Roi adjusted PT4

    • Lacas84
      Lacas84
      Bronze
      Joined: 02.04.2011 Posts: 62
      Hi All,

      I usually dont focus on stats like my c net adjusted or ROI% adjusted but i feel like im running really bad so i took a look.
      My question is how accurate are these stats on Fifty50 sng-s and what they really mean.
      For example this month i played 704 $7 Fifty50 sng-s so far. My results are -35 dollars and my roi is -0.73%. But my c net adjusted shows $1493 winnings and my roi Adjusted is 30.30%. So does this mean that the way i played i should have 1493 dollars and my roi should be 30.30% but due to variance or luck i dont?

      Thx!
  • 11 replies
    • PokerTracker
      PokerTracker
      Gold
      Joined: 07.06.2011 Posts: 642
      Originally posted by Lacas84
      Hi All,

      I usually dont focus on stats like my c net adjusted or ROI% adjusted but i feel like im running really bad so i took a look.
      My question is how accurate are these stats on Fifty50 sng-s and what they really mean.
      For example this month i played 704 $7 Fifty50 sng-s so far. My results are -35 dollars and my roi is -0.73%. But my c net adjusted shows $1493 winnings and my roi Adjusted is 30.30%. So does this mean that the way i played i should have 1493 dollars and my roi should be 30.30% but due to variance or luck i dont?

      Thx!
      Net Adjusted Winnings is a measurement of expectation, it is not a measurement of luck therefore it cannot really tell you if you are running bad or not. The calculations are VERY accurate, even for "exotics" like Fifty50s but since they are measurements of expectation, not luck, the results can be confusing to someone who does not understand how the calculations are made, or how to read them.

      Its a little hard for us to help you interpret your results via the forum, I suggest you create a support ticket on our website and submit your results along with a copy of your graph so we can review for you if you would like further analysis. Before you do this, as a precaution, I would like you to perform a re-analysis of your Net Adjusted winnings to verify that the data is correct. This is done by entering the Tournament Results Editor, select all of your results (except the ones that you have manually entered) - this can be done by clicking CONTROL-A on a Windows OS machine, or CMD-A on the Mac OS. The click Auto Detect. Adjust your tournament detection settings, then run detection on the tournaments again. This will assure the table type, buyin, place finished, and tournament types are all detected, and then Net Adjusted Winnings is re-calculated for all-in events.
    • Lacas84
      Lacas84
      Bronze
      Joined: 02.04.2011 Posts: 62
      Hi,

      Thx for the reply. But i still dont know what to make of these results. It doesnt show variance or luck but expectation. Ok, but what that does really mean? I mean if there are 3 black and a white ball in a bag and I get one out without looking I expected it to be black I know theres a chance its gonna be white. Now lets say i do this 4 times and every time i pull out the white ball even though i expect it to be black every time but somehow i always get the withe one. Does that mean im really unlucky? After calculating you know you have bigger chance to pull out a black ball so thats what you expect but for whatever reason you pull the white one? So what i mean is if i know that theres a bigger chance that event A happens then event B then im gonna expect event A to happen but more often then it should event B happens. So is that variance or luck or expectaion. Because i expect event B to happen based on the knowledge i have. So what does net adjusted winnings show then? im really confused.
    • PokerTracker
      PokerTracker
      Gold
      Joined: 07.06.2011 Posts: 642
      Originally posted by Lacas84
      Hi,

      Thx for the reply. But i still don't know what to make of these results. It doesn't show variance or luck but expectation. Ok, but what that does really mean?
      The terms Luck, Expectation, and Variance are used in statistics applications, they are defined by statisticians and mathematicians, we must use these precise definitions accordingly. The best resource for you would be to review a statistics course on these topics if you want to learn more above and beyond what we discuss in this post, but I predict you will be able to follow along.

