All-in Adjusted, In-accurate?

    • sc2zerker
      sc2zerker
      Bronze
      Joined: 21.12.2012 Posts: 162
      WARNING, I ROUNDED ALL THE MATH TO THE CLOSEST DIGIT. Answers will vary slightly if placed in PT4.

      Inc wall of text sorry. If someone can read through it and explain to me why I am wrong it would be greeeeeeeeat. Currently it makes no sense to me.

      From my understanding.

      adjusted all-in

      Amount of money won in the currency the hand was originally played when adjusting all-in pots. In those pots instead of attributing real results, this statistic attributes to the player the equity they had in the pot at the time of the all in. Pots where players call the all-in bet or raise but later fold (hence their cards are not known) are not adjusted.

      Formula: Amount Won Adjusted by All-In Equity

      ok good good, lets take a look at the all-in equity stat

      average all-in equity

      Average amount of equity that a player had at the time they went all-in in all hands where this information is known.

      Formula: (Preflop Equity + Flop Equity + Turn Equity + River Equity) / (Times Player Number of Was All-In Preflop + Number of Times Player Was All-In on the Flop + Number of Times Player Was All-In on the Turn + Number of Times Player Was All-In on the River)

      Ok cool i get it but... this is the problem.

      Lets do an easy example to start with.

      Hand details
      Seat 1: sc2zerker ($1,745.00 in chips)
      Seat 3: stefanschapendon ($1,450.00 in chips) DEALER
      Seat 5: Kukimacko ($1,190.00 in chips)
      Seat 6: misko85224 ($3,030.00 in chips)
      Seat 8: Marqui57 ($380.00 in chips)
      Seat 10: varadi123 ($1,205.00 in chips)
      Kukimacko: Post SB $10.00
      misko85224: Post BB $20.00
      *** HOLE CARDS ***
      Dealt to sc2zerker [DQ SQ]
      Marqui57: Fold
      varadi123: Fold
      sc2zerker: Raise (NF) Allin $1,745.00
      stefanschapendon: Allin $1,450.00
      Kukimacko: Fold
      misko85224: Fold
      *** FLOP *** [D2 D3 C4]
      *** TURN *** [CA]
      *** RIVER *** [C8]
      *** SUMMARY ***
      Total pot $3,225.00 Rake $0.00
      sc2zerker: Shows [DQ SQ]
      stefanschapendon: Shows [SK DA]
      sc2zerker: wins $295.00
      stefanschapendon: wins $2,930.00


      Hero all-ins with QQ and gets called by AK.

      Hero is given 56% equity, hero all-ins on pre-flop. hero is reward with 200 all-in adjusted chips in PT4.

      Cool works perfect? kind've...

      Look at the following example

      Hand detail
      Seat 1: lllukas18 ($1,500.00 in chips) DEALER
      Seat 3: Amrons ($1,500.00 in chips)
      Seat 5: sc2zerker ($1,500.00 in chips)
      Seat 6: stanislas777 ($1,500.00 in chips)
      Seat 8: tombolio33 ($1,500.00 in chips)
      Seat 10: ozzymick ($1,500.00 in chips)
      Amrons: Post SB $10.00
      sc2zerker: Post BB $20.00
      *** HOLE CARDS ***
      Dealt to sc2zerker [DA SK]
      stanislas777: Call $20.00
      tombolio33: Call $20.00
      ozzymick: Fold
      lllukas18: Call $20.00
      Amrons: Call $10.00
      sc2zerker: Raise (NF) $1,480.00
      stanislas777: Fold
      tombolio33: Fold
      lllukas18: Fold
      Amrons: Call $1,460.00
      *** FLOP *** [H9 D5 H3]
      Amrons: Allin $20.00
      sc2zerker: Allin $20.00
      *** TURN *** [S9]
      *** RIVER *** [S2]
      *** SUMMARY ***
      Total pot $3,060.00 Rake $0.00
      Amrons: Shows [S7 C7]
      sc2zerker: Mucks [DA SK]
      Amrons: wins $3,060.00


