PokerStars homepage
  • If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.


No announcement yet.

Understanding $ (EV adjusted)

  • Filter
  • Time
  • Show
Clear All
new posts

  • Understanding $ (EV adjusted)

    Hello, I've been having a really tough month playing 45 man sit and go's. I've played just over 100 of them this month and only managed an ITM of 10.8%. This has had me doing a lot of stat checking and head scratching trying to work out where it's all gone wrong. Looking at my stats I don't see much differents in the way I've been playing lately compared to the way I was playing when winning. Not that I ever do very well, but I don't usually do so badly. After whining in the chat box about losing badly and being asked 'what's gone wrong' (by Sandtrap) I said I didn't know, but I have been racing very badly. This instantly sounded like an excuse and I regretted saying it. But it did get me looking more closely at my races tab in HM2. So, at last to the point. Looking at the races page it has a 'net won' total and an '$ (EV adjusted)' total. I've also added 'bb/100' and 'EV bb/100' in an effort to understand what I'm looking at. My 'Net won' shows -$968.00 (71.3 bb/100) and an '$ (EV adjusted)' of $77,487.43 ( 311 EV bb/100). Looking at other months the numbers seem to be a lot more balanced. Does this show that I have, in fact, been racing very badly? And explain what's gone wrong? Or should I be looking else where for my problems? There is also a $EV number on the 'Tournament Results' page, which I really can't get my head around Any help with understanding these numbers will be much appreciated umbup:

  • #2
    Hi Steveisnot,

    Have to start by saying I have never used this feature for torniments.

    That said if those where cash games and Net C won was negative and EV adjusted was positive then that would mean you are getting your chips in good and taking a bad beat.

    I would have to guess for tournys it would be simular and i hope someone comes and comfirms or says na so i can look at it for my 90 man games

    Grade b
    I am always ready to learn although I do not always like being taught. ~Winston Churchill

    13 Time Bracelet Winner


    • #3
      Think it's got to be the same. Although I think looking at chip count is no good as the value of the chips changes as the big binds go up. Does that make sense?

      Check back over many months and have to conclude that I've just been very unlucky in these this month. Can't find a month with such a high disparity in these numbers. If that's what these number mean that is.

      Or can you play in a way that invites bad beats? To many coin flips? Shoving on the flop with a low two pair?

      Can't help thinking I've started doing something wrong, but I cant see it my stats. VPIP, PFR, 3BET and Agg% are all constant.

      Maybe that's it! Not mixing it up and the regs have got me pegged

      Still love to know what the $EV number means on the tournament results page. What's it working out?

      Once again any help much appreciated


      • #4
        Don't take too much notice of the dollar amount, because - as you say - the relative size of a chip stack alters according to the blind level in tourneys. (If you finish in second place, your bust out hand might be a loss of 10,000 or more chips, which is much larger than the stack you began the tourney with, but it doesn't mean you were "extra unlucky" on that one hand).
        But bb/100 and EVbb/100 are kind of useful.
        As you've no doubt noticed, there is a lot of variance in SnGs. Often you might lose your buy-in in 20 hands or less (-1 buy-in in 20 hands is pretty brutal), while winning a nice prize of several buyins might only take 150 hands.

        If your bb/100 is way below your EVbb/100, then you have indeed been running bad; losing your stack when you got your money in good, and rarely sucking out when you got it in bad. Your EVbb/100 is a "truer" reflection of your true long term winrate, because it shows how much you'd be winning if you won your "fair share" in flips.
        It could take many thousands of tourneys before your bb/100 is close to your EVbb/100. I've run really bad in cash games. Even after 100,000 hands, my bb/100 was only 60% of my EVbb/100. I'm kind of resigned to always being below EV for my whole "career". There's nothing I can change about my luck. About half of us are naturally going to be unluckier than the other half of the population. We'll just have to keep whining about the bad beats, but trying not to get too obsessed with results-oriented thinking!
        Bracelet Winner


        • #5
          The correct answer would depend on the tracking software you use, because HEM, HEM2, PT3/PT4 are calculating the c$-EV in a different way.

          The impact of luck isn´t that big as most think when it comes to EV, because it´s simply a to smallish part of the whole variance thing to see it as a manifestation of a luck indicator. If I´ve read correct through your post you´re using HEM2 and you´re playing 45´ers. The c-EV in HEM2 has a flaw in concept which makes it almost useless for MTT/SNG. ALL REAL winning players must run over ev by definition, you can see this in extreme when it come to hypers. So your post either states you´ve been unlucky on a short sample or you´re not a winning player.

          Whatever the reason is, it isn´t changing anything to know about this graph. To determine your true winrate you need a solid sample and on a solid sample the graph isn´t important anymore. So best thing would be to forget about it and focus on your game and how to improve. Improving your game means always improving your winrate and it´s contrary to EV something you can control yourself.


          • #6
            Originally posted by DivorcedDuck View Post
            The impact of luck isn´t that big as most think when it comes to EV, because it´s simply a to smallish part of the whole variance thing to see it as a manifestation of a luck indicator.
            In cash games, all in EV ("luck in called shove situations") doesn't have a huge impact on your overall winrate, although it's not uncommon to have winnings of 65% of your EV over 100,000 hands.
            In SnGs, however, there is a lot more luck involved in making the money. You can play solid poker, and then get it in with AA vs KK on the bubble. Win the hand, and you'll probably make a decent cash. Lose the 80/20 race, and you don't win a single cent. In the short term, variance is brutal in multi-table tournaments, even those with just 45 runners, because only 7 make the money.
            Originally posted by DivorcedDuck View Post
            ALL REAL winning players must run over ev by definition, you can see this in extreme when it come to hypers
            It depends which actual stat you're referring to, but if you mean EV for ALL the races you have, then it's perfectly possible for a winning player to run below EV, just as it's possible to be a losing player that is above EV in their races. This is because, as explained above, races near the bubble or when you're safely in the money have a bigger affect on the size of cash you make. Winning players may be slightly luckier than losing players, but it's not at all correct to say that all winners are winning more than expected. In fact, it's pretty silly to say so, imho.
            Bracelet Winner


