Like us on Facebook


Follow us on Twitter





Page 17 of 51 FirstFirst ... 7151617181927 ... LastLast
Results 241 to 255 of 761
  1. #241
    Join Date
    Jan 2007
    Posts
    9,069
    vCash
    1500
    Quote Originally Posted by ReyBrutal View Post
    I agree with you to a certain degree but I think +/- is a bit better of a defensive stat than UZR and I would never use FIP it's too defense reliant to use it to compare pitchers.

    VORP has it's flaws.

    I'm losing confidence in WAR as well.
    -FIP is too defense-reliant? I'm not quite sure I follow.

    -Plus/Minus makes less adjustments to it and the way it calculates run values is decent at best. Both metrics have their flaws but UZR is the lesser of two evils. I prefer to weight it .6 UZR, .2 PMR and .2 Plus/Minus.

    -VORP definitely has its flaws and shouldn't be looked at. RARP is better but still has its issues.

    -Any reason why you're losing confidence in WAR?
    My blog- analysis of the San Francisco Giants, Baseball, and Sabermetrics.

  2. #242
    Join Date
    Jan 2007
    Posts
    9,069
    vCash
    1500
    Quote Originally Posted by ReyBrutal View Post
    Do you guys put any stock into ERA+ and OPS+?

    How about UZR and UZR/150 and +/-?
    ERA+, no. Calculation of it could be improved on and there's no real need to pay attention to ERA, at least on a single-season basis. Over the course of a pitcher's career, sure. OPS+ is (and I'm sure I might get some flack for this) basically useless. You can't add OBP and SLG together because they're not on the same scale. It undervalues OBP by quite a bit.

    Plus/Minus is pretty good. UZR is better, but neither should be taken seriously on a single-season basis (too much noise in the data). UZR/150 is being cited a lot but I don't really see why. There's really no point in prorating defensive numbers in my opinion.
    My blog- analysis of the San Francisco Giants, Baseball, and Sabermetrics.

  3. #243
    Join Date
    Oct 2009
    Posts
    5
    vCash
    1500
    Quote Originally Posted by poodski View Post
    Personally I love WPA. WPA might be my favorite stat out there, but it can be very defense reliant for a pitcher.



    How is FIP defense reliant?

    What dont you like about WAR?
    Excuse me, I confused FIP with something else, I looked at it, I was wrong on that one.

    As for WAR, I was comparing a few players the other day on Fangraphs and player 2 had slightly lower stats overall and a higher WAR, I'll try to find it again and if I do I'll post it here.

  4. #244
    Join Date
    Oct 2009
    Posts
    5
    vCash
    1500
    Quote Originally Posted by C1Bman88 View Post
    ERA+, no. Calculation of it could be improved on and there's no real need to pay attention to ERA, at least on a single-season basis. Over the course of a pitcher's career, sure. OPS+ is (and I'm sure I might get some flack for this) basically useless. You can't add OBP and SLG together because they're not on the same scale. It undervalues OBP by quite a bit.

    Plus/Minus is pretty good. UZR is better, but neither should be taken seriously on a single-season basis (too much noise in the data). UZR/150 is being cited a lot but I don't really see why. There's really no point in prorating defensive numbers in my opinion.
    I disagree with that first part, I'm a firm believer that a pitchers first job is to not allow any runs to score, regardless of how it's done. I know that the "+" stats need a 3 year basis of a stadium so YS and Citi Field are still working out little kinks... still I think that it is a fair measure since it adjusts to league averages.

    I'll have to look up +/- again but there was a reason why I saw it as a better measure. I do see reason to measure defense numerically but I think it will be done better in the coming years. Did you hear about those cameras they installed in the top of every stadium which tracks like everyone on the defender, not sure if they went up yet but I read about it and it sounded interesting.

  5. #245
    Join Date
    Jan 2007
    Posts
    9,069
    vCash
    1500
    Quote Originally Posted by ReyBrutal View Post
    I disagree with that first part, I'm a firm believer that a pitchers first job is to not allow any runs to score, regardless of how it's done. I know that the "+" stats need a 3 year basis of a stadium so YS and Citi Field are still working out little kinks... still I think that it is a fair measure since it adjusts to league averages.

    I'll have to look up +/- again but there was a reason why I saw it as a better measure. I do see reason to measure defense numerically but I think it will be done better in the coming years. Did you hear about those cameras they installed in the top of every stadium which tracks like everyone on the defender, not sure if they went up yet but I read about it and it sounded interesting.
    Well, there are certain things a pitcher has no control over, and I don't think they should be held responsible for good or bad defense behind them. There's too much fluctuation year to year when it comes to defensive support, so I don't trust ERA on a single-season basis. ERA+ is pretty good for multiyear sampling though. OPS+ is just plain misleading. As much as I love B-Ref, I wish Sean would fix a few things.