      Hero all-ins with AdKs vs villian with 7s7c (hero has 45% equity to villian's 55% equity)

      Hero has 1500chips, Hero bets 1480 chips. Villian has 1500 chips, villian calls 1480 chips.

      flop happens. board shows 9h5d3h. Equity has changed. Now it's (hero has 25% to villian's 75%)

      Hero places his 20 chips that he missed. Villian calls (they are obviously pot committed with 98%+ of their chips in the pot)

      PT4 does some magic

      The magic
      1500chips to gain or lose

      25% equity vs 75% equity

      +375(25%)
      -1125(75%)
      combine the two sums

      -750 chips ( adjusted)



      poof hero is reward with a -750 all-in adjusted chips. Is this accurate? Not really when you think about it, when someone has 98%+ of their chips in the pot there is NO WAY they are going think it's profitable to fold on the flop. So why would PT4 take the equity on the flop instead of the pre-flop? So hero should only have a -150 all-in adjusted.

      I assume it's because of user error/ bad ai / no option for it. You should most likely make a mental note to have 100% of your chips in the pot so PT4 can be accurate but guess what... Players like volume. Sometimes they are playing a lot of tables and clicking fast. Sometimes a mis-click happens. Sometimes their slider bar doesn't reach the end.

      I think all-in adjusted would work much better if it defined it's equity %s when the players are pot committed instead of the last betting street. What do you guys think?

      Maybe the system doesn't work this way. Maybe im 100% wrong but that's why i think the all-in adjusted is in-accurate. I would love to hear any feedback. I'm fairly new to poker and it's something i might have 100% wrong. Im just trying my best to get accurate results while playing a lot of tables.

      Thanks for reading!
  • 3 replies
    • PokerTracker
      PokerTracker
      Gold
      Joined: 07.06.2011 Posts: 643
      First - Please be aware that you must include ties in your all-in equity adjusted calculations. Without the inclusion of possible ties, your results will never be as accurate as PokerTracker 4. Now that this is clarified, we can continue with your example.

      Hero all-ins with AdKs vs villain with 7s7c (hero has 45% equity to villain's 55% equity)

      Hero has 1500chips, Hero bets 1480 chips. Villain has 1500 chips, villain calls 1480 chips.

      flop happens. board shows 9h5d3h. Equity has changed. Now it's (hero has 25% to villain's 75%)

      Hero places his 20 chips that he missed. Villain calls (they are obviously pot committed with 98%+ of their chips in the pot)


      All-In equity Adjusted Winnings can only be calculated when there is no further betting action remaining and there are streets remaining to be seen. In this situation you said you are all-in preflop, but that is not true - you are actually all-in on the flop - leaving us two streets to be seen, not three streets.

      [QUOTE]poof hero is reward with a -750 all-in adjusted chips. Is this accurate?[/quote]The correct calculation is
      AK = 25.56% equity on the flop
      77 = 74.44% equity on the flop

      Pot size = 3,060.00 (after rake)

      AK All-In Equity Adjusted = $782.14
      77 All-In Equity Adjusted = $2277.86

      (there are no tie possibilities on the flop, but there were tie possibilities preflop)

      We suggest that you do not look at the graph, look at the hand in the reports to determine the actual All-In Equity Winnings result. The graph is an adjusted stat, relative to the prior results so if the prior result was zero, the graph will read at +782.14, but if the prior result was -2000, the graph will read as -1217.86.

      So why would PT4 take the equity on the flop instead of the pre-flop? So hero should only have a -150 all-in adjusted.
      Because that is when you went all in. PokerTracker CANNOT READ MINDS, it is mathematically correct, it does not know your intention - and most importantly this is a measurement of EXPECTATION, not LUCK... mathematically the expectation is correct. If you wish to determine the all-in equity winnings preflop you can manually perform this task.


