Scouting Explained: The 20-80 Scouting Scale

Scouting Explained: Introduction, Hitting Pt 1 Pt 2 Pt 3 Pt 4 Pt 5 Pt 6

When I started here just last month, I promised I would write a comprehensive series of articles explaining every part of the 20-80 scouting scale. This is the beginning of that series.

Background

The invention of the scale is credited to Branch Rickey and whether he intended it or not, it mirrors various scientific scales. 50 is major league average, then each 10 point increment represents a standard deviation better or worse than average. In a normal distribution, three standard deviations in either direction should include 99.7% of your sample, so that’s why the scale is 20 to 80 rather than 0 and 100. That said, the distribution of tools isn’t a normal curve for every tool, but is somewhere close to that for most.

The Basics

You’ve probably heard people call athletic hitters a “five-tool prospect.” While that is an overused and misunderstood term, they are referring to the 20-80 scouting scale. The five tools for position players are 1) Hitting 2) Power 3) Running 4) Fielding and 5) Throwing. The general use of the “five-tool” term is when all five are at least average (which is more rare than you’d think) and I generally only use it when all five are above average. It’s a shockingly small list of players over the history of baseball that have five plus tools, but if you ask around, scouts will tell you Bo Knows.

For hitters, these are the only five tools, despite many questions from readers about why we can’t expand it. Throwing accuracy is folded into the throwing tool grade (which is mostly arm strength since accuracy problems are often fixable) while fielding range, hands, instincts and all the components of defense are folded into the fielding grade. Base running skill and good jumps out of the batter’s box are also folded into the run grade. Many organizations and I will split power into game power (predicting big league power stats) and raw power (how far he can hit the ball in batting practice) but they are often the same and it’s simply a way with numbers to better explain the components of power (and also comment on the hit tool). The hit tool includes plate discipline (the most commonly asked-for sixth tool by the internet) but I’ll get more into why that is and how we can still project contact and on-base skill with one number in the article about the hit tool.

Though some teams have scouts grade each of these components, it’s the five core scouting grades that are paid attention to universally. It’s common practice in scouting reports for scouts to explain in the comments when, say a 55 fielding grade includes some 60 or higher components and some 50 or lower components, but often a 55 means a number of average to above skills and doesn’t merit much explanation. Scouts also use present and future grades for each tool. Present grades often are 20’s for high school players while, in the upper levels of the minors, the gap between present and future grades is very small. A present 20 and future 50 grade on a tool is noted as 20/50.

For pitchers, it is much more straightforward. Scouts grade each of their pitches (fastball, curveball, slider, changeup, splitter, cutter being the most common) on the 20-80 scale, then either grade the command of each pitch separately or have one overall command grade. Some teams will do grade for components of command (throwing quality strikes) with control (throwing it in the strike zone, usually closely following walk rate), pitchability (feel to sequence pitches, keep hitters off balance, etc.) and other similar things. Some clubs go so far as to have scouts grade deception, arm action (how clean/efficient/loose the arm swing is in back) and other components that the industry feels predict health. That said, the three core pitches (fastball, changeup, best breaking ball) and command are the four core grades that scouts use to make decisions and that inform the overall grade.

Objective Tool Grades

Tool Is Called Fastball Velo Batting Avg Homers RHH to 1B LHH to 1B 60 Yd Run
80 80 97 .320 40+ 4.00 3.90 6.3
75 96 .310 35-40 4.05 3.95 6.4
70 Plus Plus 95 .300 30-35 4.10 4.00 6.5
65 94 .290 27-30 4.15 4.05 6.6
60 Plus 93 .280 23-27 4.20 4.10 6.7
55 Above Avg 92 .270 19-22 4.25 4.15 6.8
50 Avg 90-91 .260 15-18 4.30 4.20 6.9-7.0
45 Below Avg 89 .250 12-15 4.35 4.25 7.1
40 88 .240 8-12 4.40 4.30 7.2
35 87 .230 5-8 4.45 4.35 7.3
30 86 .220 3-5 4.50 4.40 7.4

This is a table showing the tool grades (fastball for pitchers, hit, power and speed for hitters) that have objective scales that every scout uses to grade. These scales will vary team to team, possibly shifted one notch in either direction, or maybe separate grades for fastball velocity for righties/lefties or starter/reliever but these are essentially industry consensus scales.

An 80 tool is called 80. It’s really rare, so why do we need another name for it? 75 is almost never used because scouts will yell at you to make a choice and many don’t use 65, though it’s much more accepted than 75. These half grades like 65 and 75 don’t have separate terms because many teams use a 2-8 scale rather than 20-80 and 2-8 is the scale that was predominant when many of today’s top scouts were starting out. Now 20-80 is more commonly used, but often you’ll hear older scouts at the ballpark throwing out single numbers like 6 or 7 while we might call that a 65 here. It helps in my situation to have more numbers describe things when I’m trying to differentiate between literally hundreds of prospects that have 50 or 55 power grades, for example.

One more important addition to the scale that isn’t shown here is solid average (52.5) and fringe-average or fringy (47.5). Since so many tools fall close to 50 but you may clearly prefer one 50 to the other, many scouts will use these terms to differentiate. Again, given the thousands of players I’ll be grading, it makes sense to use this and it will show up as 45+ or 50+, since no scout has or ever will write 52.5 or 47.5 (they just put 50 then say fringy or solid average in the comments).

Fastball velocity is pretty self-explanatory and this is used as a starting point, with many other pieces of information leading to 1-2 notch moves up or down. As mentioned above, lefty/righty and starter/reliever can be taken into account (though I and many teams don’t do that, instead considering those factors in the overall grade at the end) while command, movement and deception are common other components to move up/down from the starting velocity grade.

I’ll go more into the batting average/on base/hit tool thing in the hit tool article but it seems like even the most statistically-inclined people agree this scale is kind of agreeable for what it’s trying to do. For homers, it’s a similar situation that I’ll get into later; ideally you’d like isolated power for projection purposes, but this scale works for what it’s trying to do.

For the two different run grade scale, we have the 60-yard dash, which is a combine-style showcase measure of straight-line speed akin to the 40 from the football combine while the home to first base times from either batter’s box are functional game speed. Often scouts use the raw times (comparing them with scouts nearby to verify accuracy) then round up/down based on wind/grass conditions for the 60 or how good of a jump out of the box and effort level on times to first base.

The Overall Player Grade

Hitter Starting Pitcher Relief Pitcher WAR
80 Top 1-2 #1 Starter —- 7.0
75 Top 2-3 #1 —- 6.0
70 Top 5 #1/2 —- 5.0
65 All-Star #2/3 —- 4.0
60 Plus #3 High Closer 3.0
55 Above Avg #3/4 Mid Closer 2.5
50 Avg Regular #4 Low CL/High SU 2.0
45 Platoon/Util #5 Low Setup 1.5
40 Bench Swing/Spot SP Middle RP 1.0
35 Emergency Call-Up Emergency Call-Up Emergency Call-Up 0.0
30 *Organizational *Organizational *Organizational -1.0

* “Organizational” is the term scouts use to describe a player that has no major league value; he’s just there to fill out a minor league roster and be a good influence on the prospects, though sometimes org players can outplay that projection.

** I didn’t continue down to 20 on either scale since it’s almost never relevant for players that I’ll be writing about or any of their tools, other than speed for big fat sluggers.

There have been plenty of articles here at FanGraphs breaking down this general idea and many adjacent ideas (and there will be more). I won’t profess that the scales I’m presenting here are perfect, but it’s a good combination of the objective, research-based scales and the more subjective ones that scouts have traditionally used (but are slowly becoming more objective as front offices have stat guys tweak them).

The concept that an 80 is just one or two big leaguers at each position is a traditional one while technically there could be a bunch of 7-win guys at any position. The “Top 1-2” notation for hitters is just to give an idea of typically how exclusive each group should be, realizing it isn’t always true.

Most scouts agree there are only ever 8-12 pitchers that could be called #1s or aces at any given time, but then there’s like 20 #2s and like 75 #3s. Many fans get tripped up by this term, thinking there are 30 of each type or that every team has exactly one version of each; that’s an understandable misunderstanding. Scouts see tiers of pitchers and call them #1, #2, #3 starters and this is one of those things you only fully understand when someone takes the time to explain to you what they mean.

Relievers are hard to value in this sense, as many people and scouts would say you’re crazy to not call Mariano Rivera an 80 since he’s the best ever. The problem is that assumes he’s as valuable as Mike Trout, which significantly fewer people believe, but still some people would (with some statistical adjustments for postseason leverage giving them something to point at). The WAR framework gives us a way to figure out where most players can be described and most elite relievers max out at around 3 wins, with very few racking up multiple seasons that good. You’d take a 60 position player over a 60 starting pitcher and either over a 60 reliever (all things being equal) due to attrition and these overall grades do their best to make the comparisons simpler.

The WAR-to-overall-player-grade conversion also isn’t perfect, so don’t assume someone is an 80 for the rest of time after one 7-win season by one of the WAR metrics. It’s a guide to convert a scouting grade convention for minor leaguers and amateur players to a scale that can be understood for purposes like valuing players in trades. The WAR grade here is meant as a true talent level, so crazy BABIP and UZR swings or playing time varying year-to-year doesn’t confuse us. I also may project a player’s upside or future tool grades to be a 3-win player, but the overall grade is subjective and includes various types of risk in the determination.

Many teams call their overall grade an OFP, short for Overall Future Potential. One of the clubs I worked for called their overall grade FV, short for Future Value, as that more accurately describes what this number is trying to do. The scout isn’t just averaging the core future tool grades; he’s trying to use one number to describe how valuable this player is on the overall player market, taking into account risk, distance to ceiling and other factors.



Print This Post



Kiley McDaniel has worked in the scouting departments of the New York Yankees, Baltimore Orioles and Pittsburgh Pirates and has written for ESPN, among other outlets. Follow him on twitter for real-time thoughts on the players he’s seeing and hacky attempts at humor.


Sort by:   newest | oldest | most voted
Dino
Guest
Dino
1 year 8 months ago

This is f*cking great! Keep doing these, please.

Jeremy Russell
Member
Jeremy Russell
1 year 8 months ago

And maybe also speed up with the organizational rankings. The current pace of one every 8 days would take about 7 months to finish.

Keeper
Guest
Keeper
1 year 8 months ago

Why don’t you write them yourself then Jeremy…. or actually pay his salary to do it rather than complain.

Saul Goodman
Guest
Saul Goodman
1 year 8 months ago

someone just got “lawyer-ed”

walt526
Guest
walt526
1 year 8 months ago

Awesome article–it really tied together many things that I knew about the 20-80, but was not as coherent in my head.

Do scouts have any sort of concept of an “error bar” or something? That is, is there a typical difference between what two scouts might rate a player? For example, say the first scout rates a player as a 60 power and then the cross-checker says 70 power. Is plus or minus 10 points typical? I guess another way of asking is what the ballpark (pun intended) inter-rater reliability might be.

Anon
Guest
Anon
1 year 8 months ago

Limiting the scale to 20-80 is practical because a wider range to improve accuracy for the 0.3% outside that range would add unnecessary complexity.
Just for discussion, what would be the best tool (presently in MLB or all-time) without the limit? It would be interesting to see the ratings of Aroldis Chapman’s fastball, Babe Ruth’s power, etc.

Kluber
Guest
Kluber
1 year 8 months ago

Steve Dalkowski’s fastball.

Ron Shah
Member
1 year 8 months ago

How about Billy Hamilton’s speed?

Nivra
Guest
Nivra
1 year 8 months ago

Yah, if 97 mph fastball is an 80, what in the world is Chapman’s 102? If the scale continues as listed, every 1mph is a +5 in scale, so his fastball would be a 105?

Reptar
Guest
Reptar
1 year 8 months ago

Maybe a dumb question, but why isn’t the scale just 0-60?

G. Perry
Guest
G. Perry
1 year 8 months ago

Maybe it just seems more logical for 50 to be the average than 30. Just spitballing.

Brooks
Guest
Brooks
1 year 8 months ago

The zero to sixty scale was Branch Rickey’s system.

Cool Lester Smooth
Guest
Cool Lester Smooth
1 year 8 months ago

It’s the same as the SAT scale. Three standard deviations on each side, and the mean is 50.

Johnston
Guest
1 year 8 months ago

You’re right.

Keeper
Guest
Keeper
1 year 8 months ago

Because this isn’t Road and Track Magazine.

KK-Swizzle
Guest
KK-Swizzle
1 year 8 months ago

Yes, this is very helpful, thank you!

Hurtlocker
Guest
Hurtlocker
1 year 8 months ago

Willie Mays, 80 across the board. Mickey Mantle, 80 running speed, 80 power, 50 fielding, 50 throwing. Babe Ruth, 80 power, 80 arm, 80 hit for average, 50 speed, 50 fielding. Trout, 80 batting average, 70 power, 80 speed, 70 fielding, 45 arm. This is fun.

Cool Lester Smooth
Guest
Cool Lester Smooth
1 year 8 months ago

Barry Bonds v1.0: 70 Hit, 75 Power, 80 Fielding, 80 Speed, fringy arm.

Barry Bonds v2.0: 80 hit, 80 Power, 50 (40 at the end) Fielding, 50 Speed, fringy arm.

a martsass
Guest
a martsass
1 year 8 months ago

Barry Bonds V1.0 80 Hit, 75 Power, 65 Field, 70 Speed, 40 Arm

Cool Lester Smooth
Guest
Cool Lester Smooth
1 year 8 months ago

Fair enough. I only really remember Hulk-Bonds, so I was just going based on the stat line.

a martsass
Guest
a martsass
1 year 8 months ago

I born in 1994, but I have spent many an hour trying to learn everything I could about him. I love me some 20% walk rates.

Cool Lester Smooth
Guest
Cool Lester Smooth
1 year 8 months ago

Same, actually. Are you giving him the 65 fielding as an overall outfielder, or just relative to other LFs?

TZ has him as a +15-a-year fielder through 1998, which is why I said it was an 80 grade. He also only hit .290, which is why I said 70 hit, although I agree that he was an 80 in terms of overall hitting production.

a martsass
Guest
a martsass
1 year 8 months ago

He had three or four fantastic years with the Pirates in the field, but the rest of his numbers seem much more reasonable ~4 runs saved per year. I don’t necessarily think that he didn’t field well enough to be worth 20 runs per year, but that it would be quite unlikely for to save 20 runs looking forward in those years. Maybe I’m applying too much of a knock based on the fact that a lot of the (quite subjective) observations I’ve read don’t see him as being in the top echelon, but merely very good. I’m also rewarding him in the hit tool department for the high walk and low strike out rates based on what Kiley has said above. I’m really guessing about all of this, of course, but I don’t think the hit tool could possibly be lower than 80.

jpg
Guest
jpg
1 year 8 months ago

75 power is way too high. That grade suggests he was hitting 45+ bombs every year in his younger days. I’d say 60-65 is more like it.

Jon L.
Guest
Jon L.
1 year 8 months ago

He led the league in home runs per at-bat three times before he started juicing (by my best estimate), and was in the top 10 every year.

Colin Dew-Becker
Guest
Colin Dew-Becker
1 year 8 months ago

Here’s a report on Bonds from June 29, 1990 by Larry Monroe, a scout with the White Sox. Monroe used a few more than the standardized five categories, but more or less he had Bonds in 1990 as:

60 Hit, 60+ Power, 60 Baserunning, 65 Range, 40 Arm

https://plus.google.com/photos/117390206839200135864/albums/6015353813093091729/6055466501641395762?pid=6055466501641395762&oid=117390206839200135864

a martsass
Guest
a martsass
1 year 8 months ago

Will strike out 100 times!

a martsass
Guest
a martsass
1 year 8 months ago

I suppose that’s only 17%, but still.

Powder Blues
Guest
Powder Blues
1 year 8 months ago

80 isn’t the upper limit, tehcnically – I’d suggest this:

Barry Bonds v2.0: 80 hit, 90 Power, 50 Fielding, 45 Speed, 40 arm.

vecnyj
Member
Member
vecnyj
11 months 22 days ago

If hit includes plate discipline surely that would be 85+?

Brooks
Guest
Brooks
1 year 8 months ago

I can’t think of a more important or influential baseball executive than Branch Rickey. But he did not devise the 20-80 scouting system. In fact, his system used a zero to sixty scale. The 20-80 scale was devised in the 1970s by the MLB Scouting Bureau. It’s possible that some scouts may have used a 20-80 system before the MLB Scouting Bureau was formed, but the 20-80 system did not become universally accepted until the MLB Scouting Bureau devised it and began training scouts on how to apply it.

ralph
Guest
ralph
1 year 8 months ago

Since one of the key things we can glean from looking at data is that we have to adjust for the years in which something took place, is there any particular baseline we should use? Does a plus-plus hit tool mean the player is being projected as a .300 hitter in today’s game?

In other words, seeing as how there’s much less offense today than just 15 years ago, were there a lot players with 60, 70, or 80 hit and power tools at the turn of the of century than there are now?

Cool Lester Smooth
Guest
Cool Lester Smooth
1 year 8 months ago

So, for SPs, would a good guide be:

#1: 4.5+ WAR expected every year.
#2: 3.5-4.5 WAR expected every year.
#3: 2.5-3.5 WAR expected every year.
#4: 1.5-2.5 WAR expected every year.
#5: 0.5-1.5 WAR expected every year.

Some years a guy will beat that range, sometimes they’ll fall a little short, but what they “are” is based on that framework.

Sang
Guest
Sang
1 year 8 months ago

You’re the guy who got ripped on after that McCann article.

Cool Lester Smooth
Guest
Cool Lester Smooth
1 year 8 months ago

Huh, I guess I am.

Maybe I should take it as a compliment that I made such a strong impression on Rick? Nothing I said was wrong at the time (Chris Stewart does suck at blocking pitches and throwing runners out, and McCann’s offense had historically been worth roughly 3 wins more than Stewart’s). A shitty year from McCann (in which he has still been an above average player) and an absurd, unsustainable career year from Stewart (in which he has still been worse offensively than McCann) doesn’t change what was true in November 2013.

It’s pretty impressive how much better he was at predicting the events of April through August 17th, 2014 on August 18th, 2014 than I was on November 25th, 2013, though.

I wonder how he could have possibly done it!

LHPSU
Guest
LHPSU
1 year 8 months ago

How would you grade obscure pitches when there’s very little to compare it to, or unusual pitches like knuckleballs?

SteveJeltz
Guest
SteveJeltz
1 year 8 months ago

Ben Revere was an alleged True 20 on the power scale. But maybe after this year he’s a *21*.

Keeper
Guest
Keeper
1 year 8 months ago

I never understood the ranges Present Value/Future Value. I just thought it was scouts who couldn’t make a damn decision and didn’t look particularly helpful. Thanks for giving us the Full Monty of the Scouting Scale

Andy
Guest
Andy
1 year 8 months ago

Kiley,
I know your thing isn’t major-league scouting, but here’s a thing I would find really informative: a breakdown of what a “55 hit tool” might look like at the MLB level, and a “60 command” or whatever. Something like a [these guys] have shown a [##] grade [something] tool at the MLB level, and here’s how I break down those components: [guy #1] has great bat control, can hit anything between his chin and ankles into the gap, but won’t take a walk to save his life, [guy #2} walks a lot, but can’t hit breaking pitches from opposite-handed pitchers […] but these are all guys with a “60 hit tool.” is that feasible for or interesting to you?

