The scouting scale explained

In light of the recent release of EJ’s excellent Top 30 list (and new grading methods) and Baseball America’s Top 100, I thought this would be an excellent opportunity to delve a little deeper into the traditional scouting scale.  Often, when reading a scouting report of a prospect, you’ll see something like “Player X has average speed but a plus arm” or “Player Y is a 60 hitter”, and generally you have an idea of what the scout is trying to convey.  However, the specific meaning of these grades is often lost.  I’d like to address this topic in two parts:  First, I will give a (brief, I promise) explanation of the statistical theory behind the scale, to give a better idea of how it is supposed to work.  And second, I will provide some operational definitions for what objective measurements/statistics the grades are supposed to correspond to, with some examples from current major leaguers and Yankee prospects.

The 20-80 scale is the means by which a scout evaluates the 5 tools (power, speed, hitting, arm, defense) in the case of a position player, or the repertoire of a pitcher.  Each tool or pitch is given a grade ranging from 20 (the lowest) to 80 (the highest), usually in increments of 10 (though occasionally you will see some 5’s).  The scouting scale is theoretically based on the normal distribution (see below, and if basic statistics make your head hurt, feel free to skip the next 3 paragraphs).

What the normal distribution shows is that in a large population, the distribution of a particular variable/characteristic will follow the above pattern (often referred to as the bell curve or Gaussian distribution) with a specific mean and standard deviation.  It is a reflection of the probability of finding a particular value if you picked one subject randomly from the population.  As you can see, the vast majority of the probability is right around 0, which in this case is the mean.  In baseball terms, the mean would be the average ability of a major league player at that particular skill.  The probabilities get significantly smaller as one moves away from the mean, which illustrates the relative rarity of extreme scores.

For a reason unknown to me, on the scouting scale, the mean is a 50 score.  As such, a pitcher with a 50 fastball is said to have an average major league fastball (more on what that actually means later).  Each 10-point increase/decrease in the scouting score corresponds to a change of one standard deviation away from the mean in terms of probability.  So a player with a 60 hit tool would be one standard deviation above the mean, which approximately corresponds to being in the top 15.86% of hitters in the league (and 84.14% are below).  A 60 tool is also the same as a “plus” tool.  As we move up one more standard deviation to a 70 tool, you can see the probability drops dramatically.  A 70 tool is in the top 2.27% of all players, with nearly 98% below.  A 70 tool is also referred to as a “plus-plus” tool.

Moving up one more standard deviation gets to an 80 tool, which has no other name.  With only 0.13% of the distribution at 3 standard deviations or greater, we can see how rare it is to see a true 80 tool in the majors.  The same probabilities apply as you move leftward from the mean.  As you might imagine given the probability, there are only a few true 80 tools in the majors at a given time.  Ryan Howard‘s power, Ichiro’s hit tool, and Aroldis Chapman‘s fastball are among the major league tools that would grade out as an 80, so this grade really represents the truly elite.  For prospects, an example of an 80 tool would (sadly) take us outside of the Yankee system, to something like Bryce Harper’s power or Mike Trout‘s speed.

As to the numerical significance of these grades,  Prospect Watch provides the most complete definition that I have seen.  Speed is defined in 2 ways: in the player’s overall 60-yard dash time, and the player’s speed from the batters’ box to first base.  Fielding is more of a qualitative evaluation, since fielding measurements are at present fairly unreliable.  Arm strength is sometimes measured using a radar gun, but often is more qualitative.  For catchers, however, arm strength is often measured in pop times, which is a measure of how quickly the catcher gets the ball to the base (and also factors in quickness of transfer.  An elite pop time (an 80) is below 1.74 seconds, and a 70 is between 1.75-1.79.

Hitting ability is based on projected future batting average, with a 70 hitter projected as a perennial .300-.329 hitter, and an 80 hitter consistently above .330.  Power hitting is also pretty easy to quantify, as it is based on projected seasonal home run totals.  A player with 60 power is projected for 26-34 homers, 70 corresponds to 35-44, and an 80 corresponds to 45+.  Most pitching tools are qualitative except for a fastball, in which an 80 corresponds to 99+, and a 70 corresponds to 94-98 mph.

