The ESPN player rankings

Dave · June 11, 2007 at 11:16 am · Filed Under Mariners 

If you look at ESPN.com’s front page right now, there’s a story about a player ranking formula that they’ve created.

It’s horrible, it’s useless, and you shouldn’t care. It’s based on bad premises, produces bad results, and is a collossal waste of time. ESPN is horrible at this kind of stuff – a few years ago, they shoved Productive Outs down our throats and tried to convince everyone that they’d created some fantastic new statistic.

ESPN does some things well, but besides a select few guys they’ve hired (John Hollinger, for one), they are a disaster when it comes to any kind of serious analysis. Just ignore the player rater, because it isn’t worth your time, unless you want to learn why Sean White is better than Jarrod Washburn.

Comments

72 Responses to “The ESPN player rankings”

  1. DoesntCompute on June 11th, 2007 11:26 am

    I looked at their list and said, “Ordonez is better than A-Rod? Huh?”. I then tried to figure out their formula and went, “Huh?”.

  2. Chris Miller on June 11th, 2007 11:28 am

    Ordonez has been better offensively than A-Rod this year. I feel sorry for all those Tiger fans who are undoubtedly overwhelmed with glee right now when he falls back down to earth (not saying he’ll be bad, but there is just no way he’s this good).

  3. MarinerDan on June 11th, 2007 11:29 am

    ESPN:statistical analysis::USA Today:geopolitical analysis

  4. Carson on June 11th, 2007 11:29 am

    Agreed. Their explanation of why it’s useful is more useless than the actual rankings. I by no means think I am the most intelligent fan out there, but I feel really sorry for all the people who rely on stuff like this and Baseball Tonight to get their analysis.

  5. Chris on June 11th, 2007 11:29 am

    There’s an online chat w/a researcher for this new thing. Good luck getting them to take some criticism questions.

  6. JMB on June 11th, 2007 11:34 am

    They accounted for position — nice — but then lumped all OFers in together. Ichiro gets the same credit for position as Manny Ramirez. Ugh.

  7. RaoulDuke37 on June 11th, 2007 11:35 am

    I came over here to ask if you guys had taken a look at it. I thought the numbers were quite odd myself. So, to find out you think it is horrible, means means I must be learning something.

  8. Spanky on June 11th, 2007 11:35 am

    They do have Vidro as the worst hitter on the team…even below the Kitsap twins (Ellison and Bloomie).

  9. Chris Miller on June 11th, 2007 11:36 am

    Any offensive metric that weights R, RBI, AND team winning percentage (?!), is worthless. Actually it’s less than worthless. Man, actually taking a look at how they weight it makes me sick. It’s more like fantasy rankings, except it weights in a few random peripherals, particularly ones that ESPN would be fond of (ERA, W/L, HR, AVG). I’ll continue my stay away from ESPN.com philosophy. I’ll watch games on ESPN though.

  10. Faradan on June 11th, 2007 11:37 am
  11. carcinogen on June 11th, 2007 11:37 am

    I thought we already had something called VORP? Why not do a story on that?

  12. JMB on June 11th, 2007 11:38 am

    The chat is amusing. There are some valid questions, but then there are people who don’t get it, asking “Why is Torii Hunter on the list but not Andruw Jones?”

  13. Chris Miller on June 11th, 2007 11:39 am

    #10, the second I saw that it dawned on me that Dave was right, it’s not even worth a glance.

  14. Dave on June 11th, 2007 11:40 am

    Oh man, this answer is just awesome:

    “Park effects are not directly applied to adjust the ratings. BIg parks doesn’t seem to hurt the Comeria hitters this year or say Dmitri Young.”

    Park effects apparently don’t exist because a few guys are hitting well in pitcher’s parks.

    Just horrible.

  15. CSG on June 11th, 2007 11:44 am

    I do enjoy that JJ is getting some ESPN front page love, though.

    I love how smug ESPN is about this player ranking system. They act like they’ve broken the mold as far as player evaluation goes. They should go back to doing what they do best: broadcasting every Yankees-Red Sox game.