      [quote]I assume it's because of user error/ bad ai / no option for it. You should most likely make a mental note to have 100% of your chips in the pot so PT4 can be accurate but guess what... Players like volume. Sometimes they are playing a lot of tables and clicking fast. Sometimes a mis-click happens. Sometimes their slider bar doesn't reach the end.
      [/QUOTE]

      You could consider this user error or player error, but I don't... I see this as a player choosing a non-standard line, which results in a different calculation of expectation. Sometimes these non-standard lines are the best choice - but the result of the best choice is a change in Expectation. As a reminder, All-In Equity Adjusted Winnings is not a measurement of running hot or cold, it is not a measurement of luck either... it is just a measurement of expectation.

      I think all-in adjusted would work much better if it defined it's equity %s when the players are pot committed instead of the last betting street. What do you guys think?


      That would be mathematically incorrect. The concepts of probability, equity, and expectation are defined mathematically, these concepts have nothing to do with poker - they are long established rules of mathematics and probability theory that are applied to poker.

      Maybe the system doesn't work this way. Maybe im 100% wrong but that's why i think the all-in adjusted is in-accurate. I would love to hear any feedback. I'm fairly new to poker and it's something i might have 100% wrong. Im just trying my best to get accurate results while playing a lot of tables.


      I think this paragraph explains the crux of your issue... I think you misunderstood what All-In Equity Adjusted Winnings is used for. I suggest you review these three threads which we recently discussed similar topics within the PokerStrategy community, this may help. On the bright side, you are not alone... we find most players misunderstand what Expectation is, we often see it confused it for Expected Value or incorrectly used as a measurement of luck. Luck cannot be measured in these situations because skill is a factor that cannot be measured, and card distribution cannot be accounted for. We can however measure the luck of draws - see LeakTracker > Overall Luck Bell Curve for details.

      Street Adjusted EV for PT5?

      Roi adjusted PT4
      Pokertracker Ev
    • sc2zerker
      sc2zerker
      Bronze
      Joined: 21.12.2012 Posts: 162
      Firstly, Thanks for the quick reply!

      Before I wrote the post I was not misunderstanding that all-in adjusted was something that does not determine luck or skill. It is something that simply shows the exception of the results given your all-in equity % in certain situations.

      All-In equity Adjusted Winnings can only be calculated when there is no further betting action remaining and there are streets remaining to be seen. In this situation you said you are all-in preflop, but that is not true - you are actually all-in on the flop - leaving us two streets to be seen, not three streets.

      I understand from a technical point of view you are correct. The player didn't place 100% of their chips into the pot. therefore player does not have "all" his chips into the pot. I'm not disagreeing that. I'm saying from a logical point of view, folding is not an option after seeing the flop. So the flop is irrelevant and post-flop equities are irrelevant in this example.

      The true equity in this scenario is the pre-flop equity. Not the equity given to each player on the flop.

      I'm not saying it's program error, the program is working 100% the way it's programmed to work. It's pot commitment and User error that the program cannot account for which as a result produces in-accurate results.

      Because that is when you went all in. PokerTracker CANNOT READ MINDS, it is mathematically correct, it does not know your intention - and most importantly this is a measurement of EXPECTATION, not LUCK... mathematically the expectation is correct. If you wish to determine the all-in equity winnings preflop you can manually perform this task.
      I think reading minds would be an asset for sure. it's not very realistic though. Well maybe in the future? ;) Maybe an alternative to account for pot commitment though? Or at least user error? If not then the adjusted all-in chips will always be slightly in-accurate mathematically because it's not pulling the equity %s in the correct situations.

      I understand the program is trying to be as accurate as possible. If you can explain why in the example above taking 25% vs 75% equity is accurate compared to 45% vs 55% equity. Not simply as the program takes the last betting street even if it's 1chip. Look how poorly you can exploit this way of calculating mathematically.

      Step 1
      Pre-flop

      Hero has 25,000 chips with A :spade: A :heart:
      Villian has 25,000 chips with 7 :spade: 2 :heart:

      Hero bets 24,999 chips

      Villian Calls.

