to not to overfold so massively, but PIO chooses to put almost all of the value hands in the probe lines OTT and OTR. So are we "allowed" to build our ranges that way, because by having a probing range OTT and OTR we already have put pressure onto IPs range and are anyway allowed to overfold some part of our range as the BB defender? Are there any other reasons that I am missing. As PFA vs IP CCer we should not be allowed to overfold so massively at any node right?

Here is a pic to the example above:

Nov. 2, 2019 | 9:07 p.m.

In HRC you can setup sims where limping and openraising is allowed and the EV of a push ATC strat is higher there.

Ofc ICM is flawed but using Nash only on FTs or other spots where BFs are high is plain stupid.
In HRC you can also use FGS (future game simulation), where it will calculate a few of the next hands and will thus account for shifts in stack distribution.
If you really jam ATC in this spot (including all your premiums) they cannot exploit you by calling lighter. It hurts their EV and it also hurts yours, while the rest of the players is gaining.

Nov. 1, 2019 | 1:27 p.m.

  • 20.10 bottom right 54s
    I think you missed an open here with the 2 shorties behind and the general setup of stacks behind you.
    • 2.1k looking at the line up on your table, how much sense does this 1 make? No spots at all, a lot of very tuff regs... lol, wrote the comment 1 min before you started talking about it yourself, but you kind of cut yourself off whilst talking about it...

Really nice series!

Oct. 30, 2019 | 9:33 p.m.

Thanks for your thorugh reply as well Luc :)
I guess in the end it boils down to how much one trusts into the ICM model and ones understanding of it as well as the deviations one should make because of the intrinsic flaws of the model. A major factor is also how much one trusts in his or her postflop abilities.
I do not doubt, that your understanding of the game is much better than mine and that your approach to the spot is well thought out and makes a lot of sense.
If I eventually get to run Monker myself I will happily set up a sim for this spot and share the results here.
On a final note it is quite amusing how a seemingly ultra simple spot like this, ends up being more complex than it seems and can stir quite a discussion. So much to poker being close to solved...

Oct. 30, 2019 | 8:45 p.m.

Hi Luc,
I have made some calculations by hand. I made the calcs so, that they would support your assumptions in the best possible way (you only raise 25% from the BB and when you do not raise SB has an EQR of 100%). I have looked at 32o, so bottom of range. IRL card removal effects will also play a role, but I do not want to go down this deep.
- If you are calling Nash (66+,A9s+,KQs,AJo+), then the cEV of jam is +0,83BB and cEV of limp 0,64BB.
- If you are calling much looser (44+,A8s+,KTs+,QJs,ATo+,KJo+) the cEV of jam is -0,31BB, cEV of limp does not change obv.
The second calculation is kind of redundant though, because it is obvious that if you call that loose his bottom of range jams become really bad and he is then incentivized to jam much tighter, gaining a lot of EV with his top range and both of you loosing EV to the other players.

Obviously ICM has its flaws and I am all for not sticking to some model or simplifying our strategies down. But as I already said you can set up a sim in HRC with limps and raises and the EV for just jamming ATC will be higher than the EV of implementing a mixed strat (and actually the HRC results are skewed positively towards an approach of l/r/jam as the EQR for OOP is always 100% postflop). As for effects of shifts in stacks one can use FGS, but it will not make that much difference in this spot anyway, as even if we double we will be fourth in chips with a bunch of shortstacks still being there, forcing us to play really tight.

Oct. 26, 2019 | 5:15 a.m.

Great video as usual Seth (will stop writing this, just imagine it being pinned to my future comments :))
A quick question towards your sim config. Why do you use 999 instead of the "Add allin" box?
Also I do not understand why you do not use the 2 "... to weight" boxes. At 25.25 you say yourself, that J2s and J3s are almost non existant in our range, but because you do not have the boxes ticked, it looks like we get to the river with them 100% of the time.

