screamdustry's avatar

screamdustry

1185 points

Yeah. This is true in theory only in isolation to combinations EV's tho, right. Because if you start performing X-mixed actions as a pure actions instead and you totally mess up your frequencies that way, you gonna suffer EV loss, right?

April 26, 2022 | 5:12 p.m.

Clearly its the right play. I just can't get my head round the theory.
What do you mean by condensed. I've never used a solver so this is
fascinating.

By condensed I mean - cold caller has to have quite strong range on average (because he calls vs EP and has two players behind him), its just that he is capped.

" OOP is just automatically gaining EV with stronger portions of his
ranges." not sure I understand this point, but it seems important.

I was thinking the IP has a tonne of suited connectors and broadways
that would have to fold to a small bet. But by the sounds of it, the
solver will call these hands. So the range bet achieves nothing.

This is all not that black and white, rather this is very fluid and all close.

I wouldnt say that betting range achieving nothing, because it is still very good option, its just that having checking range and also higher cbet sizing does add some extra EV to us. I personally even like bet range more, because I see value in simplifying our cbetting strategy, I just tried to explain to OP here why 'great' players might derive from it.

Then I'm guessing the solver is also checking all it's under pair oop,
and therefore to balance you need to be checking tonnes of Ax
Therefore check is better

Does the solver just check fold OOP KQs QJs type hands. Or are thes
the hands it likes to check raise.

In the flop strategies, frequencies are almost always mixed. Like, the way you think, that solver is checking X hand always and then bets X hand always, and then check X hand always for balance, this is just not a way how it works.

Like, in this example, if you allow solver to have options to bet 2/3, bet 1/3 and check, solver is going to use all options for TT-66 for example. Some KK-JJ combos are not going to use 2/3 sizing, they gonna use only 1/3 and checking strategy. All Ax are going to mix between bet 2/3, 1/3 and checking.

Now, this might seem endlessly complicated, but your job is not to copy solver 1:1, its just kinda to get good idea how the ranges are looking and try your best in simplifying the strategy for your own use.

Like, here, some good players are going to be aware of what solver is doing on UTG, and they might just try to use RNG (random number generator) in their attempt to randomize the frequencies close to solver. Other players (like me) would just simplify and bet 1/3 range because this strategy is hardly losing any EV anyway.

If you find this fascinating, really recommend trying to use solvers, its a whole new world, and it can feel overwhemling at first, but I guess this is why we're at RunItOnce.

April 26, 2022 | 4:24 p.m.

that is true this is not polarized bluff to begin with since it can
still get value, so bet small and ditch the pot bet if you wanna put
more money in?

Not sure if I follow you correctly, but no, I think 43 is plenty strong to potbet turn, the trouble is just in the very unfavourable card river.

i dont think its deriving too far from theory since its possible that
both lines are going to be +ev and we are debating which one is the
higher ev one

I would've ditch thinking like this in poker ever. If you follow this thought train, you can talk yourself into giving massive amounts of EV/100bb just because whatever you do is EV+, even if the other option is massively better. In NLH there could be massive difference between two options that are on a surface both EV+.

For this reason, whatever is EV+ is not good enough to be considered GTO. The GTO is just most optimal (highest EV) line at a time.

April 26, 2022 | 3:54 p.m.

I wouldnt be reading and deviating this much of my strategy based on a fact that he has 30% cbet on a sample. Like, we dont even know what it means. Yes, prolly on average he's going to bet stronger than weaker, but you just cannot range him like he just bets 30% of his top range and check all weaker hands. Especially in this spot, where its so common to bet 1/3 with range. You can for example easily model him to cbet best Kx and be balanced with his bluffs, check other hands. Or to never bet best Kx with this sizing, instead use this sizing with merged hands etc. My main point is - I dont think its very wise to make strong adjustments to flop frequencies unless we know exact range composition, have view on other stats, with a decent enough sample.

Im mainly reffering here to these thoughts:

dont even think 2pair is strong enough to pot turn tbh, do it with
straights

30% cbet shows up with too many Kx here so just value town 2x when you
got the boats/straights

April 26, 2022 | 3:47 p.m.

You don't have enough of these hands to balance your value hand for a
large size. I don't think they are the best bluffs either. Your hand
is good to go. You unblock all the pocket pairs and block the boats.

He seems to have plenty of other hands to choose from to bluff tho, he has two flushdraws, all his Ax combos turning bluff, all 5x/6x combos. I think he's good not turning any pairs into bluff.

April 25, 2022 | 7:13 p.m.

Wow, these are some very non-gto advices from someone named GTO Warrior.

April 25, 2022 | 7:06 p.m.

