FanGraphs Baseball

Comments

RSS feed for comments on this post.

  1. Thanks for taking the time to put this together.

    Comment by Kroot — May 2, 2011 @ 3:14 pm

  2. Nice job, I enjoyed it. Mrs. Polanco did too, I’m sure.

    Comment by JR — May 2, 2011 @ 3:17 pm

  3. WHY DO YOU HATE MY TEAM?!?!?!?!111!?!1/oNE

    Comment by Carson Cistulli — May 2, 2011 @ 3:19 pm

  4. No offense, but are you fucking kidding me? Indians at 20?!?! They already have a 10 game lead on the 16th place Twins and 7.5 game lead on the 17th place Tigers. Granted I realize that future expections are at play here, but give the Tribe at least a measure of respect.

    Comment by The Rhino — May 2, 2011 @ 3:21 pm

  5. You didn’t read the methodology did you?

    Comment by Blob — May 2, 2011 @ 3:25 pm

  6. Mr. The Rhino, please see Mr. Cistulli’s post above, preemptively making fun of you.

    Comment by Santos — May 2, 2011 @ 3:32 pm

  7. Pointing to an ESPN-NY article as signs that “those that cover the team think the sky is falling” is ridiculous. Everyone that writes for ESPN-NY is decidedly and unabashedly anti-yankee. It’s very obvious.

    Comment by descender — May 2, 2011 @ 3:36 pm

  8. Great stuff.

    Comment by fantasystud1305 — May 2, 2011 @ 3:39 pm

  9. The Indians at 20 threw me off too until I read the methodology. That said, the FANS % is pretty worthless at this point, isn’t it? The Tribe are clearly better than anyone anticipated for a few reasons that don’t seem to be changing anytime soon.

    Comment by reidymr — May 2, 2011 @ 3:43 pm

  10. Neat rankings, do you guys have a formula for the WAR%/FAN% Ratio or are you just using judgment? I’d be curious to see what you guys came up with for that.

    Comment by sparbz — May 2, 2011 @ 3:44 pm

  11. Interesting to see the disparity between reality and perception for the Royals.

    Comment by Daniel — May 2, 2011 @ 3:44 pm

  12. I guess in some measure of support for the The Rhino, I like the idea of these rankings, but the methodology is very flawed. I like balancing previous success with realistic future expectations, but FAN% is a very poor tool for these purposes. In the case of the Twins, which The Rhino was quick to observe as overrated, the .537 FAN% reflects a team that doesn’t even exist on the feild anymore. The Astros FAN% at .370 (.050 lower than any other team in the feild) is absurd, and they will be #30 by this methodology for the entire season. A better system would use ZiPS(R)%, Marcel%, Tango% to balance FAN% (if not completely replace it).

    Comment by Terence — May 2, 2011 @ 3:44 pm

  13. I agree that the FAN% should be relegated to a smaller factor. What it’s doing to Cleveland will look laughable later in the season by this method, and seems like gross overcompensation.

    Comment by hairball — May 2, 2011 @ 3:47 pm

  14. Using pre-season Fan projections for current team power rankings is absurd.

    Comment by Nik — May 2, 2011 @ 4:00 pm

  15. I love the way this sets up this week’s big Pirates – Astros series as a head to head clash for 29th place.

    Comment by maguro — May 2, 2011 @ 4:00 pm

  16. So is basing power rankings off of 25 games.

    Comment by descender — May 2, 2011 @ 4:01 pm

  17. I really like this idea overall, but I agree with others that Fan% is probably not the best way to balance out WAR%. Perhaps Zips (ROS) or some equivalent would be preferable.

    Comment by VivaAyala — May 2, 2011 @ 4:08 pm

  18. That’s exactly why Power Rankings are usually unreadable at this point, they don’t seem to do what they’re supposed to do, tell a story beyond the W-L standings, instead they just mimic a small sampling of games. Swydan’s work is much more interesting than these.

    Comment by Kroot — May 2, 2011 @ 4:17 pm

  19. At this early in the season, I think pre-season projections are a lot better predictors of what’s going to happen than W-L records are.

    We’re only like 15% of the way into the season.

    Comment by Yirmiyahu — May 2, 2011 @ 4:19 pm

  20. I think FAN percentage is great for what this is.

    Comment by Kroot — May 2, 2011 @ 4:19 pm

  21. “Saturday night, the BABIP Gods relented long enough to allow Chris Iannetta to slam a three-run homer and push the Rockies past the Pirates.”

