Def and Fwd Rankings

Discussion in 'Leaguegaming American Hockey League (LGAHL PSN)' started by big_sleezy00, Jul 20, 2019.

  1. BigSleezy00

    BigSleezy00 EAID: BigSleezy00

    Joined:
    Jul 7, 2018
    Messages:
    107
    Discord:
    bigsleezy#0
    Trophy Points:
    1,241
    Location:
    California
    Ratings:
    +178 / 3 / -7
    This is my own ranking structure. All games included up to week 6. Check the end for best def,off,and overall for fwd's and def. only players included are those with 10 games, that's to keep the outliers out of the mix. EX. a tc comes in and tears it up for 1 game. They then take over the stat/game rankings and would be statistical outliers. So anyone with less than 10 games is excluded.

    https://drive.google.com/file/d/1-kzC3qDBuFA2x8XrQ3XDZR5wIJ_CJs5r/view?usp=sharing
     
    • Informative x 4
    • Like x 3
    • Dislike x 1
    • Funny x 1
    • Dumb x 1
    Offline
    Big_Sleezy
  2. ItsKingLyme

    ItsKingLyme twitch.tv/itskinglyme

    Joined:
    Jan 13, 2017
    Messages:
    162
    Discord:
    ItsKingLymeTTV#5441
    Trophy Points:
    1,521
    Location:
    twitch.tv/itskinglyme
    Ratings:
    +419 / 6 / -19
    How the fuck do you rank best offensive and defensive fwds smh
     
    • Dislike Dislike x 1
    Offline
    ItsKingLyme
  3. BigSleezy00

    BigSleezy00 EAID: BigSleezy00

    Joined:
    Jul 7, 2018
    Messages:
    107
    Discord:
    bigsleezy#0
    Trophy Points:
    1,241
    Location:
    California
    Ratings:
    +178 / 3 / -7
    Alright, if you want to be butthurt right off the hit and not come at me with some respect and be polite i'll do you the same justice. Since this guy doesn't have mental gumption to download the doc and look at how the math is done, i'll lay it out. @ItsKingLyme you can score all the goals you want, you can be a great nhl player, but until you learn some respect you can continue to sit at the kiddie table and be a pea brained knuckle dragger. Now let's get into it.

    The stats are all pulled straight from this website. Member stats and Advanced member stats.
    Offensive forwards are done like this:

    it ranks each and every player from first to last in these categories. (+/-)/game, penalty mins/game, goals against/game, points/game, goals/game. the ranks from all of these stat lines are averaged, then that number is ranked against all the other forwards. basically whoever has the lowest average ranks among all these stat lines show up higher in the offensive forward list.

    defensive forward is just like offensive, except it's these stat lines:
    (+/-)/game, penalty mins/game, goals against/game, penalty kill %.
    for forwards,(ta/ga)/game is irrelevant when all of the other stats are taken into account. because efficiency as a forward is calculated in plusminus/game and the other stats.

    overall for forwards, is the two rank averages added together and averaged. so whoever is lowest on both offensive and defensive forwards will show higher on the overall. This ranking system directly compares the efficacy of every 10+ game ahl player against one another. it's a peer ranking system. Statistics can be a bitch if you suck at math.

    Defense is done like this

    OD: plus/minus per game, penaltymins/game, pts/game, penaltykill %, goals against/game

    DD:
    hits/game, penaltymins/game, (ta/ga)/game, penalty kill %, goals against/game
     
    • Like Like x 2
    • Funny Funny x 2
    • Winner Winner x 1
    • Informative Informative x 1
    Offline
    Big_Sleezy
  4. LG McDonald

    LG McDonald Director of Hockey Operations

    Joined:
    Sep 30, 2013
    Messages:
    4,957
    Discord:
    McDonald#2877
    Trophy Points:
    7,096
    Location:
    Ottawa, ON
    Ratings:
    +12,036 / 113 / -852
    So Math is the best forward and I am third?

