steamer's avatar


402 points

Post | steamer posted in Chatter: Balance RIO spamming range.


Changing from never to every-fucking-day isn't the most EV - obviously it's better than never :D

I filled this in on my mailing unsubscribe page too - but I wanted to let you know directly.


April 21, 2019 | 9:02 a.m.

I'm a training course addict and I'm going through recovery so I won't be buying it just now but the outline seems like it would work!! As long as you take action - which is my failing :(

Good luck with the program. Have fun!!

Feb. 14, 2019 | 11:19 a.m.

Chris, did you have FGS on or off for these sims?
And if it was on then to what depth?
Edit: nvm I should have listened another 60secs :)

Think I'll do a couple of these with FGS for my own info. Thanks for the ideas.

Sept. 8, 2016 | 11:04 a.m.

Steve, thanks for your theory videos. I enjoy them a lot.

This topic is one I looked at a while back after reading some random 2+2 post about "bunching". Because I don't have your math and excel skills I used PPT to try and find the answers and I would like to share the method here in case it's of interest to you or any of the other commenters who are interested at looking at different situations, card distribution etc.

You can use the same method to answer the same questions for any of the games PPT simulates. (Razz, PLO whatever).

  1. Enter the folding ranges for the early positions - ie the inverse(converse?) of their opening ranges (you can make them as complex/weighted as you like).
  2. Add the questions you would like answered in the panel below. (Alternatively you can use PQL to create really complicated ones - but it's probably overkill.)
  3. Use preferences dialog to set the iterations high enough to get a sensible result.
  4. (Optional) Use the csv input/output facility to get a distribution for all 169 hands. (This is one to start running before you go out - at ~30secs a sim it'll take an hour and a half, on slower PC/higher iterations 5hours+).

I found a screenshot I took, which doesn't have particularly sensible ranges but it should give you the idea.

The hand range figures are self explanatory (about 10% higher in FR HU when folded to). After that is: rainbow boards (no change), flop no card higher than a T (12% less often in FR), straight friendly boards (no change), flops with a single A (10% higher in FR).

Sept. 4, 2016 | 4:52 p.m.

I don't see any harm thinking about these things :)

  1. The number of bluffs you would need depends on your 4B sizing.
  2. AQs/AQo/AJs/pairs pick whichever of those you would include in your folding range.
  3. Pick whichever you like - usual practice is to use the top of your folding range - but you might prefer the lower sc's. As you're folding them to a 5B their AI EV is not important only the number of combos.

Aug. 26, 2016 | 9:45 p.m.

Comment | steamer commented on multiple equllibria

Not sure I'll get these right, but I like the questions so I'll give it a go.

1. Why is there multiple equillibria even in simplified poker games?
2. Is it just the difference between very similar strategic options like raise/call nuts when he's only calling your raise with nuts etc.?
3. Do those equllibria have the same game value (pay-off)?
4. What if i play a strategy that is in equillibrium but villain also plays a strategy but from a different equllibrium?
5. Will any two opposite equillibrium strategies also result in an equillibrium?
6. Can I trick my opponent by showing him I play one equllibrium strategy and then switching to a different equillibrium strategy?

1. Even a simplified poker game has a (relatively) complex decision tree.
2. Don't really understand the question. Any strategy will have to consider all options.
3. Yes. Optimal expectation. "Optimal" is probably the wrong word, but what I mean is: the best possible expectation that can be realised, regardless of our opponent's strategy. (Zero in a lot of cases)
4. Both strategies have "optimal" expectation.
5. Yes. Though they're probably not opposite, but different. An opposite one would be the other half of a pair - so strictly part of the same equilibrium.
6. No.

Sept. 23, 2014 | 4:19 p.m.

Enjoyed this a lot Brian. I wouldn't mind seeing some analysis of the mixed game spots - we can all learn together :)

Aug. 5, 2014 | 3:29 p.m.

I see way too many people r/f off <10bb or r/c hands they should shove QJs, A4s, KJo to worry about this too much. Try doing a filter on your db and see what they're doing in your games. Sure he'll have a monster sometimes but even a stopped clock's right twice a day.

July 30, 2014 | 12:36 p.m.

If he's been regularly minraising and not shoving off this stack and particularly if we have ever seen him r/f then I'm shoving this and a lot wider. In fact he'd have to have shoved every other time for me to think about folding.

July 30, 2014 | 11:58 a.m.

Range should be wider. w antes the fact he is only raising into one player(not two) > half a bb less in the pot.

July 30, 2014 | 11:49 a.m.

You can work all this out with ICMizer using the mtt facility and adding the stacks on the other table. Not easy to do in game.

In practice I'm going to let this nut go nuts and nit it up until: I have a monster (KK+), he gets down to a smaller stack, I have a shorter stack, or a lot of the short stacks have been eliminated. To look on the positive side, we have position so can act when he folds and if he wants to donate $ev to the table by trying to take everyone out then that's very helpful. 

is it better to shove? or induce? or possibly even call?