  16. TaylorD7 on June 11th, 2007 11:46 am

    I tried asking a question; the response was kind of hilarious:

    Taylor (Escondido CA): Did you take park effects into account at all? Of course the Padres hitters are going to rank poorly and the Padres pitchers awesome because of Petco. Or, for example, Ian Kinsler (Texas launching pad) v. Jose Lopez (Safeco death to hitters), have the same OPS, yet your rankings have Kinsler higher?

    SportsNation Jeff Bennett: (2:30 PM ET ) Excellent question. Park effects are not directly applied to adjust the ratings. BIg parks doesn’t seem to hurt the Comeria hitters this year or say Dmitri Young. I have a hard time de-valuing Peavy. He is 3-0 with 1.06 ERA on road this year. I see he is ranked 31st in Win shares today behind many pitchers. That seems a little low.

  17. Otto on June 11th, 2007 11:47 am

    One thing that does stick out is Soriano is #47 on the pitchers list and Horacio is no where to be found.

  18. carcinogen on June 11th, 2007 11:48 am

    Dave…nice “edit” of post 14 :-)

  19. TaylorD7 on June 11th, 2007 11:49 am

    Ian, hit it on the head with this one…again, the response was idiotic:

    Ian, NYC: I don’t understand what your point is behind this list. There are people who have put a heck of a lot of science and research into coming up with formulas like this (Win Shares, VORP, etc), while much of what you have selected here is totally arbitrary (why exactly %10 for BA for eaxmple?), and by pretending this is somehow scientific degrades the whole field of work on this subject. Much of what you are including here has been proven to be no reflection on individual player quality (like saves, wins and RBI to a large extent), not to mention penalizing someone because they play on a bad team. Why should anyone take this list seriously?

    SportsNation Jeff Bennett: (2:43 PM ET ) Ian, I think you hit on something. There is no sucjh thing as the perfect way to evaluate a baseball player. Win Shares and VORP are great, but you can ask the same types of questions about their lists. This system is very fluid and puts players in perspective based on where they rank in the majors vs their peers. Nothing more scientific than that.

  20. Dave on June 11th, 2007 11:50 am

    Right. VORP isn’t perfect, so therefore, anyone can put any numbers they want into a blender and call it science.

    This guy doesn’t deserve any more of our time.

  21. bakomariner on June 11th, 2007 11:51 am

    i think it’s horrible, but it’s just the same as analysis on Sportcenter, baseball tonight, or any team Power Rankings…it’s just to stimulate conversations and not to be taken literally…it’s all entertainment…and it’s obviously working or we wouldn’t be having this post…

  22. Otto on June 11th, 2007 11:57 am

    this is too funny why is this:

    141 Michael Bourn PHI 35 9 .257 10 0 2 11 6 .366 9 0 8.0

    151 Jose Lopez SEA 203 60 .296 28 6 35 87 10 .333 1 1 7.5

    This makes no sense

  23. Max Power on June 11th, 2007 12:06 pm

    Case study in why large corporate organizations tend not to be very good at innovation. I’m sure this started out as a good idea – before higher ups insisted on ERA, Avg, Wins etc into the requirements.

  24. Max Power on June 11th, 2007 12:21 pm

    no one pointed out this one?

    43 Adrian Beltre SEA 3B 7.0
    44 Willie Bloomquist SEA 3B 7.0

  25. eponymous coward on June 11th, 2007 12:22 pm

    Yeah, this is a joke. Like I said in the other thread, singles are accounted for FOUR TIMES in their offense formula, walks and extra base hits show up twice (three times if they are HRs). Um… no.

  26. Evan on June 11th, 2007 12:22 pm

    It’s an index. He’s ranked players based on their position within indices (and mostly useless indices, at that).

    If I were reading a scientific or economic paper that did this I’d probably burn it.

  27. Chris Miller on June 11th, 2007 12:23 pm

    #25, I noticed that XBH are only being counted in the “20% total bases” thing, otherwise they use AVG, Hits, AND HR.

  28. Chris Miller on June 11th, 2007 12:24 pm

    #24, it’s obviously because WFB ranks 1st in grit, hussle, and scrap. Adrian only ranks highly in goofy expressions while swinging at outside sliders, and he’s not even in first place!

