All in with AK (database analysis)

magicpigmagicpig Red Chipper Posts: 94 ✭✭
Has anyone done a database analysis of what happens when a player with a full stack is all in preflop with AK in various situations?

For example:

I read one player claim he was
loser in all situations that did not involve the button vs the blinds or blinds vs the button.

Comments

  • mwolovetz1mwolovetz1 Red Chipper Posts: 15 ✭✭
    I am just running some math on that, it should be positive EV I hope!!
    Use the Advance EV equation, at the micos you get all kinds of idiots calling with anything sometimes!!
  • mwolovetz1mwolovetz1 Red Chipper Posts: 15 ✭✭
    You don't need any data base to figure this out, just not clear on math though.
  • magicpigmagicpig Red Chipper Posts: 94 ✭✭
    edited February 2
    mwolovetz1 wrote: »
    You don't need any data base to figure this out, just not clear on math though.

    I am going to write something that you may find helpful:

    Understanding population tendencies is one highly valuable tool pros use.

    Even in a solved game like rock, paper, sissors it would be better to deviate from GTO if we reasonably thought our opponent had a bias.


  • mwolovetz1mwolovetz1 Red Chipper Posts: 15 ✭✭
    sounds like work lol.
  • persuadeopersuadeo Red Chipper, Table Captain Posts: 4,002 ✭✭✭✭✭
    I read one player claim he was
    loser in all situations that did not involve the button vs the blinds or blinds vs the button.

    While I too would like to see a current mass database of this, it would only make sense that ako be a slight loser in this scenario vs tight all in ranges. So his comment should be considered normal, not unexpected.
  • mwolovetz1mwolovetz1 Red Chipper Posts: 15 ✭✭
    I don't know how you can say ALL IN AK is a loser with a tight player in the blinds or any other position, or AK is a loser in certain positions.
    The math says other wise, AK against tight(worst case) scenario QQ+, AK is a Positive EV play all in. No need for massive data base.
    EV= F($Pot)+C(%Wx$W)-C(%Lx$L)= YOUR EV VALUE.

  • SplitSuitSplitSuit RCP Coach Posts: 4,010 -
    edited February 6
    mwolovetz1 wrote: »
    sounds like work lol.

    Yes. If you want to succeed long term in poker at any meaningful stake, work is a requirement. Learn to love it =)

    As for the OP, you have to set this filter correctly AND understand a few things:
    1. When you go AIPF with AK (take a spot where they open, you 3bet, they 4bet, you 5bet rip), there are two outcomes: they fold and we pick up the medium pot, or they continue and contend for the large pot.
    2. 4-5-6 bet ranges can get SUPER narrow, and almost always AK will be the equity dog when the AIPF gets called
    3. Which means that analyzing AIPF spots with AK will result in some small-medium wins (where V folds to your shove), quite a few big losses (when V calls and holds their equity), and some big wins (when V calls and we bink our equity)

    The filtering needs to account for shorter stacks (since AK AIPF is WAY different at 40bb vs. 100bb), deeper stacks (at 160bb+ ranges can contort quickly in AIPF spots), etc. So after filtering, you are usually down to a small sample size (and ime, one that has a negative bb/100). Plus...variance.

    FWIW, even though that's negative it doesn't automatically imply that going AIPF with AK is trash. You need to compare the EVs of mashing it preflop, calling their 4bet and hitting vs. missing, just calling the 2bet in the first place, etc. It's not a clear black-n-white filter/output - so I advise against making huge sweeping strategic adjustments as a result. Rather, use the output as a jumping off point for other exploration. Take your time, review each hand that did end up in this filter to ensure your reads/math looked good, and then start exploring other lines =)


    *PS: @persuadeo essentially said all of that, in far less words lol
  • mwolovetz1mwolovetz1 Red Chipper Posts: 15 ✭✭
    "The filtering needs to account for shorter stacks (since AK AIPF is WAY different at 40bb vs. 100bb), deeper stacks (at 160bb+ ranges can contort quickly in AIPF spots), etc. So after filtering, you are usually down to a small sample size (and ime, one that has a negative bb/100). Plus...variance."

    What if player is playing with a short stack in a cash game and buys in for the minimum 40BB.
    I s this going to alter the math/or results, its either profitable at 40bb and 160bb or its not. Are you inferring with deep stacks that most if not all will push only with AA?.

    At the micro level many many with call a all in with significantly weaker holdings that makes the AK push more profitable.
    In my mind you could dance around with the raise, 3bet, all in bet, but I am talking about folding or pushing all in period.
    You can draw inferences with Villain calling, rearising, 3betting you, and I get that, which gives you(The person with AK) the opportunity to get away based on your thoughts with the rearises etc.

    I would like to hear, with maybe a example how the profit differs from a 40bb to 160bb stack, as I don't get that.
    I am a newbie, just trying to understand.
    I am assuming a person gets AK and his only decision is all in or fold?

    Thanks for great dialog.
    Also not sure on some of the short hand and abbreviations for instances, what is OP(out of position?) and FWIW? not sure on that one?

