Introducing NERD for Position Players

This is a post introducing and explaining NERD scores for position players. I’m including the results first and then the background, methodology — all that junk — second.

Curious as to what NERD is? The short answer is: it’s a number, on a 0-10 scale, designed to express the “watchability” of position players for those of the sabermetric persuasion.

For more information, consult the index right after the results.

The Results
Here are the current top-20 position players by NERD:

Here are the bottom 20:

You can find the complete results in this spreadsheet.

Index
1. The Results
2. Introduction
3. Notes on Choosing the Components for Player NERD
4. Components of Player NERD
5. Calculating Player NERD
6. The Final Equation
7. Results for 2010

Introduction
Despite at least one convincing argument to the contrary, it is, in fact, not space, but rather the objective measure of a position player’s (i.e. non-pitcher’s) watchability, that is the final frontier.

The third- and second-to-last frontiers, as the attentive reader will already know, were the objective measures of starting pitchers’ and teams’ watchability, respectively. These frontiers have been settled, incorporated, and heavily urbanized (metaphorically speaking) by the introduction and enthusiastic use of NERD — that is, the number, on a 0-10 scale, designed to express the “watchability” of a game or pitcher for the learned fan — available in each and every edition of One Night Only at this site.

This omission of a NERD score for position players has been a glaring one, however, and one that has made for the occasionally awkward interaction with Messrs. Appelman, Cameron, etc. around the work of modern art that is the FanGraphs’ water cooler.

Accordingly, in what follows, I’ve endeavored to right this wrong.

Notes on Choosing the Components for Player NERD
The question with which I started — with which any sensible person would start — was, What makes a position player interesting to the baseball nerd? And, on the heels of that first question, this one, too: Of the things that might make a player interesting, which of them are easily measured?

In fact, while both the pitching- and team-oriented versions of NERD contain at least six variables each, it occurred to me that, for the field player, Wins Above Replacement — especially with the recent introduction of baserunning — is almost entirely sufficient.

Components of Player NERD
• Age (Relative to League Avg)
As has been the case with the pitching and team varieties of NERD, such is also the case for player NERD: younger players are exciting. If we consider two players, identical in terms of WAR and everything else except (one is 21; the other, 29), it’ll be the younger player about whom we care more. With one exception, that is. There were players in 2010 (with which numbers I started) who were maybe more interesting because of their advanced age — Jim Thome and Jim Edmonds, specifically. For that reason, players who are greater than two standard deviations above the league-average age begin to receive a bonus again.

• WAR (WAR per 650 PA)
At this point, because it considers so many variables (park-adjusted batting, fielding, baserunnung, mustaches), WAR for position players addresses the things that are also most appealing to the learned fan. There are few players atop the WAR charts who we might expressly not like (unless, of course, said chart-topper as particularly aggrieved our favorite team). I’ve used WAR per 650 plate appearances so that all players can be put on the same scale.

• Batting Luck (xBABIP-BABIP)
I’m quoting myself and most of the bespectacled readership when I say that baseball nerds like watching regression happen. This is why, for example, Luck (ERA minus xFIP) is one of the components of Pitcher NERD. For batters, the concept of luck is more difficult to isolate. We know, for example, that BABIP generally regresses towards .290 or .295 for pitchers. It follows, obviously, that it does the same for the entire population of hitters, too. However, the individual data points (Chris Snyder, Ichiro Suzuki) are more widely scattered. To address this, a number of xBABIP calculators have been — none more easily utilized with FanGraphs’ batted-ball data than slash12′s version at Beyond the Boxscore.

Calculating Player NERD
To calculate NERD, I found, and added together, the z-scores (standard deviations from the mean) for Age and WAR/650 (for all players with 100+ PAs). To this I added the Luck factor multipled by ten (such that, say, a .033 difference between BABIP and xBABIP becomes 0.33). That done, we now have the Raw score.

Because there are fewer variable in this iteration of NERD, the Raw scores aren’t naturally as spread out as in previous versions. To help translate the Raw scores to the 0-10 scale, I’ve multipled same by 1.5.

After doing that, a constant (presently, around 4.2) is added to create an average NERD of 5.00 among the entire population.

