Rather late, but another analogy for jewbinson and others - what if, rather than having ~85% equity against ATC, AA had more like 99% equity? Is ATC still a favourite? A way to conceptualise this might be AA versus ATC on an A22 flop, in which AA has ~99.5% equity but is not the nuts. Even if stack sizes were arbitrarily large, it's just not possible for ATC to bluff enough to get an edge.

Also, reading mersenneary's recap and remembering my thoughts on the durrrr challenge made me think that a "fixed # of hands, whoever is up at the end wins" regular HUSNG would be really cool. The endgame asymmetry as one player tries to increase variance while the other tries to reduce it would be both interesting and likely to result in big edges for good players.

What would be some good a) fixed stack sizes and b) # of hands played? The goal shoud be to prevent tournaments from being unduly long or short, so I think that the number of hands should be >1.33 the stack size in BBs (to prevent a player from clinching the win after the first hand) but the number of hands can't be too large (otherwise the games might last too long, see no blind increases HUSNGs). Other factors like "roundness" should also come into it.

My first idea is 30bb and 80 hands. Round numbers, not too large or small, where stacking your opponent twice will clinch the win. I'd perhaps prefer a slightly larger stack size (preventing too many all-in pots which are all the same size) but the next relatively round numbers with these properties are 45bb and 120 hands, which I think is getting a little too long. Perhaps 50bb and 100 hands, which is "rounder" except a little less aesthetically pleasing in getting ~2/3 of the way to the win with a stack-off versus ~1/2 with the others.

This summed up my opinion on this game

Aces will be winning small pots + bluff catching, the non-aces player will be winning big pots, and losing some big pots on occasion due to bluffing (which he has to do for the sake of balance)

I don't know the best sizing for ATC. Let's assume that if he bets, he'll uses this one:
OTF 3bb into 2 bb on flop
OTT 16bb into 8
OTR shove 80 into 40

Now we're gonna work backwards:
OTR ATC has the best hand 15% of the time. When AA is faced with a bet he'll need to call 80 to win 200, so he needs to be good 40% of the time.

--> When ATC has a value range of 15% of all possible combo's, he can add 15%/60 * 40= 10% bluffs.

--> When ATC bets the 'top' 25% of his range OTR he's unexploitable and we'll assume AA will make the 0EV fold.

On to the turn. We see a 2x potbet here as well, so AA needs 40% equity again. Given that AA has to fold the river given that ATC bets the top 25% hands, he needs some rivers to go check/check to make a profit. How many rivers? --> 40% of them.

That means ATC's turn betting range consists, next to his top 25% he bets the river with, of another 40% bluffs he'll check the river with:
25% / 60 * 40 = 16,7% bluffs
--> ATC can bet the turn with his top 41,7% and AA will be indifferent between calling and folding OTT.

Now the flop; ATC bets only 3 into 2 bb, so AA needs 37,5% equity here. Again: we saw that ATC can bet the turn with his top ~42% unexploitable, so the equity of AA will come from the time ATC checks the turn (and the river).

ATC bets OTT his top 42%, so we'll add another 37,5% bluffs here as well: 42% / 62,5 * 37,5 = ~25%
So ATC can bet OTF his top ~67% and win 67bb/100 this way!


Alright, I have to admit this is not entirely true;
- Most important: it's really though for the ATC player to select his top 67%/42% and because we can't see the turn and river we'll often make mistakes and won't beat AA 15% on the river, given that AA picks up on our strategy and bets us of our 'air' hands OTT.
- Secondly this is pure theory. In practice we can't say "ATC beats aces 15% of the time OTR" We need to adjust given the board; on dry boards we haven't got a lot of value hands, so we need to bluff less hands etc. Things get complicated fast, especially on the turn I guess.


And for the people who got confused by the math above, another way to show AA can fold on 67% of flops given this strategy:

FLOP 2bb:
ATC bets 3bb / 67% of hands.