    M.


  • magicpigmagicpig Red Chipper Posts: 94 ✭✭
    edited February 7
    mwolovetz1 wrote: »
    I don't know how you can say ALL IN AK is a loser with a tight player in the blinds or any other position, or AK is a loser in certain positions.
    The math says other wise, AK against tight(worst case) scenario QQ+, AK is a Positive EV play all in. No need for massive data base.
    EV= F($Pot)+C(%Wx$W)-C(%Lx$L)= YOUR EV VALUE.

    This is incorrect.

    Against an unknown, getting all in with AK in various positions vs other positions with multiple raise sizes is absolutely a losing play in some pools, especially where more than six people are playing. There werected old threads on other forums where several players offered up large samples years ago bit I lost track of the information.

    * I noticed a series of excellent Red Chip videos on player pool analysis so....
  • magicpigmagicpig Red Chipper Posts: 94 ✭✭
    Adam “w34z3l” Jones has done excellent database analysis and produced excellent videos.

    He also coauthored a recent book on AK.

    Maybe he will have something to say on this topic (?)
  • RoblivionRoblivion WisconsinRed Chipper Posts: 288 ✭✭✭
    mwolovetz1 wrote: »
    Thanks for great dialog.
    Also not sure on some of the short hand and abbreviations for instances, what is OP(out of position?) and FWIW? not sure on that one?

    OP = Original Post (the first post in the thread) or Original Poster (you).
    FWIW = For What It's Worth.
  • mwolovetz1mwolovetz1 Red Chipper Posts: 15 ✭✭
    I am starting to get lost here, with the math and direction of this question, can someone look at my math and tell me where I am going wrong, and if/or math is correct and or my assumptions??

    Prerequisites are Hero has AK and short stacked with 40bb, Villain has us covered.
    For this example Villain call only with QQ+,AK (worst case scenario??) or approx. 3 percent of time for this example.
    Just 2 players in pot Hero & Villain.
    For this example at Mico level which equals 40bb x 0.02= $0.80 or 80 cents.
    Villain 3bb open raise(0.06 cents).
    Hero 3bet to 9bb(.18 cents).
    Villain 4bets to 27bb(.27 cents).
    Hero shoves remaining 0.62 cents.
    Equity for AK via Flopzilla=.39% against QQ+,AK.


    Here is my math, corrections very welcome if parts are wrong!?
    Not sure if I am adjusting for $w and $L correctly?
    F($pot)+C(%Wx$W)-C(%Lx$L)= EV.
    Plugging in numbers as follows.

    .97(.75)+.03(.39x1.37)-.03(.61x.98)
    0.7275 + 0.016029 - 0.017934 = 0.725595+ EV?

    Thanks

    M.









  • driller1driller1 Red Chipper Posts: 25 ✭✭
    So the classic flips are AK vs 22-QQ. Even suited, AK is a slight dog against that range. Against non pair hands AK is a significant favorite against everything but another AK.

    If I am short stacked, I'm going most of the time, but if stacks are deep, I'll wait.
  • Brian BBrian B Red Chipper Posts: 18 ✭✭
    I believe the F in your equation is incorrect. I don't think it is 97%. Yes, he might only call your shove with a 3% range (JJ+, AK), but i think F should be the percentage that he would 4bet then fold to a shove. So lets say he 4 bets (JJ+, AK, A2-A5) or 4.2%. If he only calls your shove with the JJ+, AK, he is continuing roughly 75% of the time vs your shove and F in the equation would be 25%.
  • TheGameKatTheGameKat Posts: 2,130 -
    magicpig wrote: »
    Adam “w34z3l” Jones has done excellent database analysis and produced excellent videos.

    He also coauthored a recent book on AK.

    Maybe he will have something to say on this topic (?)

    @w34z3l
    Moderation In Moderation
  • magicpigmagicpig Red Chipper Posts: 94 ✭✭
    mwolovetz1 wrote: »
    I don't know how you can say ALL IN AK is a loser with a tight player in the blinds or any other position, or AK is a loser in certain positions.
    The math says other wise, AK against tight(worst case) scenario QQ+, AK is a Positive EV play all in. No need for massive data base.
    EV= F($Pot)+C(%Wx$W)-C(%Lx$L)= YOUR EV VALUE.

    I addressed this comment earlier and I would like to add that many very tight players, especially in casinos or in some online Zone-type pools, often flat AK and even QQ most of the time in SRP's and 3-Bet Pots so it is not possible for them to show up with anything but AA KK when the money goes in. These players would scoff at AA, KK, QQ and AK being a "tight" range.

  • mwolovetz1mwolovetz1 Red Chipper Posts: 15 ✭✭
    Thank you Brian B, you shed some light on question, thanks!!

Leave a Comment

BoldItalicStrikethroughOrdered listUnordered list
Emoji
Image
Align leftAlign centerAlign rightToggle HTML viewToggle full pageToggle lights
Drop image/file