  29. Evan on June 11th, 2007 12:24 pm

    Nothing more scientific than that.

    And right there Bennett explains why we shouldn’t care.

  30. eponymous coward on June 11th, 2007 12:28 pm

    It would be relatively easy to adapt their formula to teams and use it to predict team offensive productivity via runs scored versus various other offensive measures (Linear Weights, VORP, Runs Created, and so on).

    I predict any other sabremetric measure would wipe the floor with ESPN’s player ratings…

  31. Gomez on June 11th, 2007 12:34 pm

    I thought it was funny when I saw it this morning. These mean about as much as football’s QB rating. In other words, maybe a halfway decent indicator of whether or not a player’s worth a damn at his position, but not much more, and certainly not a rank indicator of who’s better than who.

  32. David* on June 11th, 2007 12:35 pm

    28:

    Adrian also ranks highly in appealing-of-own-checked-swings [AOOCS].

  33. carcinogen on June 11th, 2007 12:35 pm

    Does this guy actually think that statistics is not science, that it is just all an elaborate guessing game with numbers?

  34. Evan on June 11th, 2007 12:37 pm

    It would be relatively easy to adapt their formula to teams and use it to predict team offensive productivity via runs scored versus various other offensive measures (Linear Weights, VORP, Runs Created, and so on).

    Nate Silver just did some quick work applying Rating back to 1997 to see what it told him. It’s on BP Unfiltered.

  35. Evan on June 11th, 2007 12:38 pm

    Does this guy actually think that statistics is not science, that it is just all an elaborate guessing game with numbers?

    He wouldn’t be the first.

    I know an engineer who doesn’t “believe in statistics”. Of course, he’s an unemployed engineer.

  36. Mat on June 11th, 2007 12:42 pm

    If I were reading a scientific or economic paper that did this I’d probably burn it.

    Not before I jotted down the author’s name so that I wouldn’t waste my time with the rest of his “research.”

    Having given a cursory glance at this rating system, it seems worse than the Elias Sports Bureau rankings that determine what type of free agent a player is. And the Elias Sports Bureau rankings are awful.

  37. Uncle Ted on June 11th, 2007 12:45 pm

    I think this is exactly what you’d get if a bunch of fantasy baseball dorks sat around drunk trying to do baseball statistics A priori.
    “no man, seriously, like saves count at least twices as much as wins, coz like pitching in the 9th inning is like at least 10 times harder and starting pitchers only need 5 innings for a win.”

  38. S-Mac on June 11th, 2007 12:55 pm

    Did Rob Neyer have a hand in any of this?

  39. msb on June 11th, 2007 1:00 pm

    what will be entertaining, though, is just who will start quoting the player rankings as proof of their point.

  40. Manzanillos Cup on June 11th, 2007 1:01 pm

    Ha. This is exactly the kind of stuff I would do with the stats off the back of baseball cards when I was 13 years old. Have these people been living in a cave?

  41. carcinogen on June 11th, 2007 1:03 pm

    39: shall we start a betting pool re: readers/commenters on this blog using Rating as demonstration of their point?

  42. bdunn02 on June 11th, 2007 1:04 pm

    What is it that ESPN does *well*? Increase your cable bill by outbidding over-air networks for events rights? Force your college team to play games on Wednesday nights? Screw over their own employees?

  43. msb on June 11th, 2007 1:07 pm

    #41– I was thinking more along the lines of Steve Kelley vs Softy ….

  44. carcinogen on June 11th, 2007 1:11 pm

    43: Ok! Now we’re talking. Here’s my top 5 in order of likelihood.

    1. Steve Kelley
    2. Dick Fain
    3. Mike Gastineau
    4. Bob Finnegan (emeritus)
    5. Softy

  45. msb on June 11th, 2007 1:14 pm

    Unranked pitcher Chris Reitsma is back on the DL

  46. Eleven11 on June 11th, 2007 1:22 pm

    All math to be valid must be “proved”. So, who would you rather have, Albert Pujols or Maglio Ordonez (I know different positions but it is a ranking)??