    Since when did the BABIP Gods preside over home runs? That’s the realm of the HR/FB Gods, isn’t it? ;-)

    Comment by DavidJ — May 2, 2011 @ 4:21 pm

  22. I like the power rankings and have no problem with using the fans expectations of the teams.
    I’m sorry I didn’t maintain the status quo of complaining about something.

    Comment by Norm — May 2, 2011 @ 4:25 pm

  23. Why? You shouldn’t really change your expectations based off of 25 or so games.

    Say, you originally expected the Red Sox to win 100 games (.617). But they’ve gone 12-15. Then you should expect them to go about 83-52 (.615) going forward. That’d add up to 95 wins.

    There’s just as much talent as there was on that team on opening day. Opposite goes for the Royals.

    Comment by Yirmiyahu — May 2, 2011 @ 4:27 pm

  24. Nice to see the Yankees on top!

    Comment by William — May 2, 2011 @ 4:39 pm

  25. I think the best way to do this would be to use a sliding scale system. a 50/50 split between WAR-based performance metrics and pre-season opinion just does not work. A simple way to fix it would be to adjust the ratios as the season goes on. After game 1, it should be closer to 98% fan projections, 2% performance. After game 161, it should be pretty much the reverse. What the exact formula is, I dont know, but I know that the 50/50 split will not work, especially as we get further into the season.

    Comment by Nik — May 2, 2011 @ 5:02 pm

  26. What about after 100 games? 161 games? Still think Fan opinion should count then?

    Comment by Nik — May 2, 2011 @ 5:05 pm

  27. PS, some credit should also be given for overall record and especially the team’s performance during the most recent time period. These current rankings wont have much credibility if in the middle of August a team that has accumulated a bunch of WAR but then gets hit by some major injuries shouldn’t be floating near the top of the rankings for weeks on end.

    Comment by Nik — May 2, 2011 @ 5:11 pm

  28. Surprised to see Minnesota so high up the rankings.

    Comment by Llewdor — May 2, 2011 @ 5:12 pm

  29. You didn’t actually read the introduction, did you?

    “A team’s WAR% is given weighting equal to the percentage of 2011 season that the team has played. Right now, teams fall in the 15-18% range, as they have played between 26 and 29 games. As the season progresses — and we learn more about each team’s true level of talent — the WAR% will receive a greater weight with the FAN% receiving less weight. “

    Comment by Yirmiyahu — May 2, 2011 @ 5:13 pm

  30. Ethier’s streak is still amazing, WAR aside.

    Comment by CircleChange11 — May 2, 2011 @ 5:27 pm

  31. Astros can sweep the Pirates and still be #30. They’re giving up .068% in the FAN% which next week will still count three times as much as as the WAR% does. That means the Astros need to be >.200 better in WAR% this time next week to claim #29. Not gonna happen.

    Comment by Terence — May 2, 2011 @ 5:27 pm

  32. The BABIP gods reach extends only as far as the outfield fence. Iannetta clearly had enough of playing their game.

    Comment by Obligatory Sox Fan — May 2, 2011 @ 5:35 pm

  33. Nah, not really.

    Comment by Jim — May 2, 2011 @ 5:45 pm

  34. Maybe you should include ALL MLB teams next time, you know, like the Giants.

    Comment by Um — May 2, 2011 @ 5:58 pm

  35. Um, check again?

    Comment by Kroot — May 2, 2011 @ 6:01 pm

  36. To help alleviate some of the concerns people have about the weighting… A square root rule could be used to assign the weighting, or credibility, of the games already played. For example a team who has played 26 games would receive 40% from WAR (=sqrt(26/162)) and 60% from preseason fan rankings. This accelerates the weight given to performance in the season without going bonkers. Love the idea of this. I actually like the current weighting structure more, just trying to throw out some alternatives to help others in the comment section who want the Indians higher.

    Comment by glassSheets — May 2, 2011 @ 6:02 pm

  37. I just wanted to say Swaydens work here’s been really great and I’m glad to see him taking on a regular feature like this. Thanks for getting us writers like this FGraphs.

    Comment by Kroot — May 2, 2011 @ 6:03 pm

  38. ! Swydan

    Comment by Kroot — May 2, 2011 @ 6:04 pm

  39. excuse me for the typo

    Comment by Kroot — May 2, 2011 @ 6:05 pm

  40. I think the fan opinion (especially preseason fan opinion) should be removed and replaced with a more objective talent measure. Perhaps a composite projection of WAR from all starting players on the roster would help (using any of the various projection systems). It seems like the fan opinion element is included with the intent of factoring in talent, but there are better ways to measure that.