23.45 It is actually using preferrably 6x and 4x for the big size. It is funny that you talk about how it unblocks IPs folding range, but ignore the fact that it is also blocking IPs strongest value hands and also IPs bluffjams, even though at the beginning of the video you were talking about these type of blocker effects.
29.00 OOP is never betting with just a naked 3, he is blocking 93s -> Flush or T3s -> 2pair. The slivers of AQ and A2 that call are not enough to vbet such a weak hand (and they also raise some time).
32.40 Again it is only betting AQs -> Flush

Oct. 25, 2019 | 7:53 p.m.

If you run the spot in HRC you will see, that when we play a jam or fold strategy, we can jam ATC here, while we are not allowed to play 100% anymore, if we implement a limp and/or raise strat.
I am totally agreeing with you, that he is for sure not doing this with his premiums, as no reg is and thus he cannot jam bottom range anymore either. I know the HRC sims are somewhat mehh as they do not simulate proper postflop play, but it is still a hint as to how a valid strat can be constructed.
I get your point, that you think, by splitting our range into jam, limp and raise we generate more EV than by simply jamming, but tbh you cannot prove this and yes, while being not 100% accurate, the HRC sims still point towards a just jam ATC. The only possibility to "prove" this would be to set up a Monker sim, but no one will go through the troubles of doing this...

Oct. 25, 2019 | 1:31 p.m.

7.40 pretty surprised at your comment. He can jam ATC here, as you have pointed out yourself, your calling range is very narrow (Nash: 66+ A9s+ AJo+ KQs), so he gets to win the pot >90% of the time uncontested.

Oct. 23, 2019 | 12:52 p.m.

There is a replay of the WCOOP 2019 10k FT HR here:

There is a lot of pretty wild and interesting hands, would be great if one of the MTT coaches (maybe someone who does not play online much anymore these days) could review it.

Oct. 22, 2019 | 2:31 p.m.

Pretty interesting how you constructed his flatting range. K8s-K7s are always 4bets aswell as A5s-A3s you think? K6s is a flat?

Btw you should tick the "bars width proportional to weight" box, makes much more sense to view the outputs like that.

Oct. 18, 2019 | 8:16 a.m.

Funny you talk about the topic of rolling and randomizing in live MTTs. I was at a table with older school live/online pros a few days ago and they were making fun of me for randomizing, saying it is useless in MTTs (esp low-/midstakes). One also said that by taking a pure exploitative route he thinks you can get a lot more value in live MTTs where people are just super imbalanced.
I still think that it cannot be that bad to base your game on approximations of GTO and deviate slightly, rather than making huge adaptions based on a few hands seen (or even just because you think grandpa xy will do this and that).
So if I understand correctly you are also proposing a kinda middleground solution?

Oct. 18, 2019 | 8:12 a.m.

BB should not have any sets here. If you bet, SB folds and BB x/r to not a big sizing he can pĆ¼lety of stuff, sometimes even valuehands that you dominate + you have BD Straight outs. So unless he makes it more or less committing huge, I would peel one off.

Oct. 11, 2019 | 9:19 p.m.

OTF both bet and check are fine. If we bet and get x/r I would not fold unless the x/r is huge or x/r comes from SB and then BB comes along.
OTT pretty much the same holds true, both small bet (30-50%PS) and check should do well.

Oct. 10, 2019 | 10:15 a.m.

"I know, we shouldnt be results orientated too much" This is what you are in this and thus probably in many other spots.

"Do I want to take a 60/40 Situation for my tourney life now?" There are very few instances where you would not want to take this spot and these are situations with extremely high ICM pressure.

The hand in question is an easy gii. As you have already written, reraising to 30BB is the same as jamming so just jam. I think in this spot you want to do all 3things, i.e flatting the AI, jamming over the top and also min-raising. AK is the perfect hand to jam, as the flatter has to fold many of his pairs that are a slight favourite and also many other hands that may be slight dogs, but will still realize their EQ if you just flat. Plus he will not even have KK+ here at a 100% frequency

Oct. 10, 2019 | 10:10 a.m.

In the first hand you can also just call... You will not have much FE anyway after going through the blinds so if you really think he is not folding anything vs your rejam just see a Flop and if you flop any kind of EQ you can go with it.

Oct. 10, 2019 | 10:03 a.m.

Load more uses cookies to give you the best experience. Learn more about our Cookie Policy