  47. Gomez on June 11th, 2007 1:23 pm

    This system is bunk but some of your comments, guys, are hilarious.

  48. Orlandu on June 11th, 2007 1:33 pm

    Apparently Chone Figgens is better than Jose Guillen. Interesting.

  49. Max Power on June 11th, 2007 1:40 pm

    Hey – maybe it’s not all bad. Jose Vidro rates as the worst rated offensive player for the M’s – behind Jason Ellison (!), Willie Bloomquist and Ben Broussard.

  50. Orlandu on June 11th, 2007 1:40 pm

    Figgens is also better than Lopez and Betancourt. Don’t they realize he’s the Angels’ answer to Willie Bloomquist?

  51. Chris Miller on June 11th, 2007 1:43 pm

    Chone Figgins > WFB.

  52. colm on June 11th, 2007 1:48 pm

    Did anyone ask Jeff Bennett if he were a GM would he trade, say, Albert Pujols for Magglio Ordonez? Or Jose Guillen for Chone Figgens?

    At the end of his essay, Politics and the English Language, George Orwell says something like: “Break any of these rules sooner than write anything barbaric”. When the results you get are as barbaric as some of these rankings appear to be you need to throw out the rules altogether.

    Back to the spreadsheet for ESPN.

  53. hub on June 11th, 2007 1:53 pm

    As silly as these rankings are…Anything that gives Putz the credit due him, is a good thing. If nothing else, casual fans outside the AL West will soon be saying, ‘Whoa…this JJ-guy is doing great! When is Cashman gonna get off his ass and trade [insert useless scrub here] to get this guy for us?!’

  54. kenshin on June 11th, 2007 1:55 pm

    I don’t understand how ESPN could publish this tripe. Don’t they care about their reputation/product value at all?

  55. Chris Miller on June 11th, 2007 1:56 pm

    The Ordonez stuff isn’t fair, he has been the MVP so far. He probably won’t be come year end, but if you’re measuring value created, he’s been similar, if not better offensively than A-Rod, and Pujols has just started hitting again.

  56. Chris Miller on June 11th, 2007 1:57 pm

    #54, this stuff is basically just glorified fantasy ratings, so I doubt it’s being viewed quite as poorly by most of their readers.

  57. Eleven11 on June 11th, 2007 2:00 pm

    55, well maybe Chris but I think it points out two things, a two month rating is as silly as WFB’s ST stats and to have a rating of Albert that much lower than Maglio does suggest an “issue” or two…

  58. hub on June 11th, 2007 2:04 pm

    #56: I agree wholeheartedly. The ’5×5 H2H’ Fantasy leagues will eat this stuff up. I’m sure they are already singing its praises, and how it helps them in their trade of an elite 2B for a CL+1B. That might have been ESPN’s targetted audiance all along.

  59. Chris Miller on June 11th, 2007 2:09 pm

    #57, how else would in season ratings work? Should you regress? That would only be useful if you were trying to project the player. If you’re measuring performance thus far, A-Rod or Ordonez are tops in the AL.

  60. Steve T on June 11th, 2007 2:18 pm

    I’ve seen at least a dozen “systems” like this over the years. No, make that three dozen. They’re all the same; they want all the advantages of hard numbers but they can’t accept the basic premise of them, that they’re MEASURING something concrete.

    These sportswriters are absolutely wedded to the idea of judgment. It’s the foundation of what they do professionally; they judge who’s best, by naked observation and experience, with never any reference to a serious measurement.

    Maybe they think they are — they think BA is a serious measurement, which it is (but of the wrong thing). But the flaws of BA they MUST assume can be sorted out by experience. After all, if you could just look up things in a table, you wouldn’t even have to know how baseball is played to tell who’s best, right?

    The notion that numerical analysis based on recording actual events is even possible, let alone useful, is alien to them. So they make stuff up.

    Hmm, gotta have something about clutch hitting in there, right? My years of watching baseball tell me that…Ortiz is clutch. He gets a high mark there, whaddya think, about a 95? And ARod gets a 12. Yeah, that’s about right. This is how science works, isn’t it?