1) true
2) true
3) he has quite a lot of 4x/3x, basically any 4x/3x suited can ch/c turn
4) true

And agree, that defence for SB doesnt seem to be too complicated here.

So Im also fairly certain this is not a bluff. In theory this is not even close to be turning into bluff and I dont see any additional reasons why would you derive so hard from theory here.

April 25, 2022 | 7:05 p.m.

OOP can cbet this texture with range without much EV loss, thats true. Playing checking strategy seems to be adding some EV tho, solver doesnt go crazy in this spot and optimally checks around 50% of its hands with allowed betsizings of 1/3 and 2/3.

This is happening because of position advantage and the range of IP being quite condensed and strong on average. IP will have a lot of pocketpairs to defend, and that makes OOP maybe not gain too much from just blindly betting range.

And why check/fold after check? This is simple, IP shouldnt really stab here a lot, and if he does, T9 is literally like the weakest combo that OOP can hold, so what IP gonna do with it other than fold?

And if part of your thought process is something like - 'ok, so why does OOP not protect himself against potentially more agressive stabbing strategies by cbetting range' - he doesnt have to do it with T9. If IP is using more agressive stabbing strategy, OOP is just automatically gaining EV with stronger portions of his ranges. Prolly can start check/raising a bunch aswell.

April 25, 2022 | 6:50 p.m.

Also, dont feel like you need to perfect. Deviating few % in most spots is just whatever. No one knows for sure how much EVs are certain openings/defences/3bets having anyway due to rake structure/population adjustments.

Oct. 10, 2020 | 4:17 p.m.

Comment | screamdustry commented on 25z Triple Barrel

Surely betting river is bad here? AsXs from villain would likely of
been raised at some point prior, so the As is kind of irrelevant as a
blocker anyway? Would we much rather have something like 7s7c that
doesn't block his blank AsX type hands that would be calling and fold
river? Any thoughts on sizings?

Wow, I simmed the spot and you're totally right, solver is not bluffing these, becasue Villain is supposed to raise all Ahigh flushes on either flop/turn. Which kinda makes sense actually. Nice.

Still, I think that if you 3barrel here, there is no way that population is comfortable here facing another large bet in a spot where you naturally are way more nutted than they're. So I guess the sim changed my mind from thinking its definetely slam dunk to thinking now that its propably ok. :)

Oct. 10, 2020 | 4:12 p.m.

Agree with Jeff_ that river we just want to put all money in, whatever the way.

If it comes to flop size, it is 140bb deep, so we either make flop bigger, bet turn solid and jam river for around pot, or we go small flop, bet big turn, but now we have to overbet river?

And I dont think its difficult to make argument, that overbetting river for value is also not perfect and will likely make Villain fold statistically more.

So yeah, maybe we fold more of their junk OTF with bigger sizing, but if alternative is overbetting river, I dont mind.

Oct. 10, 2020 | 3:58 p.m.

Yeah, this spot is definetely difficult for regulars to bluff enough when they're checked to OTT/OTR, becasue they need to turn pocketpairs into bluffs, as texture is not providing enough other hands that they could bluff with.

So im likely just cbetting flop range and then check/folding these pocketpairs, trying to see showdown.

I dont mind your play and sizing OTT, it feels viable, but im not interested in adding complexity by splitting my sizings here.

Oct. 10, 2020 | 3:49 p.m.

I thought betting out 22% pot was a great way to get value out of
those hands since a T isn't really a scare card for Villain unlike an
A which I could've been floating withe

Yeah, because its not scare card I would just go like 3/4 for value. If he has a pocketpair that he thought is strong enough to raise flop and bet turn, he's just not going to have easy time saying good bye to it on a relatively blank river to any sizing, really.

I agree with rest of your analysis in general.

About flop/turn decision between call and raise -> I think I would always prefer to call against 100bb, becasue at this stack depth you dont have too much troubles in putting all your money into the pot anyway.

I feel like deepstacks are giving you more natural incentive to reraise with nut portion of your range tho, so I would propably just try to squeeze max value from his whatever worse 2x or overpair by reraising.

Oct. 2, 2020 | 9:24 p.m.

There is also a fact that the implied odds are going to be better in multiway than in heads up on average, so in a case of having a pair I would still propably defend here. Fold more weaker pairs than in heads up tho.

Oct. 2, 2020 | 9:13 p.m.

[Let me know your thoughts on how I can improve]

I additionaly think that you can just upgrade your thinking process by simplyfing it a lot. Im not sure how it looks in-game for ya, but I usually try to ask simplest question

1) Is Villain bluffing here?
- No, given that SB just put most of his money, its extremaly unlikely that he's bluffing here.

