Announcement

Collapse
No announcement yet.

Does WAR literally mean that player earned the team those wins by himself?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Does WAR literally mean that player earned the team those wins by himself?

    Or is it just an expression? Like if you take Trout off the Angels and put in a replacement would they lose 10 games undoubtedly? And how would they determine this?

  • #2
    No. It's a cumulative stat. Players earn parts of wins and lose them all the time due to good and bad performance.
    46 wins to match last year's total

    Comment


    • #3
      Originally posted by SamtheBravesFan View Post
      No. It's a cumulative stat. Players earn parts of wins and lose them all the time due to good and bad performance.
      I see.. so a 10 WAR player doesnt LITERALLY win his team 10 games in a season by himself correct?

      Comment


      • #4
        Originally posted by donnie72191 View Post

        I see.. so a 10 WAR player doesnt LITERALLY win his team 10 games in a season by himself correct?
        By definition "The idea behind the WAR framework is that we want to know how much better a player is than a player that would typically be available to replace that player. We start by comparing the player to average in a variety of venues, then compare our theoretical replacement player to the average player and add the two results together."


        It theoretically is the number of wins above a standard fill in replacement player (who would get some wins himself). So if a player has 10 WAR, he might account for 13ish Wins total himself (don't know the exact number, using this as an example).
        R.I.P. Farmio
        Play the Who am I? game in trivia and you can make this signature line yours for 3 days (baseball signatures only!)

        Go here for a link to all player links! http://www.baseball-fever.com/forum/...player-threads

        Go here for all your 1920's/1930's OF info

        Comment


        • #5
          Correct, Toledo. It's the marginal wins above replacement level. Replacement level is, itself, not zero wins. A team of nothing but replacement level players would win approximately .294 percent of their games, 48 in a 162-game season. If Mike Trout plays 162 games and his production is worth 10.0 WAR, what this is saying is that a team with all replacement players plus Mike Trout is a difference between 48 and 58 wins, not a difference between 0 and 10 wins.
          "It is a simple matter to erect a Hall of Fame, but difficult to select the tenants." -- Ken Smith
          "I am led to suspect that some of the electorate is very dumb." -- Henry P. Edwards
          "You have a Hall of Fame to put people in, not keep people out." -- Brian Kenny
          "There's no such thing as a perfect ballot." -- Jay Jaffe

          Comment


          • #6
            Originally posted by donnie72191 View Post

            I see.. so a 10 WAR player doesnt LITERALLY win his team 10 games in a season by himself correct?
            What do you mean by literally? His presence, in theory, adds 10 wins to the team's total. This is relative to replacement, but it's assumed that if that player isn't on the team, a replacement level player will be instead. You wouldn't have a player for very long who is below replacement level.

            It's not that you can pick out 10 specific games during the season in which Trout made the difference between a win and a loss, but that over the course of the season, the team would be expected to win 10 more games.

            Originally posted by Chadwick View Post
            A team of nothing but replacement level players would win approximately .294 percent of their games, 48 in a 162-game season.
            That has been my understanding. But an article at FG today about projections claims that teams with replacement players only win 43.66 games. I don't understand where he got that from.

            https://blogs.fangraphs.com/a-quick-...-playoff-odds/
            Last edited by Stolensingle; 02-20-2020, 07:58 PM.

            Comment


            • #7
              Originally posted by Stolensingle View Post
              That has been my understanding. But an article at FG today about projections claims that teams with replacement players only win 43.66 games. I don't understand where he got that from.

              https://blogs.fangraphs.com/a-quick-...-playoff-odds/
              I got 48 wins from a Baseball-Reference blog post a few years back. It's possible that has changed, although I thought that BRef and FG were (now) using the same replacement level?

              I have no problem with Ben Clemons' statement - he'd know better, first-hand, than me - so just change my illustration from 48 to 44 games. If my information was incorrect, I'm glad to have it corrected.
              "It is a simple matter to erect a Hall of Fame, but difficult to select the tenants." -- Ken Smith
              "I am led to suspect that some of the electorate is very dumb." -- Henry P. Edwards
              "You have a Hall of Fame to put people in, not keep people out." -- Brian Kenny
              "There's no such thing as a perfect ballot." -- Jay Jaffe

              Comment


              • #8
                Someone else raised that question in the responses to Clemens. He acknowledged the discrepancy, and suggested maybe it was because their summing of individual player's WAR was too high, because they weren't allowing for enough games missed due to injuries.

                Clemens begins with 81 x 30 = 2430 games won. He then adds all the WAR projected for individual players, and subtracting that total from 2430 gives replacement wins. The sum of his projected WAR is 1120, whereas conventionally, 1000 total WAR is assigned to a season. 2430 - 1000 = 1430; divide that by 30 teams, and you get close to 48 wins. So i think the problem is summing the individual WAR of the players. i would have thought that would work out close to 1000--it usually does at the end of a season, i think--but when using projections, there may be other errors that creep in, such as in the injury problem Clemens suggests.

                i would continue to go with 48 wins.

                Comment


                • #9
                  If someone is 10 WAR, it means that the team won 10 more games than it would have won with a "replacement" level, which you can think of as a "bubble" player, or veteran free agents being signed to minor league contracts.