            • #7
              Originally posted by ArtySmokesPS View Post
              This is because, as explained above, races near the bubble or when you're safely in the money have a bigger affect on the size of cash you make.

              before making such a statement you should have a look how HEM2 is calculating the EV and how the graph is made. the graph isn´t based on the EV itself but on a model nearing the EV and the EV itself is calcualated streetwise(!) which is unique as HEM1/PT3/PT4 don´t do that. the streetwise calculation has an effect of distortion in multiway spots, which are very common in late game, because it prefers the equity improvement on later streets and winning players are more frequently the ones with bigger stacks you know? for the same reason that effect is way bigger in hypers as in hypers there are obviously way more allins and therefore more multiway multi street allins. a real winning player can´t run under EV that way on a huge enough sample which makes save he´s a winner. before claiming someone is silly when pointing that out you should get some knowledge on that first...
              Last edited by DivorcedDuck; Wed Mar 20, 2013, 12:22 AM.


              • #8
                Duck, I'm still not sure which particular stat you are referring to.

                EVbb/100 is your expected winrate in big blinds per 100 hands IF you won your "equity share" of chips in every pot you played where there was a called shove. This stat (in HEM2) is calculated whenever you're all in, or calling an all in, whether the pot is heads up or multiway, in the money, or 1000 places from the bubble. It's the same stat that is used in cash games, and it's a reasonably good estimate of your "true" winrate. It would indeed be difficult to be a winner if your EVbb/100 was negative, but there are plenty of players (probably around 50%) that have an EVbb/100 that is lower than their bb/100.

                The tournament $EV figure, by contrast, seems to apply when you are close to the bubble or in the money and have a big flip. It refers to actual currency, not chips. I believe it's based on ICM, but can't be sure. (e.g. If you were heads up in a winner takes all tourney with equal stack sizes and took a 50/50 flip that would mean you won $1 if you won the flip, and nothing if you busted, then your $EV would be 50c, and your actual result would either be 50c above $EV or 50c below $EV).

                The long term total $EV figure would tend to be close to your actual winnings, especially in 45-man 25c games, because the difference between 7th and 6th, for example, is only a couple of cents.
                And yes, a long term winner would therefore have a positive total $EV, because they have made a profit in those games. Whether the $EV number is above or below the actual $won is down to pure luck. Some winners have a $EV that is above their actual winnings. Some winners have a $EV that is below their winnings.

                It seemed from your earlier comments that you believe all winners are winning more money than expected. If that's not the case, then I apologise for the misunderstanding.
                Bracelet Winner


                • #9
                  [Duplicate post removed]
                  Last edited by ArtySmokesPS; Wed Mar 20, 2013, 01:14 AM. Reason: dupe
                  Bracelet Winner


                  • #10
                    we´re talking about the $-adjusted ev stat for sng/mtt. it´s a column in tournaments and tournaments result tab in HEM2 just on the right of the Net Winnings and it´s not calculated the way you described it and i´m not talking about luck or real winnings. i´m saying that a winning player is "running above EV" in HEM2 by definition of the stat on the long run which is common knowledge for the reason i told you. it´s not calculated to the point where the allin was, it´s calculated street by street in HEM2.

                    i give you an example.

                    3 handed
                    blinds 25/50

                    UTG t500
                    MP2 t1500
                    BU t1500

                    UTG minraises pre, MP2 and BU call.
                    UTG goes allin on flop, MP2 and BU call.
                    MP2 bets 1/2 pot turn, BU shoves adn MP2 calls.

                    c$-EV for UTG is calculated on flop but only for him.
                    c$-EV for MP2 and BU are calculatd on turn for the whole pot.

                    the dilemma is when you calculate it street by street, you have to make a decision wether to calculate for the whole pot or partially when it´s multiway. both is possible and both has valid points. in HEM2 it´s done for the whole pot which favors the big stacks. as already told you and you for sure agree to that, a winning player will be frequently more in the position of the big stack and therefore it comes to the mentioned distortion of the graph. it has nothing to do with luck. it´s logic and math.

                    in addition for performance reasons the graph isn´t showing the total ev adjusted winnings but only a nearing value.

                    no matter how often you repeat the stuff you wrote it´s not becoming true for HEM2. PT3/PT4 do it the way you described but not HEM2 and the question of OP was about HEM2. neither you or me knows about the samplesize of OP. so the correct answer is OP is either unlucky on a short sample or a losing player on a huge sample and mostlikely something in betweeen but we don´t know. however, it´s totally unimportant no matter what you say, because OP has no influence on variance.

                    a last thing to add...

                    The long term total $EV figure would tend to be close to your actual winnings, especially in 45-man 25c games, because the difference between 7th and 6th, for example, is only a couple of cents.
                    based on the assumption we have a perfect ev-calculation, no matter if heads up or multiway and it´s calculating correct against icm, which has it´s limitations either, your assumption is totally opposite to what i would normally expect in a fair game. law of big numbers...


                    basically it says for poker that the percentual difference will decrease as the numbers getting high but the absolut difference, which we know as money, will increase. as you referred to the absolute number it´s in realiyt the other way round from what you described and this has nothing to do with luck either. variance has no memory and knows nothing about buyins. variance in poker is equal to the cards which are dealed and no more.



                    X Cookies Information

                    We have placed cookies on your computer to improve your experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.