Traditional Scouting in the Texas League Part 1
Before I begin, as always, you can reach me via e-mail at mrkupebaseball@gmail.com. In addition, in a long-overdue and much-anticipated development, I'm now on Twitter. Feel free to follow me at @mr_kupe (I think I got that right), where I'll be providing links to content and quite possibly some at-the-game observations. I'm looking forward to interacting with the prospecting community on Twitter and asking entirely too many questions about various players.
In an effort to produce more easily accessible writing, thus far in my Texas League reports I have tried to stay away from using the traditional 20-80 scouting scale. Nonetheless, the 20-80 scale is a very important aspect of my evaluations, even if it can be frustrating. Still, I've found that while numerical grades can be quite subjective, it's certainly preferable to qualitative statements on the same subjects. As an example, say that I suggest that Player A has a "good" arm and that Player B has an "really good" arm . . .it's hard to figure out what I'm talking about, as I give no basis for comparison. Using the 20-80 scale, though, I can tell you that Player A has a "55" arm (meaning above-average) and that Player B has a "60" arm (also known as plus, or well above-average). As I mentioned before, these grades can still be rather subjective; one man's average is another man's above-average, and so you can get dramatically different grades on a player even using the 20-80 scale . . .but at least you're getting more information than "good arm".
Going back to the subjective nature of numerical grades, there are a couple of things that we can address with some certainty. Batting average, for example, fits in nicely on a 20-80 scale, since we already know that the overall batting average in the major leagues is typically about .260. I prefer to use a range of batting averages, and for this I use John Sickels' suggestion that batting average can fluctuate about 20 points north or south of a player's true skill level. In other words, a 50 grade in "hitting for average" means that the player can be expected to hit between .250 and .270 in a given season. From there I add/subtract ten points for each upwards/downwards movement of 5 points on the scale; for example, a 35 grade in hitting for average means that the player can be expected to hit between .220 and .240 in a given season. The other thing that we can easily grade is a pitcher's velocity . . .but I'm getting ahead of myself here, as grading position players and grading pitchers are two entirely different things, and for the moment I'm just going to focus on position players.
In addition to hitting for average, the rest of the five tools are hitting for power, running speed, fielding ability, and arm strength. I know that some like to break these grades down further, such as grading fielding ability using sub-grades for range and hands in the case of a middle infielder, but I'm not too picky on this. I can keep that in the back of my head and write that in my comments on a player, and there's something to be said for not micromanaging the process. Sometimes a player is just bad at fielding their position and that's that. Of course, the problem with the last four of the five tools is that they're difficult to measure. How do you grade the power of a player who hits 15 doubles and 20 home runs versus the power of a player who hits 40 doubles and 10 home runs? How do you grade arm strength? Do you grade the running speed of a player in the field, out of the batter's box, or running from first to second? Unfortunately, the best answer I have here in most of these cases is: you need to watch baseball, and by this, I mean watch a LOT of baseball. Over time, you'll figure out what "average" looks like, and from there you'll be able to determine what is "above-average" and so on. Does this sound terribly unappealing in a way? It probably should; as I said before, one man's average is another man's above-average. It's this same sort of highly variable evaluation process that makes Hayden Simpson and Cito Culver first round draft picks, for better or for worse. But there is plenty of insight to be gained from even one person's subjective opinion, and if you're like most people, you'll have access to multiple opinions. Put all of those opinions together and, in the spirit of spaghetti forecasting, you'll have a consensus opinion . . .which may or may not prove to be correct. Bummer.
I originally wrote this piece at the behest of t ball, who wanted me to go into some detail on how the 20-80 scale relates to evaluation of Texas League prospects. As the Texas League is a Double A league, there are plenty of players with at least some potential to contribute to a major league team one day, and there are even a few who profile as true impact players in the majors. A good number of the pitchers have at least average major league velocity, and the ones who don't tend to grade out very highly in other areas of the game. In addition, as you may have noticed, many major league teams are jumping their players from Double A straight to the major leagues now, and some players actually do quite well. One of the biggest tricks that your mind can play on you when you're evaluating players at this level is the misbegotten impression that all of these players are far away from the major leagues in terms of skill. Many of them really aren't, even if they might need some time to adjust in the major leagues and show their true level of present skill. In A ball, with regard to hitting for average and power, grades in the 30s are quite common, and anything above that is very impressive. 20 year old Brandon Wood in the 2005 Cal League had a 50 present grade for his power, which sounds unbelievable . . .until you realize that Wood had nearly 100 extra-base hits that year, and even in the Cal League, that is just plain amazing. In AA ball, you see lots of 40s and even 45s, and the very best players will have some grades higher than that. This might surprise you, but it does make sense; if the best players in AA are capable of playing in the major leagues at this very moment, then they almost certainly have some grades that are major league average or above. This gets a little trickier when it comes to pitchers, but again, for the moment I'm just focusing on position players. While I'm not focusing on Triple A here, you'll see a lot of players there with plenty of major league average grades. Again, this makes sense - there are lots of players in AAA who are on the fringes of the major leagues but don't quite have the full package needed to stick in the majors. In other cases, you'll find some prospects who are getting that final preparation before a major league career.
