Equity and Pot Odds vs EV Line

    • Verre
      Verre
      Bronze
      Joined: 24.02.2009 Posts: 708
      I have been running into this interesting situation while playing CAP sh nl50.

      Pot odds and equity have no relation with your ev line. And if the goal is to make the best decision on every street based on a players given range on that street and the pot odds offered against the equity of your hand and said range then in reality the EV line in PT4 or HM is completey irrelivant!

      Lets give a simple example to explain.

      You have AsKs in the sb of a CAP (20bb stacks). The button makes it 2bb and you 3 bet to 6bbs. The bb folds and bu calls and we see a flop. The pot is 6bb +6bb +1bb =13bb. Our effective stacks are 14bb

      The flop comes Qs9h2s. We have 2 overs and the nut flushdraw. We cbet 6.5bb and are called by the bu. The pot is 13bb +6.5bb +6.5bb=26bb. Our stacks are 7.5bb.

      The turn is a 6h. For simplicity sake we see opponents hand. He currently holds Th9h. This means we have ~34% equity against his exact holding. The right play is to check for a free river card, but after we do villian shoves his 7.5bb. We have to pay 7.5bb to win 41bb. Obviously to make the call profitable we onlly need ~18.5% equity, and since we have almost double that its a no brainer to make the call.

      Here is where the issue comes in. Your EV line in your tracking software comes along and now tells you that you made a -EV play. You played every street against a range as best as you possibly could, but your EV line still declined.

      With this said, is EV line even worth looking at? I've played a sample size of CAP that is not enough to truly determine anything, but enough that a strong trend has started. My winnings is north of 5bb/100 and my EV is a solid 0 dollars and always hovers there. I started to notice in my session reviews that when I play a great session (not in winnings, but rather in decision making) my ev line never reflects it.

      Another point is that this is also more obvious in a CAP game than it is in a 100bb game. Its easy to move your ev line with non showdown winnings in deeper games and its also easier to not get yourself in high variance all in situations.

      Anyways unless someone else has a good point why I should look at my EV line and that I actually have a major leak in my game and I am just running way above average, I will be turning off my EV line. I would love input on this as these are just my own thoughts.
  • 10 replies
    • carrapa
      carrapa
      Bronze
      Joined: 09.12.2010 Posts: 1,295
      Great post! I've been thinking about it a lot, and you're totally right, I just look at the EV line to see if I'm playing bad or just being unlucky. Usually, I don't think that I'm playing bad if my winnings go up and the ev goes hugely down, because almost everyday I try to improve instead of "being lucky" and play every hand.

      :D
    • Riverdiver
      Riverdiver
      Bronze
      Joined: 12.07.2012 Posts: 1,126
      You know, that's why it's called allin-ev. The line only differs from the green line when you go all-in. If you don't go all in or there is no showdown, the green line and the ev line are the same for that hand. You can't make a +ev decision every time, but when you can't you should make the decision that's the least -ev. If your ev line goes up over a big sample, it means you're probably a winning player and that if you play enough you will make money from the game
    • Verre
      Verre
      Bronze
      Joined: 24.02.2009 Posts: 708
      @ Riverdiver

      I agree with you when it comes to the deeper games in principal. When you get all in in 200bb pots chances are you think you have the best hand. Very rarily will the pot odds be laying you a correct call with a fd or sd or overs or whatever.

      On the other hand, for an sss or a mss or a cap player, this just isn't the case. These hands come up so frequently, in fact I would say that at least 1/3 of all of my all ins on the turn are not because I think I have the best hand, but that the pot odds are laying me such a great price I would be foolish to fold against his range. In these games its not out of the ordinary to see a villian shove second pair on the turn when you check oop after showing aggression preflop and on the flop.
    • Torg0th
      Torg0th
      Black
      Joined: 26.03.2010 Posts: 56
      You are not looking at it the right way. You won't always be the favorite when all-in (= your EV goes down) but that doesn't mean it's not the right play. Like in your example, by making the right plays and calling with the correct odds you "reduce the EV lost" as much as possible given the cards/board that were dealt.
    • Kyyberi
      Kyyberi
      Coach
      Coach
      Joined: 09.07.2010 Posts: 10,511
      The EV stat doesnt say anything about your decision. It only shows you how much your actual winnings differ from your "expected" winnings, which is your EV.

      In your example, on the turn you call 7,5bb shove to win a pot of 41bb with 34% equity.

