"Best" hand rankings vs Suggested Ranges

Roy RentesRoy Rentes Red Chipper Posts: 14 ✭✭
Hi guys,

I have some confusion when trying to construct ranges. I don't understand how some ranges are constructed, and would love some feedback, as i would like to construct something as close to "optimal" (impossible?) as possible.

1 - There are several sites that rank the opening hands from 1-169.
e.g: http://www.preflophands.com/
2 - There was a study I found that said that the most-winning online players played around 15% of their hands in 9-ring cash games.

Therefore, we take the top 15% and the resulting range looks absolutely nothing like the ranges anywhere else on the net, including those suggested by RCP
For example, 15% ranges often include A5o+ and A2s+. They also include 22+.
BUT, a lot of these single ace hands and small pairs do not show up in the best 15% hands as suggested by the sites that rank hands. So why are they there????

I've not seen any site explain this. I really must be missing something.
Any advice or insight would be much appreciated.

Roy.

Comments

  • LeChiffreLeChiffre NetherlandsRed Chipper Posts: 197 ✭✭
    edited May 15
    15% of your range is around 200 combos. (of all possible 1326 combos you could have)
    15% of the 169 possible hands, while it might indicate 25 different hands, does not reflect 15% of your entire range.

    Example
    If I told you to play 1/169th of your range, according to the ranking you would only play AA. However, 1/169 is around 0.6%, which equates to 12 combos (AA and KK) out of all possible 1326 combos.

    Conclusion:
    If you want to consider X percentage of your range, this should not be taken from this hand ranking, but should be based on combos. That's what they mean, anyway.

    Which 15% of combos to take is a different issue, but looking at the hand ranking won't get you far. You have to consider a lot and frankly I'm not quite qualified to answer well. What I can tell you is that raw equity is not enough of a reason to include a hand in your 15%. Have to account for position, mistake propensities of opponents, stack depths, playability, blockers, etc.

  • persuadeopersuadeo Red Chipper, Table Captain Posts: 3,505 ✭✭✭✭✭
    In classic RCP tradition, i will quote myself:
    persuadeo wrote: »
    Ranges are themselves constructed from showdown backwards in a preflop stack off scenario mitigated by depth

    and, i should add, position, which you will quickly see, is where the real rub is. Sounds like a lot of work because it is.
  • Roy RentesRoy Rentes Red Chipper Posts: 14 ✭✭
    LeChiffre wrote: »
    15% of your range is around 200 combos. (of all possible 1326 combos you could have)
    15% of the 169 possible hands, while it might indicate 25 different hands, does not reflect 15% of your entire range.

    I accounted for that. I input the hands into my range application, starting at 1/169 and worked down through the list until i had reached 15% of combos.
    Works out at 35/169 hands (or 204/1326 combos)
  • TheGameKatTheGameKat Posts: 1,086 -
    Duplicate post from the other thread:

    I believe the older rankings from 1-169 were based on hot-and-cold equity. That is, how does each hand perform against every other hand assuming we just run out the board.

    But that's not how poker works in practice. The potential to bet on each street folds out equity. So playability and the ability to reach showdown with the possibility of multiple bets has to be factored in to actual ranges.

    In more detail, different range constructions put differing amounts of emphasis on issues like board coverage. For example you'll often see an opening range that includes the specific hand 76s. The overall range might include less than 20% of hands. If you look at a 1-169 list of hand rankings I think 76s comes in around 45%. But constructing a range that includes hands like 76s allows us to connect with low flops.

    Going further, stack depth can impact range construction. Hands like 76s do better when stacks are relatively deep, whereas in a shallow-stacked situation you might throw out suited connectors and some suited aces in favor of offsuit broadways.