One nice feature of the new BA list is that it includes the top tool and grade on that tool for every player on the list.  With 6 Yankee prospects on the list, it is a good opportunity to get an idea of how the scouting community grades these prospects (on their best tool, at least).

Jesus Montero checks in at #3, with his 70 power checking in as his best tool (though I have heard that his hit tool was also a 70 in the prospect handbook).  What these grades signify is that Montero projects to hit consistently over .300 with 35-40 home run power, production that will play at any position.  I’d be interested to see what grade his defense received, but I imagine it wouldn’t be pretty.

Gary Sanchez is somewhat surprisingly the second Yankee on the list at #30, and his 70 power demonstrates the huge offensive potential that this kid has.  I imagine his hit tool is probably not a 70 (at least not yet), but with his better defensive projection (as compared to Montero), it’s easy to see why everyone is so excited about Sanchez, and why he is so highly ranked despite his young age and lack of significant experience.

Manny Banuelos is the 3rd Yankee on the list at #41, and his best tool is his 60 command.  I imagine his fastball and changeup would probably be approaching “plus” territory as well, but this grade is consistent with what has long been Banuelos’ strength: his command and pitchability.  That he can back it up with two potentially plus offerings and a decent curveball is icing on the cake.

Dellin Betances is #43 on the list, with his fastball grading out as a 70.  This confirms what we have been reading about Betances: his big velocity is back, and he is living pretty consistently in the mid-90’s.  If I had to guess, his curve would probably grade out as a 60.  With 2 knockout offerings, only a checkered health history and development of his changeup stand in the way of becoming a frontline starter.

Andrew Brackman also makes the list at #78 with a 70 curveball, while his fastball would likely grade out as a 60.  Like Betances, he has 2 nasty pitches and a developing (but solid) changeup, but command and mechanical consistency are the big concerns.

Last but not least, Austin Romine just makes the list at #98 with a 60 arm.  While his receiving skills still need work, Romine’s arm strength and pop times have always been impressive.  While Romine’s offensive tools probably are in the 50-55 range at this point, that production would be quite acceptable if Romine can continue to develop his defense and become an above-average defender.

Like every other grading tool, the scouting scale certainly has its flaws.  It is often inconsistently applied among observers, and the 5-10 multiple gradation does not give a ton of room for precision.  It also does not include any valuation of the relative importance of the tools, which varies significantly depending on position.  Additionally, there is often the uncertainty of whether the grade is representative of a ceiling, a likely outcome, or some other determination of talent/value.  Despite its flaws, however, its another useful method of analysis to keep in the toolbox, and hopefully after reading this post you will be able to better understand the scouting lingo that gets tossed around.

11 thoughts on “The scouting scale explained

  1. Keith Law just said this post wasn’t accurate. Is it a novel concept for a writer talking about scouting like he knows anything to actually call a real scout for opinion,analysis and facts?

    • He said it wasn’t really accurate, which is different. We are working on finding out what he meant. Is it a novel concept to post with civility in a comments section?

      • how wasn’t the reply civil? Shouldn’t we assume the writers here would check their facts and do all the nec research before submitting a work? If one is going to do a post about scouting one would assume said writer at least spoke with a few scouts first.

        • I think it’s a legitimate question, albeit perhaps not so diplomatically phrased. I did my research, and while it’s true I didn’t speak to anyone, I did do a lot of reading on the subject. If I knew Law’s specific objection to what I wrote it would be easier to address/fix.

        • The implication of the words ‘like he knows anything’ was antagonistic, to me. If it wasn’t meant that way, I apologize for my presumption. And I have asked Keith about it, we shall see what he says. I will say that Eric has been writing about prospects for a long time, and this is a concept you tend to pick up in the process. So I have no real issue with him not contacting a scout about it. Now that a scout has told us it is not entirely accurate, we are working to correct it.

  2. Pingback: Weekend Round-Up – 2/27/2011 | Fire Brand of the American League