I'm going to fold. AQ has blockers which don't matter much because villain is bonkers and even if we induce, he shoves and shows us 64o we still can't call. In theory if we shove he should fold almost everything but if he will call with 87s or KJo and similar we shouldn't shove. Yes he's making a huge ICM mistake but we're still screwed. Calling is really meh, you don't flop QQx often enough ;) and even on A high flops you aren't far enough ahead to fistpump call a shove so calling is wasting chips. Plus calling stops the shorties from getting involved with villain - let them take the risk, they have to and it costs them a lot less.

if we assume villain knows nothing about ICM how do we adjust?
Mainly as above but you can put his ranges into ICMizer and work it out.

at what range can villain 4b shove and our call will be break even.
Never - at 100% we're losing $. This is a guess (you can work it out, depends on the size of the induce - but if he'll always shove, inducing too big is the mistake anyway not calling.)

July 30, 2014 | 11:33 a.m.

It's a reasonable prejudice given that 1. Poker is a negative sum game due to the rake and 2. The number of losers greatly outnumber winners over time.

The chances that a poker player, including those with positive results, but without a statistically significant number of hands (probably several million) is a losing player are really high.

July 30, 2014 | 10:49 a.m.

Comment | steamer commented on final table $3 rebuy


July 18, 2014 | 9:06 a.m.

Comment | steamer commented on final table $3 rebuy

Depends on stats: Villain's VPIP/PFR/CB/TCB/Fto3B/4B & Call Open/Squeeze of players to act. I'd never jam this in a $3 unless villain is one of the idiots who calls w J9o. Almost always flat (flicking in 2.2bbs/folding won't change your $EV much - you could use an ICM calculator to see how much), sometimes 3B induce (either a 4B or squeeze jam) and sometimes fold if he's ridic nitty - even that's unlikely because we win those post if he misses w AK - maybe fold if he's one of those nits that open tight and CB big 100%.

July 17, 2014 | 12:47 p.m.

Good point about the game being unbalanced. In my mind an mtt is like playing loads of different cash games with different structures and you have to adjust instantly.

July 17, 2014 | 12:33 p.m.

Well, I want this to stop. How do I get better, where do I first look? Where do I even start?

This got kind of tldr; but my suggestion is at the bottom :)

Yes MTTers suck (at cash) but if this is true then: "Cash players suck at MTT's" is also true. Because what we're really saying is that different skills apply to cash and tournaments and that someone who has devoted all their time and energy to one, is expert in that at the expense of expertise in the other.

Ralph points out that you need to study more to do well in cash. I think this is true because the lowest skill level at which you can make a sustainable profit is probably higher in cash. I would suggest that to be among the best requires even more work in MTTs because you need to know everything that cash players do and everything that relates only to MTTs. (like anybody ever had that amount of time!)

Your biggest problem is that you have everything you need to play cash profitably, learned at the level of unconscious competence - it's automatic to you and almost all of it is wrong for MTTs :)

It's a completely different game because NEVER (except heads up) does cEV = $EV and the relationship between them changes constantly and sometimes drastically. That is counter-intuitive, particularly for long term cash players. 

If I asked you what range you should 3B/call a BU 3x w antes in the SB, given villain's range and BB's stats, w 30bb eff, you could tell me without thinking. Any mtt player is going to want to know the answer to loads of other questions first because they change the right answer. (WTF? How can the "right answer" change?) Eg What are the payjumps? How soon will the blinds increase? Do I want BU to survive to prolong the bubble? What is the utility of my stack if I win(lose)? blah, blah....
Playing deep, your cash expertise will overcome your MTT deficiencies, but shallow the reverse is true. That's a very large problem because all the big $EV decisions happen at shorter stacks. It does mean that you can probably do ok in deep mtts but this is because you're effectively getting to "buy" 2 or 5 or 10 stacks for the later stages by virtue of your early game play.

Where to start:
1) Watch some of Akira's earlier videos (can't remember which sorry) for good tips on paying attention to stack sizes and what options are open to you at different stack depths.
2) Check out the comments threads in Nick Rampone's essential vids and watch (at least - I'd watch them all) the ones that have a discussion about ICM in the comments.
3) Have a play/study with the web calculator at Holdem Resources.
4) Get an ICMizer subscription. You can do 3? a day free but it's not enough.
5) Sit and think about different hands/ranges in "edge" cases. Eg. KK on the bubble of a satellite
6) Play some micro hyper/turbo SNGs and use holdem resources to build ranges for all the common spots and ICMizer to analyse a load of hands - the chart hand against range is very useful.

July 17, 2014 | 12:20 p.m.

Comment | steamer commented on Staking

Yes. But it would be a really long recursive equation for each pay spot. Assuming you invested regularly, it wouldn't make a difference and if it's just a one off it doesn't really matter unless you are buying such a high percentage that the player will need external motivation to climb the payouts. But then you'd be better picking another horse.

July 17, 2014 | 11:05 a.m.

What Tom said. Or you 9x and when you(2) folds you sit out because he's too reggy.

July 15, 2014 | 9:20 p.m.

Seems standard. Just one of those things. Do think you should CB though, don't want to give a free card or make the board any scarier for worse hands that will call or raise now.

July 15, 2014 | 9:14 p.m.

Comment | steamer commented on hot44 - snapcall?

Saw this question again and checked out just how tight UTG+2 would have to be shoving for us to lose chips. We make chips at 8%.

July 15, 2014 | 11:29 a.m.

Comment | steamer commented on hot44 - snapcall?

Call AINEC cEV for 9h * 67k stacks is 33.9%, 22+ Ax+ K4s+ K9o+ Q8s+ QTo+ J8s+ JTo T7s+ 97s+ 86s+ 76s Holdem Resources Payjumps are at 18, 15, 12 and FT iirc so not much ICM implications.

July 13, 2014 | 2:08 p.m.

@Raphael: Variance does depend on field size. You make a good point that the large field MTT's have worse players and therefore our ROI should be higher. At a certain improved ROI our variance will decrease because the increased ROI overcomes the variance caused by the larger field.

The factors involved in differences in variance are: ROI, field size and prize structure.

You can think about edge cases like winner take all or only final table paid MTT's and how difficult it would be to realise even huge ROI's reliably, in big fields with these structures and compare that to the very small ROI's the best HUSNG hyper players make a great living from.

In any event, the tournament variance calculator (link in earlier post) works it all out for you - the only hard part is estimating your ROI accurately in a given format.

July 5, 2014 | 10:29 a.m.

You can work it all out with this calculator. Tip: You can put different types of MTT's in a single calculation to see how best to balance higher variance types with lower variance and work out what you are comfortable with.

July 4, 2014 | 5:27 p.m.

I feel for you vanway - I think it's really hard to get over thinking of these hands as bad beats or bad luck. Truth is you're 60/40 which means you only win 6 out of 10 and more importantly if you have all your chips in as 60/40 4 times you are only going to survive 2 out of 10 times. (.6*.6*.6*.6 = .216 = 21.6%)

You played the hand fine, you should be happy :) But if you worry about it at all then I think you're right - you don't have the right mindset. I hope you can get it, but if not, so what? No point beating yourself up about something that's painful - give it up and do something you enjoy.

July 3, 2014 | 11:47 a.m.

Thanks Sam. Appreciate the input. I'm trying to see this from the point of the "pro-shovers". Checking it out in ICMizer (cEV) it looks like we win 1.5-2bb w/o antes or 2.5 -3bb 9max w 12.5% antes by shoving (depending on exact ranges). So if we make less than that by calling we should shove?

Which now that I worked it out makes me want to shove if there is enough in antes.

June 29, 2014 | 2:03 p.m.

You're right about the equity for sure - instead of just using the Sklansky quote I should have probably said K8s or A3o or some other hand with better equity if called but low playability OOP.

Stack sizes, dead money and FE are the same regardless of our hand and tbh in the above situation it's unclear as to whether we're value shoving or bluffing - 66 is a bluff JJ is value, 88? Who knows.

My point really is that you're taking a hand which can profitably call and turning it into a bluff. If, as Sam says you'll see some ridiculous folds (seems right) then jamming is more about fold equity than showdown equity and some raggy hand which has no post-flop playability but a blocker seems like a better one to do it with.

Full disclosure: I'm an older guy who looks like villain so I never shove 88 here live because I'll be in horrible shape when called - but I'll happily jam 76s and expect villains to fold AQ and JJ face up. I've had people sigh call with KK :) 

June 27, 2014 | 12:56 p.m.

Why not try the free trial and see for yourself? :) I'd recommend it.

June 27, 2014 | 1:15 a.m.

I'll add my vote for Nick's videos.

Also Akira has a way of describing not only what he's doing but also what he would have done if things had been a bit different and also some good insights, well explained, about how changes in stack depth effect your strategy.

If you're transitioning from cash I'd recommend ICMizer very highly - just work through tons of spots you'll see in even the ss 180's - getting past thinking in cEV at all is the major change from cash, imo.

June 25, 2014 | 8:14 p.m.

It's only face up if we only jam AK - which would be kind of silly. TBF a lot of people only do that and for a lot more than 30bb. I love them :)

No idea what good ranges would be, obv depends on player pool and dynamic, probably good homework but I'm feeling lazy - So I'll guess at:
jam TT - QQ, KK+(red), AQs, AKo;
flat KK+(black), AKs(red) + some other pp and stuff;
3B/c rest of KK+, AKs;
3B/f AQo (50%) and suited A to taste.

June 25, 2014 | 7:55 p.m.

Comment | steamer commented on Range Advantage

Thanks James. I've wanted a clear explanation of this for ages.

June 17, 2014 | 11:24 a.m.

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