The Evolution of the Perfect Stat

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts
  • Sultan_1895-1948
    Prince of Pounders
    • Sep 2005
    • 11171

    #16
    I actually like WAA/650PA a lot. Offense, defense, and baserunning (even the SB% heavy Rbaser) is included.

    We take the career WAA total and divide by their individual PA. Sure, a guy who stuck around racking up a lot of PA to get a milestone and played below league avg, will be hurt. Rose for example. But that's his career, it should be evaluated and in a sense it was a sacrifice. Why should a player like that get his cake and eat it too.

    Anyway, for further perspective, I'm thinking about putting in parentheses, the career WAA total, and then possibly each players top five WAA seasons along with PA in those seasons.

    So it would look this this: Here's Ripken and Gehrig. Different positions for a reason; keep in mind this stat rewards Ripken heavily for his position.

    --------------WAA/650------------BEST 5 WAA-----------BEST 5 WAA/650

    Gehrig ---- 5.287 (78.6) --------38.0 (3506 PA)---------------7.04
    Ripken ---- 2.684 (53.2) ------- 32.5 (3561 PA)---------------5.93

    Ripken doesn't look so bad when you look at best five years. In fact pretty darn good.

    Ripken is gonna look okay as long as fielding is included. When you just look at his hitting, it's pretty bad unless it's only compared to middle infielders. Hell, Ripken's career bWAA is about exactly half of Rose's (19.60 to 38.39).

    Actually, gonna look up some other middle infielders WAA numbers.

    First is sorted by career total, second by best 5 WAA/650

    --------------WAA/650------------BEST 5 WAA-----------BEST 5 WAA/650
    -----------(CAREER TOTAL)

    Wagner-----5.068 (91.6)--------38.8 (3018 PA)----------------8.35
    Collins------4.259 (78.9)---------34.7 (3307 PA)---------------6.82
    Arod--------4.417 (77.1)--------35.4 (3497 PA)----------------6.58
    Morgan-----3.631 (63.3)--------37.8 (3257 PA)----------------7.54
    Ripken------2.684 (53.2)--------32.5 (3561 PA)----------------5.93
    Gehringer--2.887 (45.5)---------26.5 (3515 PA)---------------4.90
    Grich-------3.439 (43.5)---------24.0 (3081 PA)---------------5.06
    Larkin------3.028 (42.2)---------22.4 (2982 PA)---------------4.88
    Trammel---2.779 (40.1)---------24.1 (3107 PA)---------------5.04
    JRobinson--4.401 (39.3)---------29.9 (3169 PA)---------------6.13
    Sandberg--2.668 (38.1)---------26.0 (3418 PA)---------------4.94
    Jeter-------1.759 (32.4)---------21.1 (3550 PA)---------------3.86
    Alomar-----2.018 (32.3)---------21.2 (3422 PA)---------------4.02
    Banks------1.788 (28.6)---------29.4 (3350 PA)---------------5.70
    Kent--------1.799 (26.4)---------18.1 (3318 PA)---------------3.54
    Nomar-----2.571 (24.2)---------23.4 (3273 PA)---------------4.64
    Tejada-----1.101 (15.6)---------15.8 (3556 PA)---------------2.89

    --------------WAA/650------------BEST 5 WAA-----------BEST 5 WAA/650
    -----------(CAREER TOTAL)

    Wagner-----5.068 (91.6)--------38.8 (3018 PA)----------------8.35
    Morgan-----3.631 (63.3)--------37.8 (3257 PA)----------------7.54
    Collins------4.259 (78.9)---------34.7 (3307 PA)---------------6.82
    Arod--------4.417 (77.1)--------35.4 (3497 PA)----------------6.58
    JRobinson--4.401 (39.3)---------29.9 (3169 PA)---------------6.13
    Ripken------2.684 (53.2)--------32.5 (3561 PA)----------------5.93
    Banks------1.788 (28.6)---------29.4 (3350 PA)---------------5.70
    Grich-------3.439 (43.5)---------24.0 (3081 PA)---------------5.06
    Trammel---2.779 (40.1)---------24.1 (3107 PA)---------------5.04
    Sandberg--2.668 (38.1)---------26.0 (3418 PA)---------------4.94
    Gehringer--2.887 (45.5)---------26.5 (3515 PA)---------------4.90
    Larkin------3.028 (42.2)---------22.4 (2982 PA)---------------4.88
    Nomar-----2.571 (24.2)---------23.4 (3273 PA)---------------4.64
    Alomar-----2.018 (32.3)---------21.2 (3422 PA)---------------4.02
    Jeter-------1.759 (32.4)---------21.1 (3550 PA)---------------3.86
    Kent--------1.799 (26.4)---------18.1 (3318 PA)---------------3.54
    Tejada-----1.101 (15.6)---------15.8 (3556 PA)---------------2.89

    I may have to move Morgan up a bit in my rankings.