    Hope that helps a little. It's a big topic.
    Moderation In Moderation
  • WickedWicked Red Chipper Posts: 60 ✭✭
    There literately are no straight forward answers to questions about range construction. It's always some double speak about cosmic influxes of mythical questions leading to more questions based on imagined scenarios. Being new to this stuff is incredibly frustrating sometimes.....most times!
  • BigFarmBigFarm MontrealRed Chipper Posts: 15 ✭✭
    I would just reiterate the comment about stack depth being very important. If you were playing a hypothetical game where everyone has say 5 BB stacks, then your range is going to look a lot more like the top hands in the hot-cold equity list. I haven't checked but I'm guessing push-fold charts will be similar because you don't have to worry about playability. In fact, they won't be identical though because you aren't optimizing against "any two". You are optimizing against the calling range.
  • LeChiffreLeChiffre NetherlandsRed Chipper Posts: 197 ✭✭
    Roy Rentes wrote: »
    LeChiffre wrote: »
    15% of your range is around 200 combos. (of all possible 1326 combos you could have)
    15% of the 169 possible hands, while it might indicate 25 different hands, does not reflect 15% of your entire range.

    I accounted for that. I input the hands into my range application, starting at 1/169 and worked down through the list until i had reached 15% of combos.
    Works out at 35/169 hands (or 204/1326 combos)

    Right, so your confusion lies in how those 204 combos are arranged, rather than what I thought you were confused about (hands vs combos)?
  • Roy RentesRoy Rentes Red Chipper Posts: 14 ✭✭
    Wicked wrote: »
    There literately are no straight forward answers to questions about range construction. It's always some double speak about cosmic influxes of mythical questions leading to more questions based on imagined scenarios. Being new to this stuff is incredibly frustrating sometimes.....most times!

    Agreed, which is why i was trying to find a mathematically optimal solution (as a base point at least)
  • Roy RentesRoy Rentes Red Chipper Posts: 14 ✭✭
    LeChiffre wrote: »
    Roy Rentes wrote: »
    LeChiffre wrote: »
    15% of your range is around 200 combos. (of all possible 1326 combos you could have)
    15% of the 169 possible hands, while it might indicate 25 different hands, does not reflect 15% of your entire range.

    I accounted for that. I input the hands into my range application, starting at 1/169 and worked down through the list until i had reached 15% of combos.
    Works out at 35/169 hands (or 204/1326 combos)

    Right, so your confusion lies in how those 204 combos are arranged, rather than what I thought you were confused about (hands vs combos)?

    Yes.
    The 204 combos as suggested by the "best" 15% of hands don't correlate closely with the average suggested TAG ranges.
    Theirein is my confusion.
  • Roy RentesRoy Rentes Red Chipper Posts: 14 ✭✭
    Roy Rentes wrote: »
    Yes.
    The 204 combos as suggested by the "best" 15% of hands don't correlate closely with the average suggested TAG ranges.
    Theirein is my confusion.

    A specific example would be why A5o+ appear is most ranges, but not the top 15%.
    Hands ranked 1-169 only have AJo+ within that range...
    Where does the A5o -ATo come from?
  • AustinAustin Red Chipper Posts: 5,486 ✭✭✭✭✭
    start with around 10% UTG and add 4% to each postion after.

    Utg 10%
    Utg1 14%
    Mp1 18%
    LJ 22%
    Hj 26%
    Co 30%
    Btn 34%

    Co and BTN could be higher in terms of RFI.
  • jeffncjeffnc Red Chipper Posts: 4,240 ✭✭✭✭✭
    Wicked wrote: »
    It's always some double speak about cosmic influxes of mythical questions leading to more questions based on imagined scenarios.

    I think he means that's the way it seems to newcomers.

  • moishetreatsmoishetreats Red Chipper Posts: 1,705 ✭✭✭✭
    @persuadeo wrote
    "Ranges are themselves constructed from showdown backwards in a preflop stack off scenario mitigated by depth"

    Agreed.

    Let's try this example to help explain (also borrowed from @TheGameKat's post):

    A3o is a better preflop hand than 76s if you're going by which is most likely to win heads-up. That being said, which would you rather play post-flop?