                  Let me know if you have any more questions.
                  Author of THE BOOK -- Playing The Percentages In Baseball

                  Comment


                  • #10
                    The whole concept of giving an individual player 'wins' is misleading and kind of silly. A 10 WAR player is not 'worth' 10 wins. Using their logic, if Mike Trout has Cody Bellinger as his backup, then he is basically worthless.

                    Comment


                    • #11
                      Dear baseball gods...please let me see Tango go at it with Wilshad about statistics at least once in my life. It would allow me to die in peace.
                      1885 1886 1926 1931 1934 1942 1944 1946 1964 1967 1982 2006 2011

                      1887 1888 1928 1930 1943 1968 1985 1987 2004 2013

                      1996 2000 2001 2002 2005 2009 2012 2014 2015


                      The Top 100 Pitchers In MLB History
                      The Top 100 Position Players In MLB History

                      Comment


                      • #12
                        Originally posted by willshad View Post
                        The whole concept of giving an individual player 'wins' is misleading and kind of silly. A 10 WAR player is not 'worth' 10 wins. Using their logic, if Mike Trout has Cody Bellinger as his backup, then he is basically worthless.
                        All humor aside (thanx Bothrops) -- and someone correct me if I'm wrong -- but I think that essentially answers the OP query. No, WAR does not mean a literal connection to dedicated wins like pitchers' Wins do. After all, pitchers have a WAR too, and it's virtually unrelated to their Wins stat.

                        I'm really a tenderfoot in the advanced stats arena, so please be patient. From what I understand, WAR measures approximately how many wins above 110-120 team losses per season a player contributes to his team's (teams') final W/L record.

                        WAR can stand on its own if taken in its proper context. The problem -- well, it's not really a problem -- is that it's cumulative, so people tend to look at it and think "the higher the better," and in a sense, that's true. But I would think a player can have a career-high WAR encapsuled in a single game, if you will. The stat cannot be judged alone, nor without adjacent stats to help better understand what a player was worth over the course of a game, a month, a season, a career.

                        And the more they play, the more that WAR is worth. If someone plays 20 years and ends up with a 0.0 WAR, they couldn't possibly be doing much for their teams, since only a truly poor player could play that long and be a neutral contributor.

                        So, again from what I've gathered, it's important to look at several other stats when checking WAR. It was never intended to be the end all and be all of measuring a player. But there's a reason it's in the first line of every player at BReference.
                        Put it in the books.

                        Comment


                        • #13
                          Originally posted by willshad View Post
                          The whole concept of giving an individual player 'wins' is misleading and kind of silly. A 10 WAR player is not 'worth' 10 wins. Using their logic, if Mike Trout has Cody Bellinger as his backup, then he is basically worthless.
                          What is misleading and silly is suggesting that Bellinger would be a backup to anyone. Come on.

                          Of course WAR, by definition, is relative to replacement. And if someone like Trout goes down for a while, which has in fact happened in each of the past three seasons, then he's replaced by someone who is more or less at replacement level. E..g., last year, Trout played his last game Sept. 7, after which he was out with a foot problem. He was replaced in CF by Brian Goodwin, who had 1.9 fWAR for the season. That's actually about average, not replacement level, but for his career he's been a little below that. In 2017, when Trout missed about 50 games with a broken thumb, he was replaced by Cameron Maybin, who had 1.2 fWAR for the entire season. That is below average, but a little better than replacement.

                          But also keep in mind that players like that are likely to be used by teams not simply to replace an injured player, but as part-time starters on occasion, maybe sharing the load with another player who also isn't quite good enough to start all the time. In this case, they're likely to be platooned, facing only opposite-handed pitchers. So their batting stats, and thus WAR, may be enhanced over what would be the case if they were an everyday player over a longer period of time.

                          You can quibble about how good the player replacing someone like Trout is, but he's going to be fairly close to replacement level. If he weren't, he would probably be starting for the Angels, or for some other team. An average team has about 33 WAR, so an average player, with about 2 WAR, will usually be starting at least much of the time, and an above average player is not going to be sitting on the bench, just waiting for a really good player to go down. You might, of course, have a team with a surfeit of, say, really good outfielders, so the fourth man is maybe a well above average player. But that's probably not the most efficient use of resources in the long-term, so a team in that position is likely to trade to fill a more pressing need. In fact, a team that has only replacement level players to back up every starting position player is probably using its resources with maximum efficiency, because any better backup could be traded for, say, another needed bullpen arm, or perhaps packaged with another starter to upgrade at some position.
                          Last edited by Stolensingle; 03-01-2020, 04:01 PM.

                          Comment


                          • #14
                            Also WAR is related to pythagorean wins (run value) not actual wins.

                            War tries to be context neutral but a 5 war player could be worth only 3 wins or 7 wins if he gets hits in clutch or "unclutch" situations. Clutch is not a sustainable skill and thus "luck" (low year to year correlation) but of course it affects real world wins if you have a year with a bunch of grand slams instead of solos or you have a high RISP wOBA.

                            Those things aren't really skills but they count in the real world standings.
                            I now have my own non commercial blog about training for batspeed and power using my training experience in baseball and track and field.

                            Comment


                            • #15
                              I've always questioned Defensive WAR. How exactly is that even calculated? How can you go back to early years and come back with any type of accuracy? Say for example I'm curious how Dimaggio and Mays compare as center fielders... how are do they get those numbers? 6.8.0

                              Comment

                              Ad Widget

                              Collapse
                              Working...
                              X