    Comment

    • leewileyfan
      Registered User
      • Feb 2009
      • 2122

      #17
      Originally posted by BigRon View Post
      It doesn't matter. both WAR and WAA are comparing performance against a baseline. In the case of WAR, it's against a player who is only good enough, if his whole team was composed of guys with his performance, to win about 1/3rd of their games. With WAA, it's comparing to a team of average MLB performers who would win 50% of their games.
      It matters in one sense: What purpose, if any, is served by the specific
      base line for player comparison?

      The only sensible purpose I can see for WAR is one of economics in budget-
      making for optimal team performance at minimal cost. Why else
      would anyone be willing and/or eager to explore very marginal prospects,
      a collection of whom would bring you a 52-110 W-L record [or thereabouts]?

      Perhaps my problem is "Replacement," the word itself. From my
      perspective growing up, the expressions "utility," "sub," second stringer" or
      "bench" indicated a player worthy of a spot on the roster, but only as a
      fill-in in case of injury, illness, spat with the manager or tea mates, or
      other reason, rational or not. It was the guys on waivers, or "player to
      be named later."

      "Replacement" for me was a matter of players willing to cross picket lines
      during a MLB strike for a fleeting shot of playing in a MLB game.

      :Frankly, I don't see any advantage of one over the other, but it's easy to look at either. Just don't get too granular- 5.6 WAR probably really isn't any better than 5.2 WAR- there's too much noise in the system, particularly on the defensive side.
      If I am to compare players at the MLB level fairly, I want to compare them
      to their peers; or, if I want to cross over generational fences with different
      performance "climates," I want to make clear the generational differences OR
      or explain the steps taken to adjust for seasonal disparities.

      As I see it [and speaking only for myself], WAR is a front office tool.
      "Anything" 'above average' takes care of everything else.

      Comment

      • Sultan_1895-1948
        Prince of Pounders
        • Sep 2005
        • 11171

        #18
        I agree Lee, that's kinda my main problem with WAR is the "replacement" aspect. In reading about it, apparently a consensus value for this "replacement" guy hadn't even been agreed to among the saber community until recently.

        I think we should always be searching for a way to compare players to their peers. I like the out/RC stat as it shows if someone pissed away a lot of outs to create the same amount of runs as someone else. Isn't that the name of the game after all? Not making outs and scoring runs. My only problem is I'm trying to somehow tie in different run scoring environments. Runs/game has an issue because there were so many errors back in the day. I suppose using just earned R/G would get us closer. Maybe if we had league RC numbers we could get a league avg OUT/RC and get sort of a OUT/RC+ number. Anyway, just rambling.

        Comment

        • leewileyfan
          Registered User
          • Feb 2009
          • 2122

          #19
          Sultan: I like the RC/650 PA approach. In fact I have a spreadsheet that
          very nearly does the same thing, except that I used 600 PA to equate PA to
          a fairly typical regular player "season." To track several of the players you
          listed, I simply changed the "season" denominator from 600 to 650. The DR
          portion was already in my spreadsheet. We are close on many players and pretty
          far apart on others. Once I started, I kept going, cutting off the listing at 2.00
          Wins above Average per season. I later did the career totals by multiplying
          that number by "seasons" to get Career Wins above Average.

          Just a few caveats:

          1. I had neither the time nor the inclination to parse a player's RC and
          PA only at a particular position. IMO, we are looking at players from
          two perspectives [actually three at least]: defense, batting run production ...
          with the third element being the compilation of career stats [+/-] due to
          longevity in the game.

          2. I add DR into the production equation only at the primary position
          at which the player is most identified. Rod Carew, for the purpose of this post,
          is a 2B. His batting contribution is everywhere.

          Here are the players, mid-infield, from highest Wins>AVG. season as the
          numerator and career Wins>AVG as the denominator:

          Hornsby 7.38/107.5
          Wagner 6.71/99.6
          Rodriguez 5.35/93.4
          Lajoie 4.92/75.35
          Utley 4.49/39.2
          Garciaparra 4.14/38.9
          Tulowitzki 3.94/22.4
          Gehringer 3.89/60.3
          H. Ramirez 3.82/28.0
          J. Morgan 3.63/63.2
          J. Robinson 3.43/30.6
          Cano 3.35/29.8
          Banks 3.29/52.6
          Kent 3.17/46.5
          Vaughn 3.14/37.3
          Cronin 2.99/40.7
          V. Stephens 2.95/32.8
          E. Collins 2.95/54.6
          R. Alomar 2.95/47.1
          O. Hudson 2.88/17.9
          Sandberg 2.85/40.6
          Gordon 2.76/27.7
          Lazzeri 2.73/30.6
          Frisch 2.72/42.2
          Stanky 2.61/21.8
          Carew 2.58/41.9
          Grich 2.50/31.7
          Furcal 2.49/25.5
          Boudreau 2.49/26.3
          Whitaker 2.43/37.2
          Doerr 2.40/29.6
          Larkin 2.39/33.3
          Lopes 2.31/26.1
          Tr. Jackson 2.30/23.7
          Ripken 2.26/44.8
          Trammel 2.26/32.6
          G. Wright 2.11/14.9
          Appling 2.07/32.6
          Jeter 2.02/36.9