    Has to be a calculation issue the guy sucks
     
    • Like Like x 1
    Offline
    LeagueGamingOfficial
  5. Brydxn

    Brydxn Well-Known Member

    Joined:
    Jun 12, 2016
    Messages:
    42
    Trophy Points:
    2,651
    Location:
    Toronto
    Ratings:
    +26 / 2 / -4
    Yikes
     
    Offline
    Brydxn
  6. nolesfan12

    nolesfan12 Well-Known Member

    Joined:
    Jan 19, 2017
    Messages:
    61
    Trophy Points:
    741
    Location:
    York, PA
    Ratings:
    +19 / 35 / -29
    I mean I know this is far from perfect because we don't have a lot of the stats you would actually need for this, but what does pims have to do with offense? Even for defense it's a stretch honestly, but penalties have literally nothing to do with your offense
     
    Offline
    noles_fan12
  7. BigSleezy00

    BigSleezy00 EAID: BigSleezy00

    Joined:
    Jul 7, 2018
    Messages:
    107
    Discord:
    bigsleezy#0
    Trophy Points:
    1,241
    Location:
    California
    Ratings:
    +178 / 3 / -7
    But they do have a lot to do with a players efficiency and it impacts the game. In the end, pims per game matter when you rank players against each other.

    Also, when I'm comparing offense vs defense ranking for a player you treat it like it's own category. Which is why you include some of the same stats. Not perfect by any means, however you can glean a lot of information from it.
     
    Offline
    Big_Sleezy
  8. Userpic1306

    Userpic1306 Warning Group

    Joined:
    Jun 2, 2016
    Messages:
    311
    Trophy Points:
    5,091
    Location:
    Pittsburgh
    Ratings:
    +674 / 15 / -45
    Having a winger as best defensive fwd is a yikes
     
    No Streaming Account
  9. XMeLLoFeTT

    XMeLLoFeTT Not a Goalie

    Joined:
    Jul 31, 2016
    Messages:
    160
    Trophy Points:
    1,602
    Location:
    Canada
    Ratings:
    +260 / 14 / -19
    Meese.
     
    • Winner Winner x 2
    • Like Like x 1
    Offline
    xmellofett
  10. WetLikeKyleLowry

    WetLikeKyleLowry Champion

    Joined:
    Feb 5, 2018
    Messages:
    62
    Trophy Points:
    2,151
    Location:
    toronto
    Ratings:
    +78 / 6 / -10
    Would be much better if you factored fights into your algorithm
     
    No Streaming Account
  11. ARoguePineapple

    ARoguePineapple Well-Known Member

    Joined:
    Nov 1, 2013
    Messages:
    267
    Discord:
    ARoguePineapple#9357
    Trophy Points:
    3,591
    Ratings:
    +425 / 2 / -10
    WOOHOO 117th best forward all around! Suck it the 127 players behind me, y'all must be doodoo if I'm better than you :p
     
    • Funny Funny x 2
    Offline
    ARoguePineapple
  12. BigSleezy00

    BigSleezy00 EAID: BigSleezy00

    Joined:
    Jul 7, 2018
    Messages:
    107
    Discord:
    bigsleezy#0
    Trophy Points:
    1,241
    Location:
    California
    Ratings:
    +178 / 3 / -7
    aggregation of penalty kill %, goals against, and pims don't lie. this is completely mathematical by the way. i haven't seen a lot of these guys play.
     
    Offline
    Big_Sleezy
  13. nolesfan12

    nolesfan12 Well-Known Member

    Joined:
    Jan 19, 2017
    Messages:
    61
    Trophy Points:
    741
    Location:
    York, PA
    Ratings:
    +19 / 35 / -29
    There are glaring holes in this and I'm trying to be nice, but you're making it difficult. You are way overselling pims. Like if I take a penalty to stop a 3 on 1 that's bad on my part? That's wild. Also your pims could be all from fighting in the final minute of the game. That makes you bad on both offense and defense?
     