The Final Equation
Looks like this:

[[AGEz + WARz + (Luck * 10)] * 1.5] + Constant

Results for 2010
For reference, here are the results from 2010 (with which I actually started, so’s to get the numbers right):

Top 20:

Bottom 20:

A full spreadsheet of all qualified (100+ PA) players can be found here.




Print This Post



Carson Cistulli occasionally publishes spirited ejaculations at The New Enthusiast.

58 Responses to “Introducing NERD for Position Players”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Yirmiyahu says:

    Mr. Cistulli,

    Your system is flawed. Jose Bautista needs the dial turned up to 11.

    +8 Vote -1 Vote +1

    • Daniel says:

      Really, though, how is Jose Bautista not even on the list for last year?

      Vote -1 Vote +1

      • He was 28th overall, with a high-end 7. That still puts him in, like, the top 7% of all qualified players in 2010.

        Ideally, there’d probably be some “breakout” additive — like, if a player found a considerably higher level of performance over the previous season. That’s interesting in a similar way that young players are interesting, I suppose, in that it allows the fan to wonder about what might be possible for the player.

        Still, 7′s not so bad. Also, this is interesting: Bautista’s .233 BABIP was pretty low, but his xBABIP only came out to .270. That’s interesting to me.

        Vote -1 Vote +1

      • Yirmiyahu says:

        On that Bautista BABIP note, it seems like your xBABIP factor doesn’t take into account how much the discrepancy actually matters. Sure, maybe Jose Bautista is vastly over- or under-performing his xBABIP, but for him it doesn’t really matter because he rarely puts the ball in play.

        Dude would have about an .800 OPS even if his BABIP was 0.

        Vote -1 Vote +1

  2. Steve Balboni says:

    Any new model must past the smell test– self-evident errors stink. Like an NBA system that says Pau Gasol is more valuable than Kobe Bryant. I think you fail, either redo it or rename the model.

    It is not possible that Vlad Guerrero is among the least watchable guys in the league. Jack Cust is a must watch, in case he runs the bases. Juan Pierre? I could watch his tiny head, crazy routes and spaghetti arm all day. Cabrera has an incredible accurate arm, watching to see if a 1b has to adjust his arm by 6″ is a game in itself.

    Most egregious: Milton Bradley? Take a look at feature film box office results: #1, Hangover II = crazy people doing crazy things.

    Vote -1 Vote +1

    • James says:

      You seem to missing the point – “watchability… for those of the sabermetric persuasion”

      Sabermetrically speaking, the aforementioned all suck now. Vlad is a shell of his former self.

      Vote -1 Vote +1

      • Steve Balboni says:

        I caught that, but are you suggesting we don’t like watching baseball? Or watch it through an excel filter?

        A model that highlights statistical anomalies should be, as I originally wrote, called something else. Like “Regression Idol” or “Survivor Outlier”.

        Vote -1 Vote +1

    • DannyG says:

      I also agree with the above poster re: Juan Pierre. I watched many a Yankees and Red Sox game just waiting to see Damon forced into a long throw, which he would undoubtedly make with the style and technique of the 12-year old girl who delivers my weekend newspaper. We could add a 4th column for ‘intangibles’ where someone like Pierre, Cust, Damon get bumped for hilarity, and every White Sox player loses points for Hawk Harrelson.

      I appreciate James’ comment about being for the sabermetic persuasion, but come on, we’re not robots here!

      Vote -1 Vote +1

      • James says:

        As a White Sox fan, Pierre makes me want to gouge my eyes out. Likewise, Harrelson offends my ears and Ken Willaims my brain.

        I posit that no one worth breathing would seriously considering watching a game solely due to any of the above players participating in said game. There is no real basis for any of them being in the show at this point.

        Vote -1 Vote +1

      • James says:

        I should temper that… I’m primarily referring to Pierre/Vlad/Cust/Bradley… Some of the players on the list have some utility as role players, but I wouldn’t pay to see them.

        Vote -1 Vote +1

    • Bill says:

      Poor analogy with Kobe and Gasol. Kobe is one of the most overrated players in basketball and the Lakers did nothing until they got Pau.

      Besides that I agree with pretty much everything else you said.