      Now in reality, you can't split the pot or win a portion of it. Either you win 41bb or 0bb. In HEM there are $ and $EV diff. columns in reports. $ means the actual money you won. Which is now about 20bb or 0bb. But you put your money in with 34% equity to 41bb pot, which means your equity share is 14bb. So if you win, it shows that the difference is -6bb. If you lose, it shows that the difference is 14bb.

      I am amazed how we have trillions of threads about EV stat, and in 99% of them the whole EV line/stat/whatever is totally misunderstood.

      Yeah in some spots you actually make -EV decision as you play against a range. Sometimes you win the flips, sometimes you lose. EV adjusted winnings (the friggin' graph line or a stat) shows you just what your winnings were if you just split the pot with your equities when the money goes in. It's like dealing the river 34 times if there were 34 cards left.

      The bottom line is this: The EV line has variance too, as your winnings line. It doesn't tell the exact truth, but it's a good tool to estimate your playing, WITH ALL THE OTHER TOOLS.

      It's not broken, it's not useless. But if you don't like it, don't use it.

      Can we have a sticky thread about how the EV stat works, after few years this is getting boring. :)


      edit: Forgot to mention. The thing that causes most confusions is that people just don't understand that the EV line will even out in a long run, just like any other line. So when you say "I had to call turn with my set cos I have odds against his range, this time he had the higher set and I was drawing dead. So my EV line went down and it's wrong, it doesn't know his range" you are wrong. You just need a bigger sample, and it will even out. With big samplesize, you will hit the bottom of his range too and then you have higher equity than you should have against his range. So it evens out.
    • YohanN7
      YohanN7
      Bronze
      Joined: 15.06.2009 Posts: 4,086
      Wow, nice post!

      I've had the impression that everybody out there are sitting with a canned version of Phil Ivey (HM2/PT4) when they are playing, and refer to it when discussing their decisions in forum threads. I have envied them a bit because I couldn't figure out how that would be possible without seeing the opponents hole cards. [EDIT:At the very least we need to know the opponents ranges.] We do see our opponents hole cards! - At least in the context of that EV thingie, which should be called something else really. No more worries :) .

      /Johan = :f_confused:
    • Verre
      Verre
      Bronze
      Joined: 24.02.2009 Posts: 708
      Nice one Kyyberi. I love the rage in your post ;)

      All joking aside, your correct. It needs time to show a true ev. I guess my point is in high variance games the line takes MUCH longer to show a true or close to true reading than a lower variance game.
    • Kyyberi
      Kyyberi
      Coach
      Coach
      Joined: 09.07.2010 Posts: 10,511
      Yeah but still the EV line is better indicator than winnings imo. As it reduces some of the variance that winnings has.
    • Verre
      Verre
      Bronze
      Joined: 24.02.2009 Posts: 708
      But like I said a sample size of 100k games or so in a lower variance setting might get you a solid EV line, where in higher variance games like (PLO/HUNL) it would take many times more than that, and in games where all ins happen mostly in every pot on the flop or turn (CAP/Hyper Turbo SnGs) it might take upwards of a few million hands to show a EV line that has a standard deviation within maybe at best a few percent?

      Obviously I'm only doing basic math on this but I know neither the EV line or Profit line are to be trusted without a massive sample size.

      What my original post was intended for was to point out that if you have a good grasp of your skills in comparison to other players at your limit that your profit or ev line in general is somewhat missleading. The easiest way to figure out if your a winning player is from HH reviews, and confirming that you are making the correct decisions and more importantly making the correct decisions more often than your opponents.

      I for one have my EV line turned off and will leave it that way from now on. It only promotes tilt when you run under it and promotes a sense of "I'm getting lucky and am not really this good" when you run over it. I'd rather check my profit once at the end of the day after I've reviewed HH's and make sure that my br is intact for the current stakes and that through hh's I made the correct play as many times as I possibly could.
    • Kyyberi
      Kyyberi
      Coach
      Coach
      Joined: 09.07.2010 Posts: 10,511
      Originally posted by Verre
      What my original post was intended for was to point out that if you have a good grasp of your skills in comparison to other players at your limit that your profit or ev line in general is somewhat missleading. The easiest way to figure out if your a winning player is from HH reviews, and confirming that you are making the correct decisions and more importantly making the correct decisions more often than your opponents.
      True.

      But EV adjusted winnings is more accurate measurement of your game than normal winnings.

      The problem in the first post was that you think the EV stat tells you that you made -EV play. Which it can't tell you. So you have misunderstood it.