    • Get Good Get Good x 1
    Offline
    noles_fan12
  14. BigSleezy00

    BigSleezy00 EAID: BigSleezy00

    Joined:
    Jul 7, 2018
    Messages:
    107
    Discord:
    bigsleezy#0
    Trophy Points:
    1,241
    Location:
    California
    Ratings:
    +178 / 3 / -7
    you don't understand how it works. just because penalty minutes play a factor does not mean it's the deciding factor. all ranks are taken from an average of ranks. if guy #1 has 10 pims and the guy right after him has 20 pims, they are still ranked 1 and 2. i'm not making this hard, i'm telling you that you haven't looked at the formula, and if you did you would see it doesn't play as much as a factor as you think. it's a factor yes, however it's not game breaking. also, much of this stuff is modeled after sabermetrics. if you don't know what that is, then you should check it out. if you don't like it, then give some constructive feedback, like suggestions. give me something mathematical to go off of. I know it's not perfect. However, it's far from nothing as well.

    -cheers
     
    Offline
    Big_Sleezy
  15. nolesfan12

    nolesfan12 Well-Known Member

    Joined:
    Jan 19, 2017
    Messages:
    61
    Trophy Points:
    741
    Location:
    York, PA
    Ratings:
    +19 / 35 / -29
    This isn’t what sabermetrics are like at all. You’re using a bunch of flawed stats that are missing a ton of important information to try to reach a conclusion. I applaud you for putting the time in, and I realize the data set you are working with is all you have. With that being said, it doesn't excuse the drastic flaws in this. You say pims aren't that big of a part of it, but everything else up until this point has suggested it's weighted equally with everything else. So where can I see the actual formula you are using?
     
    • Get Good Get Good x 1
    Offline
    noles_fan12
  16. BigSleezy00

    BigSleezy00 EAID: BigSleezy00

    Joined:
    Jul 7, 2018
    Messages:
    107
    Discord:
    bigsleezy#0
    Trophy Points:
    1,241
    Location:
    California
    Ratings:
    +178 / 3 / -7
    It's multiple cells running after the ga start from the LG stats. Unhide the cells, and currently pims are ranked equally.
     
    Offline
    Big_Sleezy
  17. nolesfan12

    nolesfan12 Well-Known Member

    Joined:
    Jan 19, 2017
    Messages:
    61
    Trophy Points:
    741
    Location:
    York, PA
    Ratings:
    +19 / 35 / -29
    So pims are a big part of it then. They are literally as important as everything else. They shouldn't be. If we were really going advanced stats you would have to factor in who each player played with in every single game as well. This isn't as complex as you are trying to make it seem. This is very low level statistics. I'm personally not willing to put this kind of effort in so again I give you props for that, but transparency is extremely important when discussing statistics. I have seen enough of these people play to know that there are extreme flaws in the data set. That's not necessarily your fault because we are pretty limited, but the flaws still exist regardless
     
    • Get Good Get Good x 1
    Offline
    noles_fan12
  18. TiberiussX

    TiberiussX 500k Dream

    Joined:
    Jan 26, 2018
    Messages:
    42
    Trophy Points:
    2,041
    Location:
    Oshawa
    Ratings:
    +31 / 2 / -3
    Statistics are a bitch
     
    Last edited: Jul 21, 2019
    No Streaming Account
  19. nolesfan12

    nolesfan12 Well-Known Member

    Joined:
    Jan 19, 2017
    Messages:
    61
    Trophy Points:
    741
    Location:
    York, PA
    Ratings:
    +19 / 35 / -29
    Yeah, low IQ individuals such as yourself struggle with understanding them
     
    • Dumb Dumb x 5
    Offline
    noles_fan12
  20. TiberiussX

    TiberiussX 500k Dream

    Joined:
    Jan 26, 2018
    Messages:
    42
    Trophy Points:
    2,041
    Location:
    Oshawa
    Ratings:
    +31 / 2 / -3
    Looks like someone didn’t like where he ended up on the list..
     
    • Like Like x 1
    • Funny Funny x 1
    No Streaming Account