      Vote -1 Vote +1

  3. Matt Defalco says:

    Something is wrong here… No jose bautista?? Something must be changed, and I feel like three our so statistics doesn’t allow for an accurate assessment. Perhaps something like pa/hr, speed score or bsr wins, with other factors. I just feel that, while WAR is a fantastic tool for displaying a player’s value, it is too cover-all to be used as such a net in such a specific and innovative statistic. Also, there should be a special component assigned to players which show a flare for the watchable. See: josh collmenter (his mechanics), or sam le cure and his recently notgraphs highlighted mustache. The value could be a simple 1 (neutral) or 2 (interesting).

    Vote -1 Vote +1

  4. Matt Defalco says:

    Also I’m posting from a phone here… Allow me some grammatical leeway

    Vote -1 Vote +1

  5. Yirmiyahu says:

    I’m not a fan of using WAR as the bulk of the score. For one thing, it’s not a rate stat, so playing time messes with things. And a guy who is above-average but not spectacular in all the WAR components might have a nice WAR, but is probably boring to watch. What’s exciting to watch are great hitters and great fielders. Positional adjustments are not entertaining to watch.

    You need to add in speed. Stolen bases, caught stealings, pickoffs, rundowns, taking the extra base, triples, etc. This stuff might not be valuable from a saber perspective, but it’s exciting to watch. Even if the result is an out.

    Hell, for that matter, terrible fielders can be fun to watch too. I miss Manny at Fenway.

    Home runs. Chicks love the long ball for a reason.

    Vote -1 Vote +1

    • ppabich says:

      It is WAR per 650 at bats. So it essentially becomes a rate stat. If you open the spreadsheet, Mike Napoli and Allen Craig are in the same WAR range as Joey Votto and Curtis Granderson.

      Secondly, bad baseball might be exciting to you, but for me there aren’t many things worse to watch than bad baseball. I don’t like to see players kick balls or run into outs.

      Vote -1 Vote +1

  6. Dave says:

    I too am outraged about Joey Bats’ snubbing of not being named to the 1st Team All Universe MVP Squad (award does not actually need to have been won for consideration). All systems that fail to recognize Bautista as the greatest player to have ever played (ties don’t count)… over the past year and half, should be considered completely bunk. Other than that I really enjoy seeing NERD scores for players. Thanks!

    Vote -1 Vote +1

    • Yirmiyahu says:

      I can’t tell if this is sarcasm. If it isn’t, I totally agree.

      Vote -1 Vote +1

      • Dave says:

        It’s unnecessarily hyperbolic, so much so it’s almost sarcastic, but not, it’s not actually sarcastic. I like the NERD scores for position players and I also like Joey Bats an almost unreasonable amount.

        I think the best thing about watching his homeruns are that even his wall scrapers are pummeled, they just don’t get much higher than the wall during their travels.

        Vote -1 Vote +1

  7. DannyG says:

    If we, the nerds, like to watch regression happen, then shouldn’t the luck factor be calculated as a magnitude? The current formula considers a player more watchable if they ‘should’ be hitting better than they are. Since we are bespectacled baseball nerds and not excitable NASCAR fans, we should be just as interested in watching players playing over their head regress. Aren’t we all curious to see how long Jose Bautista and Matt Joyce can keep hitting .350?

    Vote -1 Vote +1

    • Doug Lampert says:

      If the player is playing over his head, he presumably gets increased WAR, so adding in again for regression would be double counting luck in his case.

      Really good play should count for more than really bad luck in terms of interest, so the players who are doing well should top the list.

      Vote -1 Vote +1

  8. siggian says:

    Some other considerations:

    We like to watch league leaders. How about instead of that 1.5 multiplier, you put in some sort of bonus for the top 5 players in things like HRs (who, except pitchers, doesn’t like to see dingers), SBs, doubles, runs, hits (I’ll happily watch an Ichiro AB), range in defense (by position), plus other things.

    Walks are inherently uninteresting unless the bases are full. No bonus there.

    Batters who step out constantly to adjust everything with every swing are less interesting but I’m not sure we can measure that.

    Debuts are interesting for the first week.

    Rookies are interesting if they are doing well after the first month or so. If the rookie is struggling, he’s no longer that interesting.

    Just throwing some ideas out there.

    Vote -1 Vote +1

    • Tim says:

      I think HR/FB% could be worked in to represent the HR thing.

