FanGraphs Baseball


RSS feed for comments on this post.

  1. The metrics don’t see Fowler as the defensive wizard that he was expected to be…

    Just another reason to throw out the defensive metrics. Anyone who watched the guy play center field on a daily basis knows that he is far better than an average fielder, and borderline top shelf defender.

    Comment by David Martin — November 8, 2010 @ 4:04 pm

  2. clearly not the right place for this, but hopefully someone sees it and addresses it. WAR is the wins above replacement, where a team of replacement players will win 25 games/year. According to this site’s stats, the 2010 mets were cumulatively +16.8 WAR, for a projected total of 41.8 wins. Since this isn’t even close to what happened, what am I missing?

    (by the way, this same analysis holds true for any team – the yankees, for instance, were +32.7, for a total of 57.7 wins)

    Based on actual wins, I’d have to assume (using the mets numbers) that a replacement team would win 62 or so games, although using the brewers numbers (+29.1, 77 wins) you’d assume a replacement level competence of 48 wins and using the diamondbacks (24.7 WAR, 65 wins) you’d assume 40 wins as a baseline.

    Does WAR actually predict team success, and if not, how is it actually measuring wins?

    Comment by miffleball — November 8, 2010 @ 4:09 pm

  3. The “prediction” issue is a bigger one than can be addressed here in comments. I’ll say that I do think that it correlates fairly well.

    I do want to correct one thing — the WAR model used here sees a replacement level team as around 49- 50 wins, not 25.

    Comment by Matt Klaassen — November 8, 2010 @ 4:15 pm

  4. I think in a different article, they said replacement level for an NL team is around 48 wins.

    Comment by Mike — November 8, 2010 @ 4:16 pm

  5. @Miffleball – the discrepancy is due to the fact that a replacement level team would not win 0 games in one season. If you took a team of 25 AAAA players, they would win some games over the season. It’s that many games + team WAR that should approximate the actual win total.

    Comment by edb11235813 — November 8, 2010 @ 4:17 pm

  6. This is why I qualified what I wrote with “so far,” as even the staunchest defenders of defensive metrics would acknowledge that the sample for Fowler isn’t very telling with regard to his true talent yet.

    Comment by Matt Klaassen — November 8, 2010 @ 4:17 pm

  7. Maybe playing in Coors makes his defense look worse than it really is.

    Comment by Mike — November 8, 2010 @ 4:17 pm

  8. Throw them out based on one outfielder in a unique park.

    Comment by kbertling353 — November 8, 2010 @ 4:17 pm

  9. Based on the extremely large sample of the two games I watched (LIVEz!?!?!?!11?!/1a?1!) at Coors Field, I gotta say I was unimpressed. So I’m throwing out your throw out…… That’s right.

    Comment by My echo and bunnymen — November 8, 2010 @ 4:31 pm

  10. Also note that that’s only offensive wins not pitching war

    Comment by bender — November 8, 2010 @ 4:57 pm

  11. You are missing the fact that you are only including Batting/Fielding runs in your WAR. Add pitching runs and you will get the correct WAR totals for teams.

    Comment by Jimmy — November 8, 2010 @ 5:03 pm

  12. “He’s a real test case.”

    I believe the term you were looking for was “head case.”

    Comment by B N — November 8, 2010 @ 5:19 pm

  13. This is enough to make a Mariner fan wonder who the “Mariner-iest” non-Mariner was in 2010 . . .

    Comment by Choo — November 8, 2010 @ 7:19 pm

  14. Really wish people would quit implying that Matsui is an upgrade over Jack Cust and the A’s should sign him. If he goes to Oakland, they can kiss goodbye any playoff hopes they had. Let him rot in Seattle or KC.

    Comment by PL — November 8, 2010 @ 10:38 pm

  15. You should probably read before responding, at least most of the words.

    Comment by matt — November 8, 2010 @ 11:46 pm

  16. I think 2B and 3B are the most ‘average’ positions on the defensive spectrum. (Although I think of CF & 2B as worth 5 runs above average, while 3B is exactly average).

    Either way your list is 1 CF, 1 C, and three sluggers. Who are the most average players w/o the position component?

    Comment by Trev — November 9, 2010 @ 7:47 am

  17. I think cust may be 1 win better than matsui, 3 WAR vs. 2 WAR, but is that such a huge difference? enough to ‘kiss playoff hopes goodbye’? fact is, it’s pretty hard to be a superstar at DH, given the positional adj.

    Comment by brendan — November 9, 2010 @ 6:33 pm

  18. Matt, you totally hijacked this idea from me. I wrote an eerily similar article about two months ago and did a follow-up, end-of-season version as well. Not that I have enough hubris to genuinely believe you stole it from me, but I will puff my chest out a little that I did it first. Alas, my readership pales in comparison to that of fangraphs. Have a look, you might enjoy it. The link is to the season ending version which also has an embedded link to the original.

    Comment by Jason F — November 9, 2010 @ 8:20 pm

  19. No problem… except if you click through the links, you’ll see that I did the same sort of post a year ago, and did one even earlier than that for a now-defunct site.

    In any case, great minds think alike. As for OUR minds… well… who knows?

    Comment by Matt Klaassen — November 9, 2010 @ 10:21 pm

  20. Yeah, I don’t know if so-called great minds explore who the most average players are, but, for now, I’ll settle for like-minded. Keep up the good work.

    Comment by Jason F — November 9, 2010 @ 11:02 pm

Sorry, the comment form is closed at this time.

Close this window.

0.400 Powered by WordPress