Shoving double suited aces preflop

    • L3ST
      L3ST
      Bronze
      Joined: 21.01.2010 Posts: 413
      Grabbed by Holdem Manager
      PL Omaha $0.06(BB) Pacific
      SB ($2.64)
      BB ($5.94)
      UTG ($6.81)
      CO ($6)
      Hero ($13.71)

      CO posts $0.06

      Dealt to Hero Q:club: A:diamond: T:diamond: A:club:

      UTG calls $0.06, CO checks, Hero raises to $0.33, SB calls $0.30, BB calls $0.27, UTG calls $0.27, CO raises to $1.89, Hero raises to $6.93, fold, fold, fold, CO calls $4.02 (AI)

      FLOP ($12.90) 2:heart: 8:heart: 8:spade:

      TURN ($12.90) 2:heart: 8:heart: 8:spade: 4:heart:

      RIVER ($12.90) 2:heart: 8:heart: 8:spade: 4:heart: 2:club:

      CO shows 9:spade: 4:spade: 2:spade: 4:club:
      (Pre 29%, Flop 20.6%, Turn 90.0%)

      Hero shows Q:club: A:diamond: T:diamond: A:club:
      (Pre 71%, Flop 79.4%, Turn 10.0%)

      CO wins $12.35

      What I get for jamming good AAxx preflop. EV line up, green line down. 13 hands she lost 2 stacks. 1 hand I make up half her losses. Couple of hands later, she spews the stack over to another reg making it harder to grind it back. Another scenario is fish doubles up and leaves, money being long gone by the time I have a hand to take it back with.

      Ribbo: what's your oppinion on jamming double suited aces preflop?
      Had a discussion on the study group about this and someone stated that PLO is all about small equity differences so I should just go all in preflop if I can.

      I'm thinking risk/reward ratio here. I'm sticking my 100BB in to win 200BB. Statistically I'll be favourite against many hands. So I can get a reward of max 20% when I'm favourite and chop many times since villain might also have AA.

      I have a problem with this. Flipping in NLHE is ok since villain hits or he doesn't. But the combos of possible hands in NLHE are way less.

      If we get less than 10% profit in shoving AA preflop on long term, wouldn't a safer aproach be better since postflop edges are way better by taking a flop and winning more often when villain shows up with the 2nd best hand?


      Now that's a huge wall of text and maybe not the brightest when it comes to variance, but how big of a sample will we get to justify the huge variance that this move will bring?
  • 11 replies
    • Waiboy
      Waiboy
      Bronze
      Joined: 18.09.2008 Posts: 4,877
      Fundamental question ITT!

      Why in a game where most of the play and major errors occur post flop, would you try and push marginal pre flop edges and increase your variance v trying to setup post flop situations with much bigger edges v weak players?

      Looking forward to a more authoritative argument than I've been able to come up with.
    • Laggsy
      Laggsy
      Bronze
      Joined: 04.09.2010 Posts: 301
      Ihave this same question. Looking forward to replies.
    • Ribbo
      Ribbo
      Bronze
      Joined: 25.06.2010 Posts: 6,157
      Originally posted by L3ST

      If we get less than 10% profit in shoving AA preflop on long term, wouldn't a safer aproach be better since postflop edges are way better by taking a flop and winning more often when villain shows up with the 2nd best hand?
      Your line of thought makes no sense given:

      Hero shows Q:club: A:diamond: T:diamond: A:club:
      (Pre 71%, Flop 79.4%, Turn 10.0%)

      The example you gave did not back up your comments did it? This is a case of humans having a selective memory.

      From a mathematical standpoint, if you're 71% favourite preflop, then you're going to be favourite on the flop 71% of the time as well and since you have no way of knowing if you're ahead or behind on flops you miss, all you're doing is complicating the game unnecessarily.

      Let's say you dont jam and the flop comes 9 :heart: 8 :heart: 7 :heart: and the fish with 9 :spade: 4 :spade: 4 :spade: 2 :club: shoves, what are you doing?

      I'm guessing you fold the flop as 80% favourite.
    • SirVanek
      SirVanek
      Bronze
      Joined: 02.08.2008 Posts: 1,195
      Huh? Give me 10% profit in pre 100bb AI situations and i'm going AI everytime.
      And you have more than 10% profit do to dead money.

      While it sounds nice that Omaha is a postflop game, but come on, if i was better postflop , i wouldn't grind micros ... :)
      There is a lot of flop where 4 handed you wouldn't know where you are. Even if you hit one fd, still a lot of time you wont have bigger % differences than preflop.

      If you don't raise, pots stay small -> their error is not that big postflop -> you win less.