    And yes, I've heard of this new system- it'll fix a lot of the issues with Plus/Minus and UZR. I actually posted a thread on it a few months ago.
    My blog- analysis of the San Francisco Giants, Baseball, and Sabermetrics.

  6. #246
    Join Date
    Apr 2008
    Location
    New York
    Posts
    32,584
    vCash
    1500
    Hey , I hope someone wouldn't mind answering.

    I know tRA accounts for batted ball types in its algorithm but does it also account for "clutchiness" ?

    As in, surrendering three consecutive line drives would probably lead to a run but three line drives in different innings probably wouldn't.

    Whoever answers, thank you for your time.

  7. #247
    Join Date
    Jul 2009
    Posts
    57
    vCash
    1500
    No, it doesn't. All of the metrics of the DIPS/FIP/tRA family assume a random distribution of events (essentially).

  8. #248
    Join Date
    Jan 2007
    Posts
    9,069
    vCash
    1500
    Quote Originally Posted by Hustla23 View Post
    I know tRA accounts for batted ball types in its algorithm but does it also account for "clutchiness" ?
    Sounds to me like you're looking for something along the lines of a "value added" approach, but this doesn't account for batted ball type- the only thing that it measures is the total run expectancy based on the distribution of hits allowed (singles, doubles with a runner on first, etc.) so it's not defense-independent.

    You might be interested in RE24 (run expectancy by the 24 base-out states) for pitchers. Again, it's not defense-independent, but it'll give you a pretty good idea of how pitchers performed based on the distribution of hits they allowed.
    My blog- analysis of the San Francisco Giants, Baseball, and Sabermetrics.

  9. #249
    Join Date
    Apr 2008
    Location
    New York
    Posts
    32,584
    vCash
    1500
    Quote Originally Posted by C1Bman88 View Post
    Sounds to me like you're looking for something along the lines of a "value added" approach, but this doesn't account for batted ball type- the only thing that it measures is the total run expectancy based on the distribution of hits allowed (singles, doubles with a runner on first, etc.) so it's not defense-independent.

    You might be interested in RE24 (run expectancy by the 24 base-out states) for pitchers. Again, it's not defense-independent, but it'll give you a pretty good idea of how pitchers performed based on the distribution of hits they allowed.
    Thank you.

    I hope they do introduce something that encompasses batted ball types with its "values added."

  10. #250
    Join Date
    Apr 2008
    Location
    New York
    Posts
    32,584
    vCash
    1500
    Is Speed considered when calculating WAR?

    Or is that factored into RAR somehow?

    I get the feeling it isn't which is a shame.

  11. #251
    Join Date
    Jan 2007
    Posts
    9,069
    vCash
    1500
    Quote Originally Posted by Hustla23 View Post
    Is Speed considered when calculating WAR?

    Or is that factored into RAR somehow?

    I get the feeling it isn't which is a shame.
    RAR and WAR are the same exact thing, just shown in different format. RAR is "Runs Above Replacement," and WAR are "Wins Above Replacement." 10 runs equals one win, so when you see RAR, just divide by 10 to get WAR.

    "Speed" is a largely subjective thing and even the concept of "Speed Scores" don't tell you much. Speed doesn't necessarily translate into good baserunning. But if you're referring to SB/CS, that's included in WAR. But it doesn't add much, to be honest.

    Baserunning in general has a little impact on a player's overall contribution- you're looking at about +/- 5 runs on the basepaths. The best and worst baserunners add or subtract about 10 runs to their team.
    My blog- analysis of the San Francisco Giants, Baseball, and Sabermetrics.

  12. #252
    Join Date
    Apr 2008
    Location
    New York
    Posts
    32,584
    vCash
    1500
    Quote Originally Posted by C1Bman88 View Post
    RAR and WAR are the same exact thing, just shown in different format. RAR is "Runs Above Replacement," and WAR are "Wins Above Replacement." 10 runs equals one win, so when you see RAR, just divide by 10 to get WAR.

    "Speed" is a largely subjective thing and even the concept of "Speed Scores" don't tell you much. Speed doesn't necessarily translate into good baserunning. But if you're referring to SB/CS, that's included in WAR. But it doesn't add much, to be honest.

    Baserunning in general has a little impact on a player's overall contribution- you're looking at about +/- 5 runs on the basepaths. The best and worst baserunners add or subtract about 10 runs to their team.
    Is that a subjective conclusion?

    Or does base running really not factor much into runs produced?