When I read about PSPEX, I see a lot of numbers, but don’t really understand what they mean, beyond “OMG, this guy thinks that guy’s slider is 1 StdDev above the mean,” and I’d be thrilled to have a more nuanced understanding.

Sean
Guest
Sean
1 year 8 months ago

Great stuff! It’s nice to see where the numbers come from.

I recently saw an article that assigned WAR values to each hitting tool, so you can add them up and get a simple WAR projection. Does anyone know if this has been done for the pitching tools? Thanks in advance.

peopletocakeratio
Guest
peopletocakeratio
1 year 8 months ago

this is tremendous!

i can’t wait until we get to the how pitchers are graded. maybe change-up grades will finally make sense to me; i’ve never really understood what a “good” change-up is.

as an off-speed pitch, by definition the change-up is a pitch whose speed differs (from the fastball). so is the grade based on the difference in speed between the pitcher’s fastball and their change? is there an optimal speed difference? is the grade based on the difference in movement? does the grade have anything to do with the consistency of release points for the fastball and change-up?

also, i guess a related question would be something like…other than the difference in speed, what is distinctive about a change-up? meaning, what is a change-up doing mid-flight that distinguishes it from a fastball? is it a decrease in backspin, leading to a decrease in the “rise” we’d normally see in a fastball? obviously pitchers have many different grips for a single pitch — pitch A thrown by pitcher X doesn’t necessarily behave like pitch A thrown by pitcher Y — but is there a generalization of a change-up beyond “slower than their fastball and probably drops more”?

i’m sure i’ll post this comment and stumble across the answer within minutes. won’t that be embarrassing? oh well, maybe somebody else has the same question, so this won’t be for naught.

regardless of the asinine questions i spit out above, i love the explanation of talent evaluation and how baseball people grade the skills of players they scout.

Aaron
Guest
1 year 8 months ago

Search YouTube for Alex Claudio – he’s a Rangers prospect (just called up a couple weeks ago).

While I don’t know what his change grades out at, you’ll definitely know what a good one looks like.

ramsey
Guest
ramsey
1 year 8 months ago

If the hit tool includes plate discipline, why is it defined in terms of batting average instead of on base percentage?

What would be Adam Dunn’s hit tool?

a martsass
Guest
a martsass
1 year 8 months ago

IBBs and hit by pitches maybe?

joser
Guest
joser
1 year 8 months ago

I suspect we’ll hear more about that in the promised future article about the Hit Tool.

Jose
Guest
Jose
1 year 8 months ago

Very cool!!! Keep up the good work McDaniel!

leeroy
Guest
leeroy
1 year 8 months ago

Fangraphs: now serving beer and tacos

Matt
Guest
Matt
1 year 8 months ago

Kiley – great work as expected. The best GM I’ve ever seen at analyzing hitting talent is Neal Huntington.

Bip
Member
Member
Bip
1 year 8 months ago

So 50 is average and a 10 point variation is meant to represent about a standard deviation. That makes a lot of sense. So, I understand why the bounds are 20 and 80, because as mentioned in the article, three standard deviations on either side covers 99.7% of the sample, which corresponds to 997 of a sample of 1000 people.

This is just a point of interest and not meant to be a criticism of the scouting scale. However, I would point out that at any given time, there are 750 players on a major league roster. If we’re talking about position player tools, there are about 390 position players on a major league roster at a time. 0.3% of 390 is about 1, meaning that for each normally distributed tool, we would expect almost exactly one player to be outside that range – either 3 SD’s better or worse than average. This means that a once-in-a-generation talent lines up pretty well with a “90” grade of something.

Cool Lester Smooth
Guest
Cool Lester Smooth
1 year 8 months ago

Yup. Guys like Stanton, Gallo and Hamilton.

joser
Guest
joser
1 year 8 months ago

“…other than speed for big fat sluggers.” hee hee.

Todd R.
Guest
Todd R.
1 year 4 months ago

This is really good stuff. Exactly what I was looking for

wpDiscuz