    The reality is, without recording events and analyzing the results, you can’t tell ANYTHING about what baseball are better than others. Game observation without recorded results is WORTHLESS.

    So: the problem with this new “system” isn’t that it’s crap; it’s that it was developed along lines that GUARANTEE its crapness. Even if the ESPN ranking was EXACTLY THE SAME AS VORP, player for player, it would be garbage.

  61. Eleven11 on June 11th, 2007 2:21 pm

    That’s a good question and I don’t know the answer but in season ratings early on tend to put Jose Lopez in the All Star game.

  62. Chris Miller on June 11th, 2007 2:30 pm

    I actually kind of like the idea of a ranking that weights expected performance against actual performance for things like MVP voting and HOF, give some credit for what happened, like giving pitchers SOME credit for W/L and ERA for HOF, but still weighting that against peripherals or expected performance, but this doesn’t do it, this is junk, and does nothing to tell you about who’s gonna do good going forward, nor what they produced.

  63. Jeff Nye on June 11th, 2007 2:40 pm

    If it were something intended to be a tool mainly for fantasy owners, and is intended as such, it should be presented that way. This is being presented as an analytical tool, which it isn’t.

    I think ESPN just wanted their own in-house pseudo-sabermetric stat that they could use instead of the already established ones owned by others.

    If they use VORP, Win Shares, etc, they have to use the intellectual property of another company or person, and at the very least credit them, if not compensate them.

    Everything that ESPN does is about their brand identity, and using tools created by others interferes with their brand message.

  64. colm on June 11th, 2007 3:06 pm

    and their brand message is? Ignorant and happy about it?

  65. Karen on June 11th, 2007 3:09 pm

    #34 Evan Says: “Nate Silver just did some quick work applying Rating back to 1997 to see what it told him. It’s on BP Unfiltered.”

    In case anyone hasn’t yet found Nate Silver’s BP article, here’s the link for “Bristolian Goulash”.

  66. Xteve X on June 11th, 2007 3:55 pm

    can’t wait for the first Joe Morgan reference to this or to see how long it’ll be before they start using it on the Sunday night broadcasts.

  67. Evan on June 11th, 2007 4:14 pm

    Joe Morgan probably thinks it’s a good stat because Billy Beane made fun of it in that book he wrote.

  68. joser on June 11th, 2007 7:58 pm

    and their brand message is? Ignorant and happy about it?

    Pretty much. They do employ Joe Morgan.

    Seriously, Nye pretty much echoed my thinking. They want something that belongs to them, and the ownership is more important than the quality. Afterall, their bread-and-butter (or, apparently, beer-and-viagra) demo isn’t going to pay any attention to how the number is calculated; it’s just something that provides a nice ordering to that day’s Table of Names during BBTN. If Fantasy owners start using it, so much the better; if it provides some controversy, better still.

    Dave is right: the best thing you can do is ignore this and hope it dies a quiet, unlamented death.

  69. Jeff Nye on June 11th, 2007 8:06 pm

    Their brand message?

    Chris Berman shouting “RUMBLIN’, BUMBLIN’, STUMBLIN’!”

    ESPN is about selling the excitement of sport. They are very good at that, and have built one of the most successful cable networks ever that way.

    They are not very good at anything else.

  70. pgaur82 on June 11th, 2007 10:47 pm

    I’m curious as to why Kenji doesn’t show up at all when you select All Teams /C….

  71. The Ancient Mariner on June 12th, 2007 4:15 pm

    On another note, shouldn’t you be Jeff Nye the Baseball Guy?

  72. Chris Miller on June 12th, 2007 4:45 pm

    It’s probably best to look at 1B/BIP than 1B/AB total singles. They’re still tops in 1B/BIP, but it’s not runaway like looking at 1B/AB. They barely edge out Cleveland. Being a low walk low strikeout team, you’ll see a lot of hits because of it. I suspect they’ll regress some, but they were built and trained to be that way, so who knows, they could just keep on hitting well as a low-power, high average team.

Leave a Reply

You must be logged in to post a comment.