      Vote -1 Vote +1

    • Evan says:

      “Batters who step out constantly to adjust everything with every swing are less interesting but I’m not sure we can measure that.”

      For the most part, yes. But I always liked watching Nomar’s pre-AB routine.

      Vote -1 Vote +1

  9. OremLK says:

    I think you need to include more variables in this, such as baserunning and defense. Obviously, those are part of WAR, but I think NERD should be weighted more toward those things than, say, BB%. I can admire a guy who draws walks, but let’s face–watching a guy *not swing at pitches* isn’t half as exciting as watching a guy make a sparkling diving play.

    Vote -1 Vote +1

  10. badenjr says:

    Carson,

    I think there is an interesting distinction that needs to be made here. Are you trying to identify players that are watchable because they are exciting or because they are sabermetrically interesting? If you want to identify players that are watchable because they are exciting, you should be looking at things like home runs, triples, stolen bases, and some measure of fielding (in my opinion). If you want to get at sabermetrically interesting players, you start to look at BABIP, plate discipline, batted ball profiles, wOBA, WAR, etc.

    From a sabermetric perspective, I’d think someone like Adam Dunn or Jack Cust is pretty interesting due to their TTO nature. It wouldn’t be exciting to watch them play though. I think someone like Brandon Phillips – while perhaps overrated in general – should rate fairly high from the perspective of being exciting to watch play. From my perspective, any system that ranks how fun it is to watch players should identify early 1980s Rickey Henderson and early 2000s Barry Bonds as the most interesting guys to watch in the last half-century.

    Vote -1 Vote +1

  11. Jim Lahey says:

    I’m not really sure I like seeing Aubrey Huff, Aaron Hill, Rajai Davis, JD Drew or Vlad on the least interesting players of this season.

    These guys are all coming off of recently good seasons and aren’t so over the hill that they’re unbearable to watch. I think that if a player has had an All-Star appearance / been the team MVP / received gold glove, silver slugger, etc… within the last 3 seasons, he should be given credit.

    What about a luck factor that was based off ZIPS wOBA to current wOBA?
    Last season wOBA to this season wOBA?
    Something that isn’t weighted / walks aren’t accounted for because walks suck.

    Vote -1 Vote +1

  12. Ryan Waltman says:

    this makes no sense…VLADDY VLADDY was awesome last year!!! he was the hottest thing in sexy DH pickups! NERD looks like a way to find trendy new players to rave about…is it?

    Vote -1 Vote +1

  13. Llewdor says:

    Well that settles it. E5 is unwatchable in the field, AND at the plate.

    Vote -1 Vote +1

  14. Llewdor says:

    I object to your methodology, though.

    Players who are oddities are watchable just for that. 2011 Jack Cust, he of the walking skill, but not the hitting skill, is fun to watch. Jack Cust, this year, is either going to walk or strikeout in nearly every plate appearnce. Watching him keep that pattern going is breathtaking.

    Vote -1 Vote +1

  15. Jim Lahey says:

    I should make my point…
    I think awards should be included somehow because we like to evaluate whether players that receive the awards actually deserve them, or whether the guy that carried the team last year was lucky.

    For example: When I turn on a Giants game, I want to see Aubrey Huff. He was arguably the Giants best player last year. Last year he hit well and put up a fantastic WAR, but, sabermetrically, he’s freakin awful this year. Does that make him uninteresting? Maybe by more traditional measures, but I can appreciate how much a player sucks too, at least when he should be good. (Looking at all the Francoeur articles that have been written over the years, I’m not the only one)

    I feel like David Ortiz would have showed up on this list as the least interesting players at this point in the season 2 of the last 3 years because of his terrible start.

    Vote -1 Vote +1

    • eric says:

      I don’t know about awards, maybe saber ones (fielding bible awards, for example) but probably not mainstream media ones.

      How was Ortiz’s BABIP/xBABIP?

      Vote -1 Vote +1

  16. eric says:

    Outstanding work here

    Vote -1 Vote +1

  17. Jon says:

    Good work.

    A few ideas: Maybe factor in something for players recently returned from injury, or recent call-ups with good minor-league numbers? It’s always interesting to see the new guys, or key pieces returning (Utley, Hamilton).