    Comment by OzzieGuillen — May 2, 2011 @ 6:21 pm

  41. I’m fine with using FAN% and WAR%. And the weighting you use has the right idea. But I think you could make it a little bit better. Cleveland’s hot start has to make you think they’re actually a better team than expected, not just that their win total will be higher than expected based on the wins already on the board. Rany Jazayerli did some research on this back in 2003. I’d suggest using his findings to tweak the formula. (http://www.baseballprospectus.com/article.php?articleid=1818)

    Comment by Brett — May 2, 2011 @ 6:37 pm

  42. Honestly this is horrible for a SABR site. Fan ratings? Ugh are you kidding me. Look at Beyond the Box Score for a site that using stats for a power ranking.

    http://www.beyondtheboxscore.com/2011/5/1/2147694/april-2011-bsrized-standings

    Comment by FlimtotheFlam — May 2, 2011 @ 6:49 pm

  43. The weirdest one to me is the White Sox. I’m surprised there was so much hate for them from the fans. With their pitching and the addition of Dunn I thought they’d be pretty decent, certainly better than the Tigers and probably better than the Twins, I guess the fans knew something I did not.

    I don’t really have a problem with the Indians at 20th, it is early after all, and the only team above them that I’d say is clearly worse than them is the current Twins. The Twins with Mauer, Morneau and a functioning Liriano are a good team, without they are a bad team. It would be nice if there way a way to account for major injuries/whatever’s wrong with Liriano in the rankings, because that’s what’s driving the difference between preseason ranking and current performance in a lot of cases.

    Comment by isavage30 — May 2, 2011 @ 7:19 pm

  44. The assumption that the fan ratings are some accurate barometer of true talent doesn’t really hold up. I like the thought, but that undermines the whole rankings.

    Comment by suicide squeeze — May 2, 2011 @ 7:20 pm

  45. I think this is a really really good piece. I like the methodology. Phasing in the current season and phasing out the preseason in proportion to games played is simple but very strong. The only quibble I can see is whether the FAN% is the best preseason metric to be using, but glancing at the results nothing seems out of whack so it seems fine to me. Replacing it with a preseason projection from another source probably wouldn’t change things too much.

    I enjoyed the links too. Good work.

    Comment by BillWallace — May 2, 2011 @ 7:24 pm

  46. Since Baltimore has a higher WAR than Seattle or the White Sox and has played fewer games, how do they have a lower WAR percentage?

    Comment by Larry — May 2, 2011 @ 7:27 pm

  47. I agree that I like the theory here, but that Fan % probably isn’t the best choice. Can’t we just use some sort of preseason Zips/pecota/WAR #s instead? Then do something like:

    ((Current WP * Games played) + (Preseason projected WP * Games remaining))/162

    Comment by Everett — May 2, 2011 @ 7:51 pm

  48. it’s still a feat

    Comment by bender — May 2, 2011 @ 8:22 pm

  49. Run differential could also play a part, thus giving 2 parts actual performance, 1 part Fan/Projections.

    What has happened on the field is more important than what was “supposed” to happen … even in April.

    Comment by CircleChange11 — May 2, 2011 @ 9:06 pm

  50. Like it. I would agree that a projection system might be better than Fan’s, if only so people don’t see it as subjective.

    Comment by Ben Hall — May 2, 2011 @ 9:12 pm

  51. Maybe it’s my beer talking, but I think this is bloody brilliant.

    Comment by My echo and bunnymen — May 2, 2011 @ 9:25 pm

  52. There have been something like eight 30-game streaks since 2000, including one by Willy Taveras, whose presence on any list of ‘offensive greatness’ calls into question the second word in that label.

    Comment by MBD — May 2, 2011 @ 9:36 pm

  53. Doh! Good catch, thank you.

    Comment by Paul Swydan — May 2, 2011 @ 10:40 pm

  54. Yep – this is essentially true. The Fan’s Standings have a very high correlation to the ZIPS Standings, with most teams within a few wins in either direction. The biggest differences are teams like the White Sox (where the fans had 79 wins but ZIPS had 86) and the Marlins (fans had 85 wins, ZIPS 78), and in both cases, the fans are looking more likely to be right so far.

    Replacing the Fan’s Standings with any of the other projection systems wouldn’t make any substantial difference.

    Comment by Dave Cameron — May 2, 2011 @ 10:46 pm

  55. Sure, you could implement some form of pythag integrated as well. Just something less subjective than a fan poll (even if it is a sophisticated one).