2) Is Villain jamming here worse for value?
- Likely not. He might fold AQ on the flop, and we're also blocking these. Then, he also just might call it OTR instead of jamming it.

And here, thats end of decision tree. Villain is neither bluffing or is likely to have worse for value. In this case with additional important datapoint, that in 3way you actually need to also be even higher in range, becasue another player is going to also have better hands.

Oct. 2, 2020 | 9:11 p.m.

Comment | screamdustry commented on HU nl200

I would have same dilemma. It is pretty nice combo to bluff, but I totally agree that the Villain's defence is not looking challenging at all. I usually tend to give up in these scenarios. But Im also a puss. :)

That being said, I definetely like flop/turn play. Good to know about overbetting turn.

Oct. 2, 2020 | 8:53 p.m.

I agree with Kalupso here. Simplifying, I don't think that our job at these stakes should be to try to make fish fold top pairs on dry boards.

Oct. 2, 2020 | 8:46 p.m.

If we have to overbet jam river to get ppl at our stake to fold a weak
Ax here then it just means they are not folding enough,

Calling most Ax in Villain's range is optimal strategy in theory against triple barrel here.

Oct. 2, 2020 | 8:44 p.m.

No no, I misread the preflop action. And I totally blame that horrendous replayer, not my cognitive skills or anything ;)

Then I guess the 2nd most likely hand for value is like nut flush that he decided to 'slowplay' IP. So flushes are gonna be no question bluffcatchers over Kx.

Sept. 8, 2020 | 5:15 p.m.

The only combo that actually makes sense to check turn and raise river is KK. It kinda makes good sense for him to play like that, but its obviously super narrow to range on one combo.

I dont know. I thought to myself - I would propably like to overfold here, but not to a degree of crazy.

But then, are Kx better here to bluffcatch with than flushes, given that it reduces that one combo that makes most sense for him to have? If he's not having for sure KK, does that mean that his range is now more volatile? I dont know either.

I guess that its just one of these spots where I would give up futile process of assumption-making, given how little idea we have, and just be satisfied with anything that does look like EV+ calldown in theory.

Sept. 8, 2020 | 12:52 p.m.

@1:00 58
Any particular reason you dont attack any of these streets with 85high?

@9:23 T9
You mention that you would normally raise this texture in 100% frequency. I know that in theory, we are going to lean towards more calls without bdfd, and more raising with bdfd. So I wonder if you have any exploitive reasons to think that its just better to attack wider in this spot against general population?

@22:32 QJ
I would like your turn analysis very much if you would've played stricly against regular, but looking at your pool, there seems to around 50% chance that this new player is either really big fish or some other weak-passive type. So for that reason, I think I would always prefer to bet with your exact handstrenght.

@26:39 Q8 analysis
I would think that ch/folding any AX in this spot on the river is going to be really very suboptimal. I think you're making really very haphazard assumptions here based on such a limited information as Villain's PFR, assuming that its going to automatically translate into his postflop passivness.

Im going to agree, that in general, 50/5 is going to be more passive by average than 50/20, but these are still going to be two profiles without any clue, making bunch of erratic and volatile plays. Thus, I feel like folding anything as strong as Ax will get us into trouble here.

Also, after watching this analysis, I went to my HoldemManager to see average AFq of players that are playing high VPIP/small PFR. My criteria for a search was VPIP>30 and PFR<6 on a sample of more than 80 hands, and the average AFq of first ten players that I could find was 33%. Which, given higher VPIP than average regular, is making them still quite agressive.

BTW, even tho I dont agree with everything I see, I really like your videos and you're easily my favorite essential videomaker right now.

I really loved the closing part of the video. If I could have any requests, I would love to see similar work, but with raises against cbet flop in 3bet pots from IP player. These are not really plays that are very optimal in theory, but I see them all the time at tables, and Im kinda confused on how I should approach it.

Sept. 1, 2020 | 5:37 p.m.

@4:50 AK
Dude, thats such an obvious x1,8 bet on the flop. At least 0,015 higher EV than your lazy sizing.

Aug. 26, 2020 | 2:36 p.m.

EV difference based on what?
How he's supposed to get his stack if he's not raising the size OTT?

May 15, 2020 | 2:06 a.m.

I think the mere fact that any of us are "grinding out RIO points" is
pathetic. And I don't care how many feathers I ruffle, nobody else
seems to.

Imagine someone playing NL5 or NL10. Maybe that person also lives in non-western country, so 100$ is actually a lot of money to him. Either 20 or 10 BI at his stakes. The only way he can reasonably get it, is by consitently creating quality content via his posts, that are helpful and valuable for other members. So he goes for it, and is putting deliberate and consistent effort through the whole month.