    Maybe you could use the pre-season projections to get a gauge on how good a player is projected to be? Good way to judge who might be a hot call-up too.

    Vote -1 Vote +1

  18. Zak says:

    How can Jose Reyes not be on the list for this year?

    Vote -1 Vote +1

  19. Sven says:

    Danny Espinosa is currently my favorite player to watch. Good list.

    Vote -1 Vote +1

  20. eric says:

    Ichiro seems too low to me. Obviously he should be penalized for his slow start — a -0.2 WAR at this point is surprising — but it’s hard for me to think of him as being this low on the list: 236 out of 284. Even #170 last year seemed low. I don’t think he needs allstar or media narrative bonuses but as Llewdor comments oddities and unique skill sets are interesting. Also past performance might be a little underrepresented in this formula?

    Vote -1 Vote +1

    • ppabich says:

      well, you shouldn’t use these metrics on their own. Combine them with your own common sense, and you will get a nice view of who to watch. Ichiro will always be interesting, but guys like Gerardo Parra were not on my radar screen. Guys high on the list should intrigue you. Seeing Longoria on the list isn’t a surprise, but you wouldn’t tune him out anyways. So, the list should lead you to new interesting players.

      Vote -1 Vote +1

    • eric says:

      good point

      Vote -1 Vote +1

  21. Dave says:

    The Tigers have three of this year’s least watchable players. That sounds about right.

    Vote -1 Vote +1

  22. Mark Doo says:

    This is easily the stupidest thing I have ever seen.

    Completely meaningless and look at the absurd results.

    Vote -1 Vote +1

  23. keegs says:

    there should be a power or speed/defense element. in your bottom 10 are rajai davis and adam dunn. both those guys are awesome to watch, but for completely different reasons. Players like Inge and Hairston are what I think of as boring ballplayers, not Big Donkey.

    I don’t understand the appeal of xBABIP/BABIP thing. If I’m tuning into a game, I want to see guys who mash or guys who fly. Measuring and then ranking skillsets seems much more interesting than luck.

    Vote -1 Vote +1

  24. PaulScarfo says:

    NERD = the “Sabermetricians-don’t-do-intangibles-but-if-we-did-we’d-do-it-better-than-you” stat?

    This is truly a horrible piece of writing (analysis?) …

    Vote -1 Vote +1

  25. filihok says:

    That Juan Castro received a NERD rating of 0 leads me to believe that position player NERD is very much on the right track.

    Vote -1 Vote +1

  26. PWH says:

    The amount of Peter Bourjos I get to see each week is directly proportional to my overall happiness. Flawless formula.

    Vote -1 Vote +1

    • Andre says:

      Agreed. I love watching the kid play. I like this system!

      (and yes, even if you were being sarcastic)

      Vote -1 Vote +1

  27. KDL says:

    This has very hipster-rating-indie-bands feel to it. As soon as a player hits his stride he becomes less watchable? I like watching exciting new talent…but I’d much rather see guys like Pujols and Votto and Utley IN THEIR PRIME than check out guys on the cusp of their prime.
    If we’re going to factor in age…can we place a higher premium on year 26,27,28…or find some other way to call players in their prime more watchable than players getting there.

    Vote -1 Vote +1

  28. Josh says:

    I’m afraid I disagree with 2 of the 3 measures being used for this.
    WAR: is not that interesting. It’s a good measure of player quality, but as many have said, quality of play does not necessarily correspond to watchability. I agree with the several people that noted HRs, triples, baserunning, and defense in some fashion should contribute to this instead.
    Luck: not as interesting for batters as pitchers. A batter gets 4-5 PA/game. The sample size is too small to care about large-scale trends in their luck, whereas pitchers will face 15-25 batters per game, and while it’s still a small sample size, it’s 4-5 times more chances.

    It’s a good start and a fun idea, but please don’t stop here.

    Vote -1 Vote +1

  29. jeffrey gross says:

    I DISAGREE WITH MILTON BRADLEY. HE IS VERY ENTERTAINING TO WATCH. HE’S A TICKING TIME BOMB OF FUN

    Vote -1 Vote +1

  30. We are a group of volunteers and opening a new scheme in our community. Your web site offered us with valuable information to paintings on. You have done an impressive process and our entire community shall be thankful to you.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current day month ye@r *