    Comment by Everett — May 2, 2011 @ 10:53 pm

  56. Larry, you also need to account for WAR for pitchers, which in this case is what sinks Baltimore.

    Comment by Paul Swydan — May 2, 2011 @ 10:57 pm

  57. Great work.

    OT: Has anyone ever done any work on measuring player’s fantasy contributions on a week-to-week volatility basis? I feel like almost every year Berkman has these stretches of a month where he’s the best player in the world and then hits like crap for a different month. Knowing which guys are less volatile (like Holliday maybe) would be an advantage in H2H leagues.

    Comment by Travis — May 3, 2011 @ 1:01 am

  58. The question about the fan rankings though has to be this: Were they made arbitrarily and based mostly on last year’s records? I’m not saying the Tribe are going to the post-season, but isn’t it possible that LaPorta is starting to come through? That Sizemore and Hafner finally feel more like their old selves? That Masterson has figured out how to deal with lefties more effectively? Sure, guys like Hannahan aren’t likely to continue at their present pace, but I think that too many people were unnecessarily down on the Indians for trading star players for prospects that really needed to be allowed the opportunity to learn at the MLB level and develop, something that people seemed to have little patience for.

    Toss in top prospects coming up like Phelps, Chisenhall, Kipnis, White, and the wealth of depth at AAA Columbus and the Indians may very well have a shot at the division, especially given the Twins, ChiSox and Tigers recent lack of performance. I like the idea of splitting the fan ranks evenly with zips-ros, marcel, and oliver projections (albeit none of them were very high on the Indians either).

    Comment by Chris — May 3, 2011 @ 1:02 am

  59. No power ranking really makes that much sense. I saw a ranking recently that had the M’s at 10, which makes absolutely no sense. (26 is probably also way too low – I’d split the difference, and then add one or two to counter my fannish instincts.)

    Comment by bookbook — May 3, 2011 @ 1:33 am

  60. I disagree with using FAN%. People will naturally have bias towards their team, which obscures the statistic towards large market teams.

    Comment by TheGrandSlamwich — May 3, 2011 @ 1:37 am

  61. This is very easy to account for, and we do. Again, there’s no real difference between using the Fan’s Standings and something like ZIPS.

    Comment by Dave Cameron — May 3, 2011 @ 1:55 am

  62. Rankings are for pure entertainment, so it doesn’t bother me if Team A is lower/higher than I think it should be or the methodology is suspect. The write-ups are hilarious and informative, so good job Swydan!

    Comment by Krog — May 3, 2011 @ 2:03 am

  63. You must always remember the Juan Pierre factor – especially if he is being handed a leadoff spot.

    That’s probably what held the Sox back…

    Comment by JoeyO — May 3, 2011 @ 2:04 am

  64. I bet your favorite band is the Beatles and you think Citizen Kane is the greatest movie ever made.

    Comment by MGP — May 3, 2011 @ 4:25 am

  65. in reply to nik:

    “As the season progresses — and we learn more about each team’s true level of talent — the WAR% will receive a greater weight with the FAN% receiving less weight.”

    please read the article.

    Comment by Reuben — May 3, 2011 @ 4:29 am

  66. Did Carson just prove preconception?

    Comment by Jon — May 3, 2011 @ 4:54 am

  67. It’s gotta be that oft-whined-about “West Coast bias!”

    Comment by chuckb — May 3, 2011 @ 9:39 am

  68. Team WAR and run differential should correlate pretty well over long enough time spans. You can have one game where a team manages to strand 17 baserunners and lose despite putting up a decent offensive WAR but that’s not going to happen much.

    Comment by don — May 3, 2011 @ 10:02 am

  69. And you don’t?

    Comment by Yaz — May 3, 2011 @ 10:30 am

  70. Thanks. I knew I was missing something.

    Comment by Larry — May 3, 2011 @ 10:45 am

  71. … or the fact that ESPN is decidedly pro-red sox…

    Comment by descender — May 3, 2011 @ 1:18 pm

  72. Thank you for clarifying that for me. :)

    Comment by TheGrandSlamwich — May 3, 2011 @ 1:29 pm

  73. These rankings are a joke. For a site such as Fangraphs to use Fans Opinion as a data point is ridiculous; no wonder the Yank-me’s and Red Sox are #1 and 2, maybe because they have the highest fan opinion? You’re better than this. And yes I am an Indians fan and there’s no way the Tigers are better than them after being swept by the Indians.

    Do you guys even watch the games? The Tigers, Twins, and Padres have all been atrocious while the Indians have been killing everyone (Best run differential and top 3 in WAR, hurr hurrr they must be the 20th best team durrrr).