This is that pathetic to you?

May 4, 2020 | 7:25 p.m.

I absolutely understand your concerns Eldora and my opinion is that these things should absolutely be resolved under closed doors, so to not damage any reputations. I am not sure if lIlCitanul and forCarlotta tried that way first, but I hope they did.

And yes, I agree that I was making claims with quite a lot of confidence that might look unjustified, but I guess it was mostly coming from the fact, that I've been in this very race we're talking about several times (was top5 around 6times in recent years) and during this time I naturally spend some time observing how it works. Because of that, I dont agree with this statement:

Similarly, Flight_Risks occurence was sometimes described as a magical
'1-day-climb' whereas in reality he posted regularly all the way back
from becoming a member in March and also posted in the beginning of
April already. Then he started to up his frequency around April 20th
without changing his posting quality though. Just an observation of
when he appeared in the Top 10 doesn't do this justice.

From someone to climb from less than 42points, to 97 in one day, this is enormous anomaly. I think its possible to happen only if someone posts extremaly popular post, in hot/popular thread, that is liked by bunch of RIO staff. Otherwise, it is just not realistic that over 50% of someone's 92points distribution has a place during one day.

Furthermore, adding the facts that:
- a lot of likes from almost unused accounts [Badabing (1point) and ishflop (0point)]
- and there is clear motive to cheat the imperfect system (100$ subscription at line)

I'm not going to change my opinion that the confidence in which I made my claims were unjustified.

We decided to add an additional sixth place leaderboard winner for the
month of April to make sure that noone will be set at a disadvantage
based on this decision.

This is good decision. The same situation as above happend to me few years back, and I presented my proofs to Mikey Stotz via priv, and he agreed with me and also granted 6th membership. Mentioning it, because its almost certain that it will happen again. Maybe its not a huge problem, given that the community is not that big, so there is room for micromanagment, but the current system is absolutely abusable, and I would feel really bad for forCarlotta if he did become a victim of it.

All that being said, I really feel bad for you being put in this difficult situation Eldora, I understand that as a site, its almost like being a judge, you cant really make binding decisions with a free hand even if the confidence interval that someone is guilty, is really high, without giving the benefit of a doubt.

I think its also another argument why the system should be somehow reworked.

May 4, 2020 | 3:05 p.m.

Eldora
So let it be clear, you are now allowing account's point boosting by other accounts, as long as they're not from the same IP and you're not going to punish this behaviour in the future? Because this is clearly what happend here in this situation.

I think you're also missing a point here - I think the evaluation of the one's contribution to the forum here is really besides the point, and no one was questioning it.

It is more about the point/RIO Elite giveaway system being fair and rewarding the most helpful and quality posters. If you allow situation where the account gains most of its points in the last few days of competition, clearly boosted by two accounts that are not even active at the forums, your message is that you don't really care about any of that.

And I hope that you can see clearly see how potentially harmful to the forum it can become. If you're not going to react to LAZY attempts like this one (and it was lazy as heck, he could as least get himself boosted regularly during whole month by these two accounts), the whole rank-giveaway system can easily turn to just boosting-shitshow.

May 2, 2020 | 3:07 p.m.

You overbet river claiming he doesn’t have a 9 while you do have. Can
you elaborate a bit further because I can’t figure out what 9x would
you have

T9s/98s are propably being called with some frequency prelfop, just as any other broadways.

Although I agree with the second part of your post, that Peter shouldnt be really here with a lot of sets.

So, really props for this regular for recognizing that:
1) Hero range is incredibly narrow
2) If he has any suited broadways in his range with same frequency like T9/98s, he has like 2,5:1 ratio of potential bluffs to valuehands for his overbet.

Jan. 10, 2020 | 3:01 p.m.

Yup.

Whats the weakest hand you would continue here?

Dec. 13, 2019 | 2:23 p.m.

Was very excited about making an 'E for production' joke as you messed up first hand prelfop action, but then you made it himself later on ;(

@Last hand - QJ

Do you think its relatively better or worse to bluffcatch in this spot against regular?

Dec. 9, 2019 | 8:04 p.m.

Yes, very profitable. To give you better perspective, I would guess that they're profitable to the extent, that we still call all pocketpairs against x3,5 and I would guess also x4 and slightly larger than that too (at least 55-44).

Its always uncomfortable to play a baby pocketpair OOP, but the reality is that we're just always going to have a pair against 25%-55% range, and thats really something. On top of hitting sets, obviously.

And, also - while defending from BB, you only need to lose more than -100bb/100 to make a relative profit. So if you feel like you're not winning with these, its propably because you dont, but relatively its still better than folding.

Dec. 3, 2019 | 1:47 p.m.

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