      And yes, there are periods when in 50-100 pre AI's you are down 2-3k bb, but trust me you will get that back.
    • Kyyberi
      Kyyberi
      Coach
      Coach
      Joined: 09.07.2010 Posts: 10,508
      I told this in the study group. The problem with aces is that you have to play guessing games at flop too often. 7/8 you won't hit an ace, and you have just overpair. They are hard to play postflop, because their playability is so poor postflop. And there are many good flops for you, but opponent will shut down. If he has KKxx, he won't jam Axx flop. Or if flop is monotone, it's hard to get value for your nutflush.

      And in your example, if you call that 3bet you will have usually multiway pot with SPR of something like 0,5. Impossible to have ANY postflop edge there.
    • L3ST
      L3ST
      Bronze
      Joined: 21.01.2010 Posts: 413
      Originally posted by Ribbo
      The example you gave did not back up your comments did it? This is a case of humans having a selective memory.
      The example was the most recent result that got me thinking maybe this isn't the best way to win at PLO.

      I posted this hand last week.


      SB shows 3:heart: 4:diamond: 2:spade: 3:club:
      (Pre 29%, Flop 89.3%, Turn 83.3%)

      UTG shows J:diamond: 9:spade: T:heart: 6:spade:
      (Pre 35%, Flop 0.0%, Turn 0.0%)

      Hero shows A:spade: A:diamond: K:spade: Q:spade:
      (Pre 36%, Flop 10.7%, Turn 16.7%)

      SB wins $20.38

      He hit the flop hard with set+redraw to higher FH.
      Since we're so close in equity 3way, how many times do I have to flip with these guys to break even after the loss?
      Will they flip with me again?
      Probably won't play them again since they will bust being overplayed post-flop by other regs.
      Other regs won't go all in with 4332 rainbow.

      So in this case I risked 6.8$ to win 20.4c. For 3% win over a large sample that I won't ever get.

      So, in BB: I'm putting in 68BB to win 2BB over large sample. Even if it would be a normal distribution on let's say 200 hands. We do it all the same. I'm winning 2BB for every 68BB put in the pot. With 3% ROI that's a side of variance I don't wanna take, expecially since typical BRM for PLO is about 50Buyins.


      Originally posted by RibboI'm guessing you fold the flop as 80% favourite.
      Yes, since the flop is no good for me I'll probably fold.

      Let's have another example. Unsuited aces, guy made a mistake min-raising instead of jamming let's say.

      Grabbed by Holdem Manager
      PL Omaha $0.06(BB) Replayer
      SB ($6.84)
      BB ($10.46)
      UTG ($15.39)
      UTG+1 ($10.68)
      UTG+2 ($3.21)
      MP1 ($10.03)
      Hero ($13.16)
      BTN ($43.91)

      Dealt to Hero Q:diamond: K:diamond: A:spade: 7:heart:

      fold, UTG+1 raises to $0.21, UTG+2 raises to $0.72, MP1 calls $0.72, Hero calls $0.72, BTN calls $0.72, fold, fold, UTG+1 raises to $4.41, UTG+2 calls $2.49 (AI), fold, fold, BTN calls $3.69

      FLOP ($13.56) 2:heart: 6:heart: K:club:

      UTG+1 bets $6.27 (AI), BTN calls $6.27

      TURN ($26.10) 2:heart: 6:heart: K:club: 5:diamond:

      RIVER ($26.10) 2:heart: 6:heart: K:club: 5:diamond: A:diamond:

      UTG+1 shows 7:diamond: A:heart: A:club: 7:spade:
      (Pre 44%, Flop 40.2%, Turn 0.0%)

      UTG+2 shows 6:diamond: 8:heart: Q:spade: T:diamond:
      (Pre 38%, Flop 30.9%, Turn 0.0%)

      BTN shows 4:club: 3:spade: 3:club: 3:heart:
      (Pre 18%, Flop 28.8%, Turn 100.0%)

      BTN wins $24.80

      BTN was super aggro and spewy playing 90/35/28.8 1.7 aggresion over ~300 hands sample. Running super hot that day, he shoved a gutshot and won. Didn't mark other hands but this was more than stupid so it deserved the checkmark. This guy had a 284BB/100 hands profit, winning 52$ over 300 hands sample.

      Overall on 400 hands he has 42$ profit with an EV of -9$. About 713BB vs -59BB EV.

      Vs the same guy, I have a 65$ profit with an EV of 79$, about 934BB, HU and shorthanded mostly playing him post-flop, over ~140 hands.

      Bottom line is: for me it looks like too much of a risk for too little of a reward, while playing this post-flop at micro-limits is way safer. Even tough my 3bet range and early open-raise range might become exploitable, I'm trying to balance it with rundowns.