      A results-oriented measurement of luck could be perceived if you look at street by street equity. This is not really luck, but the brain can certainly use this to comprehend how lucky your results are even though this does not account for the luck included in card distribution or the opponent's decision tree (Collin Moshman has written about this in great detail in his great book
      The Math of Hold'em). Additionally PT4 can only calculate all-in equity adjusted winnings when all hands that had equity in the pot are known at showdown, this cannot be calculated when the pot is multiway and a player folded such as an all-in preflop, 2 calls, followed by an all-in and a fold on the flop. This results in a bias where only some hands are calculated unless you are playing purely headsup games.

      When you convert all-in equity into currency, then we are only discussing the use of expectation. Note that this is not EV (Expected Value), many players wrongly assume this is true. EV is measurement of future results vs an opponent's range (and it is not only during all-ins), where as all-in equity is a results oriented measurement of the share of the pot that we are expected to win - without factoring in skill, range, or fold equity. EV cannot be calculated by a tracker, only a human being can determine the range needed.

      All-In Equity is measured in Chips. In a cash game the value of chips is equal to the value of currency that you can win, but in a tournament a one to one correlation does not exist between the number of chips and the prize pool which means we must use ICM (Independent Chip Modeling) to determine the value of each chip, and then incorporate that value with the equity adjusted expectation of the pot to result in your Net Adjusted Winnings. ICM is a complicated topic outside of the scope of this post, it is only calculated at the final table assuming payout structure is know for the event, and it is recalculated as stack sizes change and players bust out.

      To reiterate, Net Adjusted Winnings is calculated by correlating the ICM value of each chip, the size of the pot, and the all-in equity adjusted winnings in chips to determine the Net Adjusted Winnings in a currency format - to verify that PT4 is accurate you must review each specific all-in situation at the final table rather than looking at the summed results. For example lets take a situation that is easy to calculate such as a $20 + $1 HUSNG where we know the prize pool is $40 and there are 3,000 chip at play so 40/3000 = .01333333, this means the value of each chip at play is $.0133 for our purposes. Next we consider the total size of the pot - lets say it is 2,202 chips (the villain already took 798 chips from you) this means the total prize pool value of this pot is $29.3599. Then we consider the all-in equity percentage preflop, lets use an example where you are expected to win 81.97% of the time. When you win you will get a pot valued at $29.29, when you lose then the hand cost you $21 (the price of the buyin + fee). The Net Adjusted Winnings stat is calculated by adding the expectation of the hand with the actual results, therefore the calculation is (29.3599 * .8197 - 21 = $3.06631003 which is rounded up to $3.07.

      And that is why Net Expectation results can be so confusing, even though they are mathematically correct. In that example your Net Expectation is only $3.07, even though you have 81.97% equity because your stack size is too small, and your opponents stack size is too large, to increase your expectation. As you can see, this is a VERY complicated topic!

      I mean if there are 3 black and a white ball in a bag and I get one out without looking I expected it to be black I know theres a chance its gonna be white. Now lets say i do this 4 times and everytime i pull out the white ball even though i expect it to be black every time but somehow i always get the white one. Does that mean im really unlucky?
      You are confusing probability with Net Expectation in your example, I feel its better to just use the mathematical analysis shown above than to try to use your simplified example.
    • Lacas84
      Lacas84
      Bronze
      Joined: 02.04.2011 Posts: 62
      Im starting to getting it. thx!
      Ok so i know now it is not what i thought it was. But then what can i read from this stat? what do i learn from it? how do i know if i played my hands good or bad? Or what stat in pt4 tells how well i played? So i would know if i should study more ( I know you always need to study more to stay on top :D )or im just on a downswing?

      Just one more quesiton!

      "All-In Equity is measured in Chips. In a cash game the value of chips is equal to the value of currency that you can win,"

      So lets say if me and another player goes allin pre flop and both of us have $50 then the pot is $100 (minus rake). I have AA hes got 88 roughly i got 80% and hes got 20% chance of winning. Is that 80% all-in equity for me? wich means i expect to win 80 dollars on that hand.? Or my all-in equity is $80? is that correct? And so if lets say after 100,000 hands in cash game my all-in equity shows more in $ then i won does that mean im unlucky?

