you're not gaining 1/3rd of a small bet preflop by calling. there are 3bbs sitting there... your blind and the two other bb's that they put in that will be in the pot in addition to whatever else you decide to make it. that's 3x33% for 1bb of EV based solely on equity.

it's pretty hard to lose 1bb of equity. the value of position in limit is tiny compared to nl especially in multiway pots that were 4bets preflop.

it would be less of a mistake to cap a3o than it would be to fold a3s. (barely)

not that it matters just thought that was interesting.

What ranges are people using for the Button and SB?

In the Stoxtrader/Zobags book they recommend the button opening about 40% of hands when folded to, and the SB defending (by 3-betting) with 20%. That would put A3s in the lower-middle part of the button's range, and or doesn't even make it to the SB's range. If they are following these guidelines you are in serious trouble with A3s.

How wide do the button / SB ranges have to be for the cap to be good?

Also, the section in that book entitled "Why it is better to defend with Q7o than K2o" basically sums up my concerns. Of course A3s is not as bad as K2 but it suffers from the same issues: easily dominated and plays poorly postflop. It's not all about preflop equity.

That book also recommends that you need AJs or better to defend once the small blind 3-bets, unless the small blind is very aggressive... in which case they recommend loosening up to ATs.

I'm working with looser ranges than that but it's not close vs 40 rfi 20 sb 3 bet either.

That book is outdated.

Not sure it’s a matter of it being outdated as it’s a matter of that book considering hands where rake is not trivial. At 5/10 and 10/20 your opening range is tightened up by a decent margin and the sb tightens his 3bet both to reflect that cost of rake and the tighter button open range. The impact of rake is compounded again because you then have to both consider their tighter ranges and the rake itself, which makes a huge difference on the ev of entering the pot. Even at 20/40 the impact is not negligible.

Ok i dug that book out and spent way too much time donking through it and it's quite incorrect in so many spots.

I didn't find anything mentioning rake. Obviously rake effects ranges but their range choices aren't correct in med/high rake environments either.

Ok, so, can you give ballpark figures for:

What % of hands should the button open-raise with?
What % of hands should SB 3-bet when button open-raises?
What % of hands should BB play (cap or call) when button open-raise and SB 3-bets?

That book was suggesting something in the ballpark of 40% / 20% / 10%.

A3s for the BB suggests that his number is at least 25%.

not for free

Then this discussion is meaningless.

preflop wasnt why tbis hand was posted. if you are stuck on that that it should be information that you have work to do preflop. if thats meaningless then ok.

I think you have to consider more than just what people should be doing and actually take into account what people do in practice. If you see someone do something like open 23s or 76o on the button, then obviously you have to consider how that’s going to change the ranges of the sb and yourself.

Not at all. I had the same initial reaction you did - fold A3s - but based on the comments of the more experienced posters, I've been trying to come up with a rationale for why you would cap A3s in this situation. I've learned a great deal from reading about this hand (and the other ones).

One comment I had on an earlier post was that the capper's range is not necessarily all that tight. He just capped A3s, so if he's capping stuff like that, then his range is not nearly as tight as what it would be if he stuck to Stoxtrade's guidelines exclusively. By capping A3s, he does not allow his opponents to put him on a tight range in a similar cap situation in the future. Plus, he's baiting people with better Ace hi in the future to call him down, since he may be sitting on A3. He'll probably get more action on his caps now in general.

We have to assume he has enough experience to know when he's looking at a better ace, if one comes.

Yeah, I absolutely agree with that. If the button is opening almost 100% in that spot, and the SB is aware of that he is probably 3-betting very light, and A3s is in good shape against their ranges.

But are these reasonable ranges? I would think that if the button was way looser than expected it would have been mentioned in the original post.

i’m not assuming out of line button or sb

"But are these reasonable ranges? I would think that if the button was way looser than expected it would have been mentioned in the original post"

which is why this all gravitates back to what is optimal, where those of us without solver data on preflop opens are left trying to decipher the implied opening range for bicyclekicks intimation that A3s is a clear 4bet. he also suggests that a3o would be a mistake so a3s is likely close to the bottom.

the ratios of button/sb/bb aren't 40/20/10, and we can know those ratios are messed up since if button is opening 40 you definitely want to be 3betting more than 20 given dead money and lack of rake, and the 4bet can similarly be wider than the 50th percentile of the 3bettor because you're also up against a wider button range.

if we get hit the numbers dead on do we get a prize? i say 50/35/25 when rake is zero.

There is a disconnect in this thread between posters that are using different tools and data to make their points from each other. Once one accepts the wisdom of the simulators one can’t return to the land of books and hot cold equity estimations.

I think that it’s reasonable to assume a somewhat competent opponent will be opening at least 45% from button and somewhere around 20-25% from sb. If they are somewhat tighter then capping a3s is neutral to slightly losing. If they are somewhat looser then capping it is a huge winning play.

i don't think anyone is disputing what solvers are producing. most of us just don't have the data and are working with what we've got. bk doesn't want to provide the info for free which is understandable, so someone name a price.

is selling the data itself (without the program) second hand acceptable?

I was more referring to the people debating whether to fold pre. It’s hard when people are used to trying to figure out a spot and now the new way is “I’ll ask the dream machine and trust it”.

To my knowledge the problem with accessing solver preflop data is you have to be competent with the solver to even view the answer. I’m sure someone will come along and sell some advanced charts at some point (maybe they already do?)

well, it's a pretty straight forward matrix for limit. you have optimal opens, and then optimal 3bets/4bets given the positions of the openers. I've seen a site that sells the data from a monker sim 4 handed for something like 500$ but the site seems somewhat sketchy / wouldn't trust it plus they ran the sims assuming a 1% rake for some reason.

4 handed is all that you would really need since the remaining positions can be inferred pretty easily with that info.

and it would be useful to contrast different rake structures.
ie: 10% cap 1bb (5/10) to 10% cap 1/10th of a bb (50/100)



the info isn't even that valuable / doesn't hurt the people providing it much considering how seldom big online games run and how diffuse the impact would be for live players. the fact that it's so hard to find a place to buy this data kind of makes me want to assemble something like that.

Not a limit Hold’em player but I read that Stox book a long time ago. The suggested 40% open from button is based upon empirical results of a very large number of hands played from $100/200 and higher. For example if his worst unsuited king is K9o the backup charts usually show a slim profit for K9o and negative amounts for K8o and K7o. Therefore it seems hard to say it’s too tight to fold K8o and K7o when presumably a skilled player lost money playing them over hundreds of trials with those specific hands.

Not sure how this translates to live $200/400 but rake is certainly negligible. IIRC his restealing ranges ranges simply state we should have 50% equity against the opening range so that is just based on equity simulations.

50% can't possibly apply in all situations. consider his ratio of 40/20/10.

imagine folding an 11% hand in the bb because it had 48% against the 3bet. does that make any sense? it also has something like 58% against the button for an overall equity that's clearly >33%, AND you're subsidized by a full 3 blinds of dead money. it's not even close.

the sb would similarly want to enter the pot with less than 50% equity against the opener because 90% of the time he's getting heads up with 2bb's of dead money.

if the initial open was 40 the ratio would be closer to 40/30/20.

The results are instructive as case studies but they do have some problems. Namely those are that strategy in the game has changed and so have the population tendencies. Either way, stoxtrader book emphasized playing a tighter strategy while learning how to play in tougher games and then loosening up. So I think you are somewhat cherrypicking information from the book to support your argument.

I didn’t really think I was making an argument about anything but it’s possible, I certainly have no dog in the fight

Just indicating that at least the initial opens were based on cold hard data and some people may not realize that. Nothing at all was “cherry picked.” But there was nothing mentioned in the discussion of the formulation of the ranges that indicated you should play tighter while you learned. If the results indicated it was a profitable then he advocated playing it.

But your post made me think of a question to ask, if the opens were mostly accurate back then but not now due to changes in strategy and population does this mean the solvers would have been wrong for online play back in 2004? But the solvers are “correct” for 2019?

The better you play post flop the more hands you can play preflop. So the empirically based ranges from this book are optimal for the author's post flop play. Modern ranges with solvers assume perfect play post flop, so keep that in consideration when you consider your individual preflop ranges.

Interestingly, having done some homework in my lab on this particular hand, I believe correct play is to either bet flop, check fold turn or check call flop, check fold turn to a button two barrel. If the button checks back the turn we should check call the river. So while natural on a board texture like this, it seems my double barrel is a mistake.

My iron friend is checking the flop with a large % of it's range here, a play which I'm sure was foreign to high stakes back in 2004. It still is in 2019!