    It’s articles like this that give stat-geeks a bad name when obviously you haven’t even watched any of the games and are basing your opinion solely on a valuable piece of info (WAR) and then COMPLETELY discounting it by weighing Fan Opinion (whatever the hell that is) much greater in the calculation.

    Comment by Jeff — May 3, 2011 @ 2:19 pm

  74. Guess what – if we ran it with ZIPS replacing the Fan’s Standings, the Indians would fare worse.

    Comment by Dave Cameron — May 3, 2011 @ 2:59 pm

  75. Guess what, the season has started. How bout you do away with the pre-season projection BS and focus on, I don’t know, the actual games being played? Best record in the league. #1 in run differential. #3 in WAR.

    But you guys are right. They’re worse than the Padres and the Twins whose combined winning percentage is .357.

    Comment by Jeff — May 3, 2011 @ 3:31 pm

  76. I wonder how many weeks of 0.673 baseball the Indians would have to play to crack the top 2 or 3 in the list?

    Comment by Xeifrank — May 3, 2011 @ 3:38 pm

  77. Also, no mention of the fact that the Indians went from having the 3rd worst defense last year to the best this year based on UZR/150. Replacing Jhonny Peralta and Luis Valbuena (2 of the worst fielders in the history of baseball) with Hannahan and Orlando Cabrera will do that for you.

    But don’t let informative stats like that get in the way of Fan opinion %. Wouldn’t want to lose that valuable information on people GUESSING which team is better in the preseason.

    Comment by Jeff — May 3, 2011 @ 3:39 pm

  78. If you really don’t understand the need to regress a month’s worth of data, I don’t know what to tell you.

    Comment by Dave Cameron — May 3, 2011 @ 5:07 pm

  79. the onion needs to release a new t-shirt that says “the news media entity in question is hopelessly biased against my hometown sports team.”

    Comment by tom s. — May 3, 2011 @ 5:35 pm

  80. I’m partial to Shawshank Redemption. And the Beatles are the greatest band in modern music.

    Comment by William — May 3, 2011 @ 5:36 pm

  81. This is a reply to the Dave Cameron post at 5:07 PM that is above but didn’t have a reply tag at the bottom of his post.

    If you don’t understand why people’s guesses on Win-Loss record, especially guesses made before the season started where no analysis can be done, have no statistical relevance, I don’t know what to tell you. Who cares if 50,000 nobodies think the Yankees are going to win 100 games and the Indians are going to win 60? Doesn’t provide one bit of information.

    You’re a statistical analysis website and you’re basing this ranking (statistic) on people’s opinions. That fly’s in the face of everything that statistical analysis is; how do you not see the insanity in that?

    Comment by Jeff — May 3, 2011 @ 6:41 pm

  82. http://en.wikipedia.org/wiki/The_Wisdom_of_Crowds

    Comment by Dave Cameron — May 3, 2011 @ 7:11 pm

  83. The overall principle of wisdom of crowds is fine. However, what I’d really like to know is what the correlation is. I’d feel pretty strongly that different fields would produce different levels of correlation, and even within baseball I imagine the correlation would vary depending on what is asked to be predicted. Do we have enough data to say that wisdom of the crowds is going to provide better correlation than say ZIPS or pecota or some such? If so, go for it. If not, and ZIPS are readily available, it’d make sense to use those since we can run the #s for correlation, until we get evidence that suggests that wisdom of the crowds is in fact as good or better than ZIPS (or CHONE, or Marcel, whatever).

    Comment by Everett — May 3, 2011 @ 7:46 pm

  84. Very cool idea for rankings, certainly preferable to the standard power ranking, wherein they list the teams in order of records, with few differences.

    Any thought about some kind of nonlinear function for the weights? I feel like a linear function underweights the WAR% – for example, I feel like halfway through the year, WAR% should be something closer to ~75% than 50%.

    Comment by Chris — May 3, 2011 @ 9:24 pm

  85. ZIPS and the Fan’s Standings are almost exactly the same.

    Comment by Dave Cameron — May 3, 2011 @ 10:08 pm

  86. No analysis can be done before the season starts? Does that mean that the players’ histories tell us nothing about their abilities and their likely performance in the upcoming season? You should watch more games and see if you notice any trends from one season to the next. Please do report back on what you find.

    Comment by MBD — May 4, 2011 @ 12:08 am

  87. Jeff Francouer is white and he gives a shit. As any NBA fan can attest, that combo plays well in middle America.

    Comment by Mr. Thell — May 6, 2011 @ 12:45 pm

Leave a comment

Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>


Close this window.

0.230 Powered by WordPress