      Sorry about all the questions!
    • PokerTracker
      PokerTracker
      Gold
      Joined: 07.06.2011 Posts: 642
      Originally posted by Lacas84
      Im starting to getting it. thx!
      Ok so i know now it is not what i thought it was. But then what can i read from this stat? what do i learn from it? how do i know if i played my hands good or bad? Or what stat in pt4 tells how well i played?
      There is no such stat, a tracking program only tracks results... PT4 does not judge your performance ;-) But you can use PT4 to find problem situations and then review your play such a playing in 3Bet pots, or from the Big Blind, to make sure your overall playing decisions are profitable - and if not, then use the replayer to find what your errors might be. This is what a coach would do for you, but to be honest you could certainly coach yourself using the same methods.

      So i would know if i should study more ( I know you always need to study more to stay on top :D )or im just on a downswing?
      Ignore if you are on a downswing or not, focus your attention on studying - thats how the great players become great!


      "All-In Equity is measured in Chips. In a cash game the value of chips is equal to the value of currency that you can win,"

      So lets say if me and another player goes allin pre flop and both of us have $50 then the pot is $100 (minus rake). I have AA hes got 88 roughly i got 80% and hes got 20% chance of winning. Is that 80% all-in equity for me? wich means i expect to win 80 dollars on that hand.?
      If you have 80% equity in the pot, and the pot is $100, then your Expectation of the pot is $80. Keep in mind you will win $50 80% of the time ($100 - $50 investment in the pot), and lose $50 20% of the time.. but it is displayed as $80 because that is your expectation if you play the same hand the same way and get the same call infinite times again and again -


      And so if lets say after 100,000 hands in cash game my all-in equity shows more in $ then i won does that mean im unlucky?
      No. This has nothing to do with luck, it is only expectation. We could only measure luck if you played the same hand 100,000 times in a row, vs the same opponent, with the same hand played the same way. Then we would take all of your actual winnings and compare it to your theoretical winnings of 80*100,000 = 800,0000. If your real world winnings were lower than $800,000, then we could honestly say you are unlucky. But since this scenario is impossible, it means that All-In Equity cannot possibly be a measurement of luck... it is just your expectation of the pot, nothing more than that.

      Sorry about all the questions!
      Don't worry, you're asking great questions that even many world-class players do not know the answers to! We wish more people would ask these questions, and more coaches would help explain the difference between expectation and luck; and how EV is a misnomer in tracking tools. Thank you so much for asking!
    • Lacas84
      Lacas84
      Bronze
      Joined: 02.04.2011 Posts: 62
      Thx again!

      So basically I can not make any use of these two stats? Cause they show irrelevant statistics i can not use to improve my game. Right?

      Oh btw! Wouldnt that be a measure of luck if we look at only all in hands and their all in equity. Then see if i win as many times as i should with the better hand? So we wouldnt care about the pot value or actual dollar amount as it is always different every hand but only if i won the hand or not. Then we could see that oh ok i had AA and went all in 10,000 times and i only won 65% of the time vs one player. or something like that.
      What do you think?
    • PokerTracker
      PokerTracker
      Gold
      Joined: 07.06.2011 Posts: 642
      Originally posted by Lacas84
      Thx again!

      So basically I can not make any use of these two stats? Cause they show irrelevant statistics i can not use to improve my game. Right?
      We disagree. These are VERY relevant and accurate stats, but they do not serve the purpose you thought they did. Expectation is very important even though it does not show luck as you thought it did!

      Oh btw! Wouldnt that be a measure of luck if we look at only all in hands and their all in equity. Then see if i win as many times as i should with the better hand? So we wouldnt care about the pot value or actual dollar amount as it is always different every hand but only if i won the hand or not. Then we could see that oh ok i had AA and went all in 10,000 times and i only won 65% of the time vs one player. or something like that.
      What do you think?
      That would be a biased measurement of luck, not a true measurement of luck, hence it is not statistically relevant. Remember we cannot factor in if your play, or your opponent's play was correct - nor can we factor in card distribution.
    • Lacas84
      Lacas84
      Bronze
      Joined: 02.04.2011 Posts: 62
      " We disagree. These are VERY relevant and accurate stats, but they do not serve the purpose you thought they did. Expectation is very important even though it does not show luck as you thought it did! "

      Ok so what can i learn from it? Tell me please! Lets see my example 704 7$ fifty50 sng-s and my results are -35 dollars and my roi is -0.73%. But my c net adjusted shows $1493 winnings and my roi Adjusted is 30.30%. So what is the conclusion from this. Knowing the adjusted stats results how this helps me?

      " That would be a biased measurement of luck, not a true measurement of luck, hence it is not statistically relevant. Remember we cannot factor in if your play, or your opponent's play was correct - nor can we factor in card distribution. "

      Why would it be biased? its just two hands compared together. Hand 1 has 20% chance Hand 2 has 80% chance then we see the results. And then we see how the results go after 100k hands. I dont see any reason to think its biased in any way.
      Hand distribution doesnt matter cause we wouldnt care how many times one player gets AA and how many time the other. thats not important. the only thing matters is that when we have those hands how they perform compared to chance (in this case 20-80%).
    • PokerTracker
      PokerTracker
      Gold
      Joined: 07.06.2011 Posts: 642
      Originally posted by Lacas84
      Ok so what can i learn from it? Tell me please! Lets see my example 704 7$ fifty50 sng-s and my results are -35 dollars and my roi is -0.73%. But my c net adjusted shows $1493 winnings and my roi Adjusted is 30.30%. So what is the conclusion from this. Knowing the adjusted stats results how this helps me?
      I have already answered this question in prior posts. You must first redetect all of the tournaments to assure your results are correct. Once that is done you can submit a support ticket with a copy of your graph and our team will be happy to review.

      PS: We suggest you look at each individual action, rather than the net sum total... this may help you to understand the functionality of expectation better.

      Why would it be biased? its just two hands compared together. Hand 1 has 20% chance Hand 2 has 80% chance then we see the results. And then we see how the results go after 100k hands. I dont see any reason to think its biased in any way.
      Hand distribution doesnt matter cause we wouldnt care how many times one player gets AA and how many time the other. thats not important. the only thing matters is that when we have those hands how they perform compared to chance (in this case 20-80%).
      Once again this is beyond the scope of this thread, therefore I suggest consulting with a statistician. I'm afraid we don't have the free resources available to devote extra time on this topic now that we have clarified the functionality of Net Adjusted Winnings, we are always happy to explain the reasons why our results are statistically accurate and the definition of our stats but if you want to explore general statistical concepts its probably best to discuss this with the general poker community or a specialist on this topic who can help communicate the concepts for you. Alternatively I recommend reading the excellent book from Colin Moshman
    • PokerTracker
      PokerTracker
      Gold
      Joined: 07.06.2011 Posts: 642
      Lacas84 - This thread inspired us to create a new guide to help explain how to calculate Net Adjusted Winnings, and how to understand it's purpose. This is something that we can certainly spend the time to do for the good of all of our users, where as it could be too time consuming for just a single post in a forum - so we are going to make this heavy investment thanks to your inspiration. Much appreciated!

      PS: I thought of a new way to explain what Net Adjusted Winnings is yesterday, its so simple that I am surprised we did not convey this earlier. The formula is Expected ICM - Realized ICM (Actual winnings). So for example lets say your ICM expectation is 80, and if you win your Realized ICM Win would be $140. If you lost the hand in a HUSNG and in turn lost the match, then your winnings would be $0 so 80-0 = 80.... this means the Net Adjusted Winnings graph point would be $80 ABOVE your actual winnings (if filtered to show only this hand it would show as $80). If instead the Hero won the hand, the Net Adjusted Winnings would be 80-140 = -60 BELOW your actual winnings (if filtered to show only this hand it would show as $-60).

      We will post the guide once it is ready in this thread.
    • Lacas84
      Lacas84
      Bronze
      Joined: 02.04.2011 Posts: 62
      Great News!
      Thx!