    If an ace comes on the board, you're entirely guessing with A3o -- not a comfortable (read: profitable) way to play. On the other hand, with 76s, you'll have two distinct advantages: (1) it will be pretty clear if you have the equity to continue or if you have minimal to no equity; and (2) you have a likely chance to improve to the best hand.

    If you're playing heads-up and if V shoves pre-flop and turns over 76s, then, yes, you should call with A3o: it's simply a better starting hand in that scenario. Otherwise, 76s is a faaaaaar more playable hand -- both in terms of decision-making and profitability post-flop.

    @TheGameKat explains the next steps in your thinking in his post.
  • persuadeopersuadeo Red Chipper, Table Captain Posts: 3,505 ✭✭✭✭✭
    Austin wrote: »
    start with around 10% UTG and add 4% to each postion after.

    Utg 10%
    Utg1 14%
    Mp1 18%
    LJ 22%
    Hj 26%
    Co 30%
    Btn 34%

    Co and BTN could be higher in terms of RFI.

    Reasonable but he is asking where it comes from, while mixing up hot and cold equities with postflop EV nuances.
  • AustinAustin Red Chipper Posts: 5,486 ✭✭✭✭✭
    persuadeo wrote: »
    Austin wrote: »
    start with around 10% UTG and add 4% to each postion after.

    Utg 10%
    Utg1 14%
    Mp1 18%
    LJ 22%
    Hj 26%
    Co 30%
    Btn 34%

    Co and BTN could be higher in terms of RFI.

    Reasonable but he is asking where it comes from, while mixing up hot and cold equities with postflop EV nuances.



    Video is just a guideline and i know part of the range or how it is described some players here, including myself will disagree with.

    Helps explain some of the playability vs raw equity regarding preflop hands.

    Example in the video is 22 vs AKs
    Ts7h3d board which hand would you rather have postflop? 22 has 68% equity but given there are bets to be made on 3 streets AKs plays way better.
  • jeffncjeffnc Red Chipper Posts: 4,240 ✭✭✭✭✭
    I've always liked the 22 vs AKo vs 98s example. Which hand do you want? Depends on what hand you're up against!

    98s beats 22, 22 beats AK, and AK crushes 98s! (AK ends up on top in the manage a trois, by the way).

    That's like a math equation that says A>B, B>C, C>A which is impossible.

    Which just goes to show, you need some flexibility in your thinking of hand rankings, and also some consideration of the fundamentals of poker, to evaluate.
  • moishetreatsmoishetreats Red Chipper Posts: 1,705 ✭✭✭✭
    Austin wrote: »
    start with around 10% UTG and add 4% to each postion after.

    Utg 10%
    Utg1 14%
    Mp1 18%
    LJ 22%
    Hj 26%
    Co 30%
    Btn 34%

    Co and BTN could be higher in terms of RFI.

    Question: Doesn't this present a board coverage issue? Or is that something that you're not worried about being exploited at lower-stakes?
  • AustinAustin Red Chipper Posts: 5,486 ✭✭✭✭✭
    Austin wrote: »
    start with around 10% UTG and add 4% to each postion after.

    Utg 10%
    Utg1 14%
    Mp1 18%
    LJ 22%
    Hj 26%
    Co 30%
    Btn 34%

    Co and BTN could be higher in terms of RFI.

    Question: Doesn't this present a board coverage issue? Or is that something that you're not worried about being exploited at lower-stakes?

    As @kenaces pointed out to me in another thread I was likely leaking money at the low stakes raising some SCs from EP. I used this as board coverage and to stay some what active on the table. Don't wanna appear as super nitty. I felt my read was good enough to not get into trouble with these hands when i flop say top pair and face aggression. Most of the time you end up MW and don't know if your low straight for example 98s on JTQ is good when they call with Ak or K9s. Most low stakes players I notice are not 3 betting AK.