      (equity was 88.97% vs 11.03%)

      Step 2

      Flop

      7 :diamond: 7 :heart: 7 :club:

      (equity is now 0.1% vs 99.9% )

      Hero bets 1 chip moves and moves all-in
      Villian calls all-in with 1 chip

      Pokertracker assigns expected results

      +25chips
      -24975chips
      -----
      -24950chips

      on average hero is losing 24950chps ( this is true because of the flop )
      Hero is given an -24950 chip adjusted in his report

      So now you have a nice big -25k chip adjusted to your report which is not even close to accurate.

      What is PT4 took the equity that was given to each player pre-flop in the following example? Wouldn't that be much more reasonable? Wouldn't that be much more accurate?

      Just my thoughts on matter, I don't mean to offend anyone. I think the program is great and im really enjoying it. I also want to get the most out of it. I think to do that I should use the all-in adjusted stat which I might be mis understanding!

      Thanks in-advance.
      Much appreciation and sorry for being a huge pain in the side about it.
    • PokerTracker
      PokerTracker
      Gold
      Joined: 07.06.2011 Posts: 643
      Originally posted by sc2zerker
      Firstly, Thanks for the quick reply!

      Before I wrote the post I was not misunderstanding that all-in adjusted was something that does not determine luck or skill.It is something that simply shows the exception of the results given your all-in equity % in certain situations.
      I think you mean Expectation, but otherwise correct.

      I understand from a technical point of view you are correct. The player didn't place 100% of their chips into the pot. therefore player does not have "all" his chips into the pot. I'm not disagreeing that. I'm saying from a logical point of view, folding is not an option after seeing the flop. So the flop is irrelevant and post-flop equities are irrelevant in this example.
      I'm afraid we are right both logically and technically. We cannot bend the rules of mathematics in this situation. The flop is HIGHLY relevant because this is how you chose to play the hand.... we do not see this an error, we see this as a choice.

      The true equity in this scenario is the pre-flop equity. Not the equity given to each player on the flop.
      We agree the most important equity situation in this hand is preflop, but it is not the "true" equity. The point of the all-in is the only point where we can correctly calculate the all-in equity.

      Maybe an alternative to account for pot commitment though?
      Pot commitment is subjective, we do not intend to measure this.

      If not then the adjusted all-in chips will always be slightly in-accurate mathematically because it's not pulling the equity %s in the correct situations.
      Once again, it is 100% accurate mathematically and it is pulling the equity in the correct situation, which is at the point of the all-in. If you wish to measure equity and adjust for pot value on earlier streets, you are free to do so manually - but this is not something that PokerTracker is designed to do - nor is it something that is requested by your fellow players (it would be far too confusing to track adjusted equity on all streets)

      So now you have a nice big -25k chip adjusted to your report which is not even close to accurate.
      Our method is tried and true, it is accepted universally by mathematicians as well as players; therefore rather than argue your point perhaps you would be better served if you consulted with a poker coach that specialises in understanding equity adjusted winnings? Or perhaps posting your theory with your peers in a strategy forum instead of a software support forum to ask for guidance? Unfortunately your misunderstanding is outside of the scope of our support goals, one of these two options will probably help to point you in the right direction better.

      Just my thoughts on matter, I don't mean to offend anyone. I think the program is great and im really enjoying it. I also want to get the most out of it. I think to do that I should use the all-in adjusted stat which I might be mis understanding!

      Thanks in-advance.
      Much appreciation and sorry for being a huge pain in the side about it.
      Don't worry, we are not offended at all, we are just trying to help to the best of our ability within the scope of our support. We are actually glad you're asking these questions, we would rather see you educated and empowered than misunderstand a key concept such as all-in equity adjusted winnings.

      We often advise players to ignore their All-In Equity Adjusted Winnings results, because the concept of expectation is so commonly misunderstood. You would be better served focusing on your tough situations such as filtering to review your play in 3bet pots, or play from the Small Blind and Big Blind - these are spots where you can make more money or lose less, and are far more valuable to review than All-In Equity.

      Good luck, we hope you come to understand this concept better, and learn why the flop is not a viable inflection point to consider if there is further action remaining to take place.