The 20-80 Scale, Saber Style (Again Yet Different)

After taking a look at how scouting grades for tools might grade out statistically, I’m back to look at how some of our other commonly used metrics measure in regard to standard deviations. On Tuesday my goal was to look at each tool individually, and I focused, as a result, on making sure there were enough innings pitched and at-bats to at least indicate there was something going on. Today, I’ll look at metrics that look more at the whole player, so while I’ll continue to use the 20-80 shorthand, we’ll also talk about where players slot in the grand scheme of things. As you will see, some of the bins will match up with some of our preconceived notions, but it’s always good to confirm some things. Without further ado …

Grade wRC+ Player WAR/600 Player
80 165 Jose Bautista 7.5 Mike Trout
70 141 Giancarlo Stanton 5.6 Jacoby Ellsbury
60 117 Yadier Molina 3.7 Carlos Beltran
50 94 James Loney 1.8 Reed Johnson
40 70 Gerald Laird -0.1 Justin Smoak
30 46 Wil Nieves -2.0 Emmanuel Burriss
20 22 Ramiro Pena -3.9 Joe Mather

Leading off is wRC+. I used three-year averages again (2010-2012), but I lowered the PA qualifier to 200 in order to get the distribution within standard error. Nothing here should really shock you. Most players will fit into the 70-120 range. Jose Bautista is not the only one with an 80 score this time, however, as Miguel Cabrera and Joey Votto sneak in ahead of him. All the way on the other end, Ramiro Pena and former prospect darling Josh Bell hit the bottom of the scale harder than they do a baseball.

Taking a look at WAR, I broke into WAR/600 to give you an idea of what it would look like per season (roughly). Trout, as one might imagine, tops the list with almost 8.5 WAR/600, but I was surprised to see who came in second (answer at the bottom of the post). The rest of the breakdown is as expected – 0-2 wins is generally a bench player, 2-4 is a regular, 4-6 is an All-Star-caliber player, and a 6+ win player is an MVP candidate, with the 7-and-above players as your real contenders.

Grade OBP Player wOBA Player
80 .413 Prince Fielder .421 Joey Votto
70 .384 Ryan Braun .385 Buster Posey
60 .355 Norichika Aoki .349 Matt Joyce
50 .327 Kendrys Morales .312 Tyler Flowers
40 .298 Tyler Colvin .276 Everth Cabrera
30 .269 Cesar Izturis .240 Jack Wilson
20 .240 Drew Butera .204 Ramiro Pena

Moving on to OBP, the numbers probably aren’t much different than one would expect. Somewhere in the .320 range is pretty average, and every 30 points or so further away, you reach a new level of talent (or lack thereof). Prince Fielder, Miguel Cabrera, and Joey Votto are your 80-grade hitters here, and it shouldn’t be surprising that a bunch of first basemen (and one should-be-once-was first baseman) head the list. Mauer rolls in at #5 at exactly what it would take to get a 75. On the flips side of the spectrum, we have the Terrible Troika of Drew Butera, Jeff Mathis, and Brandon Wood.

wOBA paints a pretty similar picture. Fielder tumbles a bit out of the 80-grade range, and Jose Bautista continues to sit just below the threshold, which basically means he could be included. Down at the bottom, we again find Pena, Bell, Mathis, and Burriss all just a hair above the .204 mark, which again basically means they can be added.

WAR SP/200 Player RP/60 Player
80 6.8 - 2.33 Craig Kimbrel
70 5.4 Josh Johnson 1.64 David Robertson
60 3.9 Madison Bumgarner 0.94 Wade Davis
50 2.4 Joe Blanton 0.25 Jon Rauch
40 1.0 Bronson Arroyo -0.45 Kyle McClellan
30 -0.5 Jeff Suppan -1.14 Chuckie Fick
20 -2.0 Ryan Rowland-Smith -1.83 Brian Tallet

The pitching statistics were a little more interesting as far as parsing out the details went. Looking at their overall WAR numbers first, I separated the relievers from the starters, and I used 200 and 60 IP as the ratios to, again, get an idea of their per season rates. No one was able to get to the top grade in the starting pitching section, though Stephen Strasburg (6.4) got fairly close, but Craig Kimbrel did take 80 honors on the reliever end, with Mariano Rivera, Aroldis Chapman, and our second trivia answer all above 2.0 per 60 IP.

FIP AL SP Player NL SP Player
80 2.32 - 2.24 -
70 2.93 Justin Verlander 2.80 Clayton Kershaw
60 3.54 Dan Haren 3.36 Mat Latos
50 4.15 Kevin Slowey 3.91 Tim Stauffer
40 4.77 Jeanmar Gomez 4.47 Randy Wolf
30 5.38 Javier Vazquez 5.03 Ross Ohlendorf
20 5.99 Ryan Rowland-Smith 5.58 -

 

FIP AL RP Player NL RP Player
80 1.48 - 1.34 Craig Kimbrel
70 2.30 Junichi Tazawa 2.15 Aroldis Chapman
60 3.13 Esmil Rogers 2.95 Jason Motte
50 3.96 Tim Collins 3.76 Wesley Wright
40 4.79 Aaron Laffey 4.57 Josh Roenicke
30 5.62 Manny Delcarmen 5.37 Livan Hernandez
20 6.45 Brian Tallet 6.18 -

Moving on to FIP, I split the pitching groups a bit further into AL and NL because the run environments are a bit different for each league. As expected, the NL FIPs for both starting and relief pitchers were lower for the NL. Cliff Lee was the closest AL starting pitcher to an 80 grade as he tossed 2.58 FIP ball for the Mariners and Rangers a few years back, and Strasburg was close yet again at 2.47. Halladay’s 2011 (2.20) or Medlen’s run at the end of last season would have been good enough (2.22) to merit an 80 grade, so it takes one heckuva run to be that good right now. Medlen’s teammate Kimbrel (1.23) was the only reliever to get an 80 grade in his FIP as he was almost a full run lower than anyone else on either list. He is the clear-cut best reliever in baseball at the moment, and that’s the sort of talent you need to grade out an 80 player.