    I'd consider the ability to steal bases a pretty big factor in scoring runs IMO

  13. #253
    Join Date
    Jan 2007
    Posts
    9,069
    vCash
    1500
    Quote Originally Posted by Hustla23 View Post
    Is that a subjective conclusion?

    Or does base running really not factor much into runs produced?

    I'd consider the ability to steal bases a pretty big factor in scoring runs IMO
    No, not a subjective conclusion at all. You can look at the best and worst baserunners here. Michael Bourn was the only player to be worth over 1 win in baserunning runs. Only 5 players from 2006-2009 have amassed over a win on the basepaths.

    As for the value of a steal, it's relatively straightforward- just use a run expectancy chart.

    Say we have a man on first base—the average run expectancy of a runner on first is 0.552. If he steals second base, the run expectancy increases to about 0.726 runs. The difference between the two, 0.174, is the run value of a stolen base. If he’s caught stealing, the difference is between having no runners on base (0.302) and having a man on second (0.726), or -0.249 runs. If we do the same thing for third base, we find that the run value of stealing third is worth 0.211 runs and a caught stealing is -0.635 runs.

    If you find the average value between steals and caught stealing of second and third base, you'll find that the run value of a stolen base is about ((0.174+0.211)/2) = 0.19 runs added, and a caught stealing would be worth about ((-0.249+-0.635)/2) = -0.44 runs.

    So, Jacoby Ellsbury, who stole 70 bases in 2009, added 13.3 runs via steals, but cost his team -5.3 runs by being caught stealing 12 times. So that means he added a net gain of around +8 runs added. Baseball Prospectus' EQBRR has Ellsbury adding +7 runs, but their methodology is MUCH more intricately detailed, and includes things like park effects.
    My blog- analysis of the San Francisco Giants, Baseball, and Sabermetrics.

  14. #254
    Join Date
    Apr 2008
    Location
    New York
    Posts
    32,584
    vCash
    1500
    Quote Originally Posted by C1Bman88 View Post
    No, not a subjective conclusion at all. You can look at the best and worst baserunners here. Michael Bourn was the only player to be worth over 1 win in baserunning runs. Only 5 players from 2006-2009 have amassed over a win on the basepaths.

    As for the value of a steal, it's relatively straightforward- just use a run expectancy chart.

    Say we have a man on first base—the average run expectancy of a runner on first is 0.552. If he steals second base, the run expectancy increases to about 0.726 runs. The difference between the two, 0.174, is the run value of a stolen base. If he’s caught stealing, the difference is between having no runners on base (0.302) and having a man on second (0.726), or -0.249 runs. If we do the same thing for third base, we find that the run value of stealing third is worth 0.211 runs and a caught stealing is -0.635 runs.

    If you find the average value between steals and caught stealing of second and third base, you'll find that the run value of a stolen base is about ((0.174+0.211)/2) = 0.19 runs added, and a caught stealing would be worth about ((-0.249+-0.635)/2) = -0.44 runs.

    So, Jacoby Ellsbury, who stole 70 bases in 2009, added 13.3 runs via steals, but cost his team -5.3 runs by being caught stealing 12 times. So that means he added a net gain of around +8 runs added. Baseball Prospectus' EQBRR has Ellsbury adding +7 runs, but their methodology is MUCH more intricately detailed, and includes things like park effects.
    Ahh, I see.

    Thank you very much for the information. It was very well described and informative.

    But you said that SB/CS doesn't affect factor in much.

    But one win would be a significant number wouldn't it? Considering, it is one win based entirely on speed.

  15. #255
    Join Date
    Jan 2007
    Posts
    9,069
    vCash
    1500
    Quote Originally Posted by Hustla23 View Post
    Ahh, I see.

    Thank you very much for the information. It was very well described and informative.

    But you said that SB/CS doesn't affect factor in much.

    But one win would be a significant number wouldn't it? Considering, it is one win based entirely on speed.
    Ah, but baserunning ability isn't reliant on speed. The best baserunners tend to be speed demons, but it's ultimately the player's instincts and their ability to capitalize on an opportunity that makes a runner good. Ryan Zimmerman, for example, isn't fast- but he's smart, so he helped add 5 runs to his team. Kurt Suzuki isn't fast either, but he added 4 runs to the Athletics this past season by virtue of his smart baserunning.

    One win is significant, but the lack of players to reach that mark is extraordinarily low, and the amount of players that reach even half of that mark is also low. It is significant for certain players, yes, but for most, not very.
    My blog- analysis of the San Francisco Giants, Baseball, and Sabermetrics.

Page 17 of 51 FirstFirst ... 7151617181927 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •