Grading on the Curve

facebooktwitterreddit

I’m a fan of the Bell Curve because it can be used to measure almost anything. If you aren’t familiar, it’s a bell-shaped curve that tapers at both ends, illustrating the idea that under normal distribution people cluster toward the center (average), then are infrequent toward the edges (far above or below average).

You may not know that scouts grade on a curve as well, using a scale between 20 and 80. When they are out scouting a player they assign each tool a number from 20 (Horrible. Think Ben Revere‘s arm or Jim Thome‘s speed.) to 80 (Elite. Think Giancarlo Stanton‘s power or Billy Hamilton‘s speed.), with 50 representing the Major League average. It’s thought that Branch Rickey invented it way back when.

Scouts are so versed in this scale that they can toss numbers around among themselves while really sharing a picture of a player’s skillset. I’ve begun to do this in my head as I’m “scouting” fantasy baseball players, but it’s important that you really understand the scale so that you properly weight players.

Look at the image above. W’eve established that 50 is average. One standard deviation above and below represents approximately 68% of players. So almost 7 out of 10 players should have their skills fall between 40 and 60, which is important to baseline so you aren’t indiscriminately tossing around “grades” too high or too low. 95% of players fall between 30 and 70. Only he players who are truly, truly horrible (bottom 2%) with a particular skillset should be graded in the 20’s. Likewise, only the truly, truly elite (top 2%) should get a grade between 70-80. (Note: You can also chop off the zeroes and just scale from 2-8.)

  • 20 = Horrible
  • 30 = Well below average
  • 40 = Just below average
  • 50 = Average Major League regular
  • 60 = Above average
  • 70 = All-Star
  • 80 = Elite, Kate Upton

As I said, I do this with fantasy players and it’s pretty fun and remarkably helpful once you’ve had a little practice so you can grade on auto pilot. I’ll also scale other things. I just had sloppy joes and I can give them a solid 50, but the dessert pizza I had afterward gets a comfortable 70. But back to fantasy baseball. Giancarlo grades an 80 power, 50 average, 65 for RBI, and 35 in terms of speed. Mike Trout grades 65 power, 80 speed, 80 in terms of run scoring, but just a 50 with RBI, simply because he bats leadoff.

But that’s just for stats and doesn’t really stretch the utility of this tool. It can also be used to scale confidence in a couple ways. Kris Medlen probably pulls down just a 35 in terms of confidence because he doesn’t have that many innings pitched to build a reputation that we can trust him. A player like Edwin Encarnacion grades a similar 35 in terms of confidence, but it’s because his skills took a huge jump last season. The moral of that lesson would be to not draft a team of players that all grade low on your confidence scale.

We’re just starting to have fun with this and we haven’t even talked about grading a player’s ceiling / floor (potential), but I want to tell you about something we’re doing for our draft kit this year. We’re going to grade for risk / reward.

We’ll grade player for risk. That is, does he have an injury history, is it a small sample size, is he getting up in age, or has he simply shown erratic skills in the past? Those we consider dead average get a 50 risk. While no one can guess when a broken leg will strike, they should be pretty darned dependable. Some we’ll score low risk because they’re as steady as they come, while others will score high risk and you won’t want many of those on your team.

We couple that with reward. Some players offer a higher shot at reward. If a player offers high reward with low risk, you can count on paying more for them at the auction table. Inversely, a player who offers low reward, but high risk should be avoided like the plague.

Conclusion

I share this to give you another way to judge players. It’s versatile in that you can grade skills, abilities, risk, potential, or sloppy joes using it. But despite it’s versatility, it all uses a common scale that is easy to get your mind around.

Second, I share this as a way to give you a sneak peek at a tool we’ll use in the draft kit that gives a nod to the world of real life Major League scouts, who aren’t all the stiff-necked luddites portrayed in the Moneyball movie.

Finally, I share this because it’s fun to grade this way. With a little use it grows to offer remarkable utility. Now go practice on your sloppy joes.