Assume AA calls, he invested 3bb

TURN (8bb)
ATC bets 62,5% of the time or the top 42% of all combinations.
AA immediatly profits 37.5% * 8bb = 3bb when ATC gives up,

If AA elects to call he'll invest another 16BB and gets to the
RIVER (40bb)
ATC bets 60% of the time (or his top 25%).
AA wins the pot when ATC checks: 40% * 40bb = 16bb

If AA calls he'll win vs the 40% bluffs for a total of +120bb
or loses vs the 60% value range for -80bb

Which makes him once more indifferent between fold & call.

yeah i agree that this would be a massive levelling war and a balancing act. very crazy idea and i love the thread. i dont even think it comes down to mathematics here its more levelling. that AA hand on the 4 flush board is sick tho, nice shove, nice call, but its essentially 50/50 to fold or call right?

GTO play is obviously to go allin on flop (also on turn if you forgot to go allin on flop). You win the pot 100% of the time. You can't do any better than that. Doing anything else is just hoping you trick him into calling incorrectly (or actually allowing him to call correctly), he can always just fold 100% anyway and do as well as he did vs the allin plan.

Edit: Scanned the thread a bit more, I see this has already been discussed. If you think anything other than shoving is GTO, you're confused about what optimal means. It's possible that you can maximally exploit your opponent by betting less, but you're counting on him to make a mistake, which I think he did. That's exploitive play, not optimal. With your actual betsizing, it's likely that his correct response is to fold everything other than straightflush draws and a small fraction of his 76o, but continuing with only those hands would be better for him that facing a shove and having to fold them too.

fun thread. did this ever get finished or did they just talk a bunch of theory in the end?

They played 300 hands ATC won but really hard to draw any conclusions from that.

If people are really interested I may try to make a web based version so people can play without the whole AIM thing. Would have to wait till the weekend though

Despite the arguments to the contrary, I still think AA should be betting some non-nut flops and turns but never rivers obviously. It seems like there are too many times where AA has huge amounts of equity and should not allow ATC to realize all of theirs. AA are bluff catchers in some sense but they're also huge favorites on a lot of flops. On a K82r ATC are ahead ~3.3% but have a little more than 11% equity. If that particular board gets checked down AA's EV is something like +.8BB given that ATC will have the best hand 11% and bluff us out a similar amount. It's hard for me to imagine how AA doesn't do better than +.8BBs here. On this board it just doesn't seem like there are enough made hands for ATC to bluff AA out. If AA bets pot, and I'm not arguing that's the right size, can ATC actually raise profitably? And if not, why wouldn't you want to bet?

Hey, I remember this now. Does anyone else like it?

Cool thread.

I will give whoever on here AA and the rules is you can only limp. Meet me at any casino in the sf to sac area we can play whatever limits you want. I will let you take any 2 aces out of the deck that will be your starting hand and then we will shuffle the rest of the cards and deal them out. We play 4 hrs minimum. At least 1000 nl. Message me if your in the sac to sf area and we can set this up at the casino i can get a dealer to do it for us and it will only be a 1 dollar drop per hand.

exactly but im willing to to give anyone those odds i think my post flop skills are good enough that it will make it a interesting game. So message me all meet at any casino and the dealer will set up heads up matches to 1 dollar drop so its not to bad on the rake. but its got to be at least 1000 nl you have to put at least 1000 dollars up the rake is definitely going to eat away at our stacks playing every hand. So i wont do it for less then that.

It depends on how deep each player is. If the players are deep, I expect this to be ev-. If the players are shallow, ev+.

All the non-AA player has to do is to limp every hand and wait for two pair or better. The AA player will not know if this has happened. The non-AA player can then simply add in a few air-bets.

All it would take is a single push from AA at the wrong time for the non-AA to seriously hurt the AA player.

AA is strong, but not that strong if your opponent knows what you have.

Shouldn't ATC only limp 100% of the time if he believes he has a post flop edge? With A6o we are only 6.44% but need 25% equity to make limping profitable. Assuming it gets checked down to river we made a mistake by calling pre.

The best thing to come out of this thread for me is seeing that jewbinson last posted in February