    My range is not something you just slide over and get a percentage with.
  • WickedWicked Red Chipper Posts: 60 ✭✭
    First off, I would like all of you to know that I very much respect the time you take to try and impart your considerable knowledge on newbies to these concepts like myself. I meant no disrespect with my post. I am a very literal thinker. Flip switch, light comes on. Don't need to know the theory behind electricity for this. I haven't played since black friday, and when I started back up I realized I needed to up my game, so here I am. My problem with range construction is that all of you are on a vastly higher level of thinking with this already. I am trying to get the " I'm an idiot" version of bare bones range construction concepts together before moving on to GTO optimal construction for every situation and position. Thank you all again for your help with this and I apologize for being a slow learner.
  • ChipXtractorChipXtractor Red Chipper, Table Captain Posts: 1,184 ✭✭✭✭
    edited May 17
    Oops. Didnt mean to post twice ... sorry.
    Twitter = @ChipXtractor
  • WickedWicked Red Chipper Posts: 60 ✭✭
    I feel like I hijacked Roy’s thread here. My question was if I construct a range and open a hand from that range and it misses the flop, if I decide to continue, isn’t that hand now in my bluff range? If someone opens in front of me and I decide to 3bet with my hand isn’t it now part of my 3bet range? Aren’t all these different ranges just subsets of my original range on a super simple level? I realize that with time and experience these get fine tuned, but isn’t this thinking correct on a base level?
  • jeffncjeffnc Red Chipper Posts: 4,240 ✭✭✭✭✭
    Wicked wrote: »
    I feel like I hijacked Roy’s thread here. My question was if I construct a range and open a hand from that range and it misses the flop, if I decide to continue, isn’t that hand now in my bluff range?

    Not necessarily. If you raise AK and get called by QJ, and the flop is T64, and you cbet, are you bluffing?

  • WickedWicked Red Chipper Posts: 60 ✭✭
    I'm asking in a way zoomed out view. I realize there are a million variables, but Im asking as a bare bones base to start from. am I totally off base with this thinking?
  • jeffncjeffnc Red Chipper Posts: 4,240 ✭✭✭✭✭
    edited May 18
    Wicked wrote: »
    Aren’t all these different ranges just subsets of my original range on a super simple level?

    Yes, but that seemed kind of obvious so I wasn't sure that was your question :)

  • WickedWicked Red Chipper Posts: 60 ✭✭
    I thought it was obvious, but getting that answer took a lot of time and several posts. Having that base and knowing that is correct lets me start expanding. I just didn't want to start off on the wrong line of thinking. Thank you!
  • Roy RentesRoy Rentes Red Chipper Posts: 14 ✭✭
    Hi lads.
    Cheers for all the feedback.
    For anyone who may be confused as to what I meant, here is a screenshot of the ranges I constructed. Based on 15% overall range taken from ranking site ranking hands 1->169

    cv1y70bfkwqc.jpg

    I understand this is not ideal in the real world due to playability, coverage etc.
  • persuadeopersuadeo Red Chipper, Table Captain Posts: 3,505 ✭✭✭✭✭
    edited May 21
    Alrighty, a rational starting point - a truly linear range, a kind of automatic, air-bagged car for a scary world full of crazy drivers.

    Now we test it out, and we'll assume the simplified game of 100 bbs.

    Assume you are three bet from the button and laid 2:1. Investigate the branches of what happens next.

    What folds, what calls, what 4 bets (and then folds or calls off)? If we don't like where we end up at any one point, what do we adjust?

    In fact, how do we even think about that?

    Anyway, as we examine the interaction of this range with our opponent's range, we'll learn where our biggest leaks are quickly - as well as earn a general idea of what range we have in any one spot. More importantly, we'll learn why we have it, because this toy game is likely not to be ever repeated, and it's the principles that end up truly mattering.

Leave a Comment

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