          Below the 2.00 level, but relatively high in Career "Wins:"

          Yount, Career 35.4
          Biggio, Career 35.6
          Buddy Myer, Career 23.4

          EDIT NOTE: When I posted, I had forgotten that each player's
          RC/PA was calculated with the approximate average during seasons
          of play, in gaps of .0025. Thus a player in a weak batting climate
          might be compared to a model of .1050, while a player in a robust
          offensive climate might be compared to a .1225 base.

          I found an error with Eddie Collins.

          He should be 3.40/64.2.
          Last edited by leewileyfan; 03-19-2014, 08:17 AM.

          Comment

          • abolishthedh
            Hopeful traditionalist
            • Mar 2003
            • 1932

            #20
            With all apologies for the buzzkill, I don't believe there should be a perfect stat for ballplayers. Baseball stats for individual players, especially at a specific point in time, should be uniquely descriptive.

            As with the weather, each day is best described as windy, humid, hot, cold, a chance for rain, how much it rained or snowed, wind chill, heat index, allergy index, or the change in temperature throughout the day. Finding one stat for all of that would be pointless.

            I like having different stats for ballplayers, especially for the same ballplayer over the course of his career. It shows how things change, or how his position and role on the team adds up, whether he is paid what he is worth, or whether he should be benched or traded or sent back to the minors. One stat which is one-size-fits-all is one stat too many.
            Catfish Hunter, RIP. Mark Fidrych, RIP. Skip Caray, RIP. Tony Gwynn, #19, RIP

            A fanatic is someone who can't change his mind and won't change the subject. -- Winston Churchill. (Please take note that I've recently become aware of how this quote applies to a certain US president. This is a coincidence, and the quote was first added to this signature too far back to remember when).

            Experience is the hardest teacher. She gives the test first and the lesson later. -- Dan Quisenberry.

            Comment

            • Yankillaz
              I smell Sex and Candy
              • Feb 2006
              • 4924

              #21
              I understand. I wanted to ask something taking what abolish wrote: how can we adjust eras? Like for example: The 90s and the 2000s were at an all time offensive this will be valued as a full 1.00. But how about the 60s? Shouldn't we take into account that due to the low enviroment.
              "I am not too serious about anything. I believe you have to enjoy yourself to get the most out of your ability."-
              George Brett

              Comment

              • leewileyfan
                Registered User
                • Feb 2009
                • 2122

                #22
                Originally posted by Yankillaz View Post
                I understand. I wanted to ask something taking what abolish wrote: how can we adjust eras? Like for example: The 90s and the 2000s were at an all time offensive this will be valued as a full 1.00. But how about the 60s? Shouldn't we take into account that due to the low enviroment.
                What I find to be a simple and direct solution to generational differences
                is to have a single source of reference for the offensive climate each season.

                I use Baseball-Reference. For each season, I refer to BATTING
                and jot down Runs Scored and Plate Appearances. With Runs as the numerator and PA as the denominator, I divide and get a rate of run generation to 4 decimal places.

                With this data, I have a basis for comparison within a season. Or, if i set up a spreadsheet that defines decades, general "generations" of play for less specific "averages," I can make comps between generations, even if they are far apart.

                A general rule of thumb might be an overall "average" of .1200 runs/pa; but hat will be biased against the outliers like 1908 and 1968, when the reality was more like .095.

                Comment

                • Yankillaz
                  I smell Sex and Candy
                  • Feb 2006
                  • 4924

                  #23
                  Good Lee, I will try something like that. Thx.
                  "I am not too serious about anything. I believe you have to enjoy yourself to get the most out of your ability."-
                  George Brett

                  Comment

                  • Second Base Coach
                    Tabletop Baseball Fan
                    • May 2006
                    • 4552

                    #24
                    Some info for you if you are interested in using runs scored per year per league:

                    National League Batting Year-by-Year Averages, Batting Totals and more on Baseball-Reference.com


                    American League Batting Year-by-Year Averages, Batting Totals and more on Baseball-Reference.com
                    Your Second Base Coach
                    Garvey, Lopes, Russell, and Cey started 833 times and the Dodgers went 498-335, for a .598 winning percentage. That’s equal to a team going 97-65 over a season. On those occasions when at least one of them missed his start, the Dodgers were 306-267-1, which is a .534 clip. That works out to a team going 87-75. So having all four of them added 10 wins to the Dodgers per year.
                    http://www.youtube.com/watch?v=p5hCIvMule0

                    Comment

                    Ad Widget

                    Collapse
                    Working...
                    X