——

People never seem to underestimate what it takes to get an 80-grade anything. Statistically, only 0.1 percent of the population, or 1 in 1000, people are 3 standard deviations above (or below) the mean. 750 players will start the season on an Opening Day roster, which means there might not be an 80 in the bunch, and if we use the entirety of professional baseball, we have … 30 teams * 25 players * 8 levels … carry the 4 … 6000 players, meaning about 6 players will grade out as an 80 at something (though enlarging the population like that would mean changing the mean and deviations). Hopefully, this exercise has given us a bit of perspective on what it takes for one player to be clearly on another level than another player, and I hope it will help us appreciate just how great (or how bad) a performance really is.

——-

Trivia Answer: Yasmani Grandal came in second on the WAR/600 list demonstrating the impact he brought to the Padres in his brief time, and Sean Doolittle is the surprise reliever for his performance. Obviously, these are small samples, but I thought they were interesting, nonetheless.




Print This Post





19 Responses to “The 20-80 Scale, Saber Style (Again Yet Different)”

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

    Another awesome article. Craig Kimbrel is beyond good

    Vote -1 Vote +1

  2. Request Box says:

    Really dig this work. How hard would it be to add a column to player pages, whereby wOBA / FIP / whatever gets a 20-80 grade? Now that the legwork (and presumably math behind it) is complete, this could be another tool in the arsenal for discussions / arguments, as well as “who’s the most avg at …” articles. Make it happen!

    Vote -1 Vote +1

  3. philosofool says:

    Could you publish your standard deviations and means?

    Vote -1 Vote +1

    • reillocity says:

      He did essentially. Those can be inferred from each table with reasonable accuracy. Just look at what the stat value associated with the 50 score for each league is and that’s basically the league mean. The difference between the 60 score value for the league and the 50 score for the league is basically the standard deviation (notice that every 10 point change amounts to the essentially the same change in the stat, which is nearly the standard deviation).

      Vote -1 Vote +1

  4. Jason461 says:

    Are you including pitchers in your hitting samples at all? If so, they’ll skew the numbers down a bit, won’t they?

    Vote -1 Vote +1

    • guesswork says:

      Regardless, he is only including players with 200 PA so the results are going to be biased anyway. Typically players need to be well established or do well in a small sample to get 200+ PAs, which will skew everything upwards.

      Vote -1 Vote +1

  5. guesswork says:

    Again, I am going to recommend you redo this analysis using sample quantiles instead of a normal approximation. The normal approximation is nice, but may give unrealistic results for the players at the extremes. I think it would be interesting to see how the results differ as well. Chances are they will be very similar except at the tails.

    Of course, there are issues, namely a larger sample size would be needed (2008-2012 may suffice) and it is restrained to the sample. Thus it assumed that both a 20 and 80 player exist, but at least it makes no distributional assumptions.

    An even better approach? A nonparametric Bayesian approach that uses scouts’ actual grades as priors. No sample size issues or distributional assumptions! Of course, acquiring scouts’ grades is not easy as far as I know.

    Vote -1 Vote +1

    • JBImaknee says:

      There are explicit tests for whether a given distribution is appropriate for data. For most of this, given the sample sizes, assuming Gaussian is probably sufficient, and I’d imagine that the data would hold up. The one case where it may not is perhaps something like fastball velocity in the last article, where there would be expected to be a skewed distribution since pitchers can always throw slower whereas biophysics constrains upper velocity.

      That said, I think the real contribution here is the use of a distribution that makes “great” a function of average and variance. Nothing says that the best and/or worse of a population is something we should be in awe of, and all too often people try to force players into what should be rarified territory

      Vote -1 Vote +1

  6. Brandon says:

    Your math is wrong on the WAR/600.

    Mike Trout has had 774 career plate appearances for a total of 10.8 WAR. The proper way to calculate this would be to divide the 10.8 WAR by 774 PA = .013954 WAR per PA. Now multiply that by 600 = 8.37 WAR per 600 PA.

    I did not check the other players but the math may be wrong on those as well.

    Vote -1 Vote +1

  7. Jason B says:

    I thought 0.25%, rather than 0.1%, were above/below three standard deviations from the mean. (Or, stated another way, I thought 99.5% was within three stadard deviations of the mean.) But I may be mis-remembering basis statistics. At any rate, it doesn’t change your basic premise about the uniqueness or rarity of these types of players (and ‘two-and-a-half-in-a-thousand’ doesn’t have the same ring as ‘one-in-a-thousand’ does anyway).

    Vote -1 Vote +1

  8. SC says:

    This is Great!! Thanks for your work!!

    Vote -1 Vote +1

  9. Baltar says:

    I’m surprised that Burriss ranked that high.

    Vote -1 Vote +1

  10. Jon L. says:

    Yasmani Grandal, of course! Had him on the tip of the tongue.

    Vote -1 Vote +1

  11. Jason says:

    I’m not a fan of the WAR/600 results. I think it needs to have a minimum PA cap for position players; maybe 1500? The first two players on that list: Trout and Ellsbury stand out as having 1 great season only. Trout may never repeat it and Ellsbury probably won’t repeat his 2011.

    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 ye@r *