      SirVanek: 10% was an estimate, at best. Can't give you that, I'm a micro stakes grinder with aggresive BRM :D
    • Waiboy
      Waiboy
      Bronze
      Joined: 18.09.2008 Posts: 4,877
      Hey L3ST

      Firstly, this is an important thread, and I'm sure there are always going to be folks that struggle with the logic. I know I tried to get to a coherent argument and failed before my first post. Hopefully this one is a bit better.

      To your last example above, the problem is that they could also show up with AKK* and AQQ*, in which case you've got a massive edge in a massive pot:

      AA** 49.6778% Wins Hi: 49.2890%, Ties Hi: 0.8078%
      AKK* 26.0872% Wins Hi: 25.7091%, Ties Hi: 0.7865%
      AQQ* 24.2350% Wins Hi: 23.8400%, Ties Hi: 0.8203%

      Sometimes you'll go in with AA** and be against AA** and T987ds and be behind.. but with the dead money in the pot by the time the action comes around again, you have the pot odds to get it in and you lose money by calling and folding on ugly flops with the best hand or sufficient equity to make the call +EV:

      AA** 28.3340% Wins Hi: 15.0774%, Ties Hi: 26.5145%
      AA** 28.4164% Wins Hi: 15.1484%, Ties Hi: 26.5374%
      Ts9d8s7d 43.2496% Wins Hi: 42.9102%, Ties Hi: 0.6800%

      And in the last example hand you posted, you're forgetting all the dead money that make going all in correct and a +EV play. If you don't go allin pre and fold your equity out on (most) flops you miss (as Kyyberi notes, you have fewer flops you can hit and have a lot of equity and the AA** has less playability than with a hand like T987s, but pre flop you (almost) always have an edge.

      It looks to me like some sample size issues and confirmation bias is making you tough to convince.

      If nothing else, understand you have some pretty smart people (and don't for a second think I include myself in this) suggesting that shoving your AA** pre flop is the best play, and offering reasons. In fact, I can't remember any literature or vids anywhere suggesting anything otherwise.

      Probably the better quote from Ribbo is:

      all you're doing is complicating the game unnecessarily.
      Here's hoping this makes some sense!

      :f_cool:
    • SirVanek
      SirVanek
      Bronze
      Joined: 02.08.2008 Posts: 1,195
      These are 3way AI.
      But most of the time you have better equity's than there. Lot of time they will have Axxx, KKxx, maybe even QQxx where you will have a clear preflop equity advantage.

      And please lets not talk about 300 hand samples :)
      I have over 10k hand on some regs, and in that 10k they show profit. But overall they are loosing.
      Or just look at my last month graph.
    • L3ST
      L3ST
      Bronze
      Joined: 21.01.2010 Posts: 413
      Could someone with a decent samplesize please filter AAxx preflop and post a graph with EV line too? in BB, $ results are not relevant.

      I just want to see the success of this move, not the statistical probability.
    • Kyyberi
      Kyyberi
      Coach
      Coach
      Joined: 09.07.2010 Posts: 10,508
      You mean winnings and EV when shoving AAxx preflop?

      edit: well I made this:

      I have sample size of 124K hands, all my nicknames and all limits (tilting sessions, drunk sessions and them all), with 6max/5max games.

      First is the line of just filtering AAxx hands. They do include AAAx and even AAAA.



      And here is the same thing, filtered with preflop all-ins:



      Winnings for all AAxx hands is 271bb/100. And for going all-in preflop its 863bb/100. Both sample sizes are running pretty much exactly with EV.

      863bb/100 is pretty nice winrate.

      And another thing. Going all-in preflop with aces is pretty much always +EV move. Even if you know that it's going to be 3way allin and someone else has aces, there is enough money on the pot already that it will be +EV to shove. There are situations when you might actually fold aces preflop. Tight player opens, and another tight player 3bets. And you have crappy aces and haven't put any money in the pot. But those situations are super rare.

      And making +500bb/100 bulletproof +EV plays is something that you can't bypass in plo.
    • SirVanek
      SirVanek
      Bronze
      Joined: 02.08.2008 Posts: 1,195
      From 53k hand plo 25 sh at Stars , only AAxx filtered and preflop AI.

      As you can see in the first ~25k hand i was loosing in pre AI's with AAxx.





      For the record, AA's where i coldcalled :




      And all AAxx :




      Again, as you can see i have better EV bb/100 in pre AI's than in total .


      One more, where i did 3bet with AA (only 6 hand where i called a 3bet, so no reason to post that) :