To return to the Texas League, while I've spent a lot of time and text getting to this point, the formula for grading and projecting a player on the 20-80 scale is actually quite simple. You have the five tools: hitting for average, hitting for power, running speed, fielding ability, and arm strength. Put a number next to each of those tools from 20-80 (stick to base 5), take the average of those five numbers, and you get what is known as an OFP (Overall Future Projection), which is a simple snapshot of what kind of prospect you're looking at. I just use the traditional scale, which is:
80-65 - Star
64-50 - Regular
49-40- Fringe
38-39 - Organizational
Sounds simple, right? Well . . .kind of. OFP isn't the be-all and end-all when it comes to prospecting, but I do find it can provide surprisingly useful results even when you wouldn't necessarily expect it to. Just as an example, first basemen (who generally have poor arms and running speed) get put on equal footing with defensive whizzes at shortstop (who grade out quite well even if they cannot hit very well). You may find some variations in the formula, but for myself, the only adjustment I care to make is when it comes to catchers, for whom running speed makes very little difference. I personally prefer to not incorporate running speed into the OFPs of catchers, as it also helps to adjust for the positional value of a catcher, but to each their own.
You'll also notice that OFP doesn't incorporate intangibles or plate discipline, both of which are important to the overall equation for eventual major league success. I handle this by allowing some leeway in the overall OFP score for intangibles, and by including plate discipline issues in the overall writeup of a player. Plate discipline is one of the hardest things to project, because it has much more to do with the mental rather than physical aspects of the game. Nonetheless, I do recognize that in more extreme cases some flexibility is useful here as well. I wouldn't adjust OFP more than a point or two in either direction based on these things, and even then only in rare cases. It's a good formula, and I have no intention of breaking it.
Let's go a little more in-depth by looking at a prospect that I haven't written up yet because I haven't seen him enough, San Antonio SS Drew Cumberland. These are preliminary grades, but I would be surprised if any of them moved more than 5 points in one direction or the other. The focus here is the process rather than the result.
Drew Cumberland, SS, San Antonio Missions
Scouting Grade Report - Preliminary
Hitting for average: 40/60
Hitting for power: 35/40
Running speed: 70/70
Fielding: 45/45
Arm strength: 50/50
OFP: 53
There's a lot of numbers here, so I'm going to go through everything step-by-step to demonstrate the utility of the 20-80 system as a tool for both present and future evaluation. I'll actually begin, however, with the OFP. Cumberland's OFP of 53 might not sound terribly enticing at first, but it actually suggests that he projects to be a major league regular with some margin to spare, which would place him among the upper tier of Texas League prospects. You can actually take the summation of these grades a bit further by looking at the current rather than future grades - Cumberland's OCP (or Overall Current Projection) as a major league player is 48, placing him as a fringe major leaguer and most likely a bench player at present.
By this point you're probably saying: wait a second, Cumberland was just promoted from the California League, and you're telling me that he could be a major league bench player NOW? I am, but that's not really something to get too excited about. Cumberland's present hitting for average grade corresponds to a batting average range between .230 and .250 in the major leagues. This might sound very high for a player just arriving in Double A, but keep in mind that Cumberland hit .365 in Lake Elsinore in the California League this season, and the major league equivalent of this level of performance in San Diego is actually about .270. I don't think he's quite that good of a hitter (you'll have to wait for the writeup to hear why), but as one of the best hitters in the Cal League this year, it's not hard to believe that he's an average Double A hitter, and so everything comes together quite nicely, and I haven't even gotten into the qualitative analysis yet. Cumberland's value as a major league player at this point would be as a no-hit middle infielder with fringe-average defense and good speed on the bases . . .in other words, not a player that most teams would have much use for in the major leagues. And if he was a better player at present, then he probably wouldn't be in Double A. Notice that he only has one tool at present that is better than major league average, so no worries, I'm not suggesting that Cumberland should be up with the major league team next week.
While you can expect the hitting grades to increase over time, the running speed and arm strength grades are relatively stable. You might see a player slow down a bit as he gets older and fills out, but it depends on the body type that we're talking about here. Cumberland doesn't look like a player who will lose much if any speed, and he actually looks to be better physical condition than he was in the video I've seen on him in the past. There are some cases in which a player might add arm strength, but Cumberland most likely won't - his arm strength has remained consistent since high school, and he's getting absolutely everything out of his arm that he can at present.
The last grade I wanted to address is the fielding grade, and the reason for this is because positional changes can have a significant impact on OFP. An exceptionally poor fielding third baseman who becomes a league average defender at first base could see his OFP jump, although there he'll have to contend with other first baseman who might match his fielding and out-do his hitting. Again, the qualitative analysis is very important here - what kinds of things make a poor fielding shortstop a good candidate to move to third base, second base, or center field, for instance? Cumberland is an interesting case, as I'm expecting there to be a vigorous debate about whether or not he should play shortstop at the major league level, and my personal opinion is that he's right about at his ceiling as a fringe-average defender there. In this case, he'd profile as a offensive-minded starting shortstop with passable defense. If he were a second baseman, I'd have no problem projecting him to be a plus defender there, and perhaps even better than that in the mold of a Dustin Pedroia. Now, what does that do to his OFP? Well, if we replace that 45 for fielding ability with a 60, then Cumberland's OFP suddenly jumps to 56. If his defense gets any better than that, or he ends up being a better hitter than currently projected, then we're now talking about a potential above-average regular who could end up in some All Star games. In other words, he's a good looking prospect, as you would expect from a former supplemental round pick who is enjoying a breakthrough season.
Trying to separate your perceptions of a player in the present from your projections of him in the future is insanely tricky work, and the real scouts should receive no small amount of credit for staking their professional futures to such endeavors. I could easily spend ten more pages writing solely about the qualitative analysis aspects of scouting (also known as "so how did you get to that 40 hitting for power grade, anyways?"), but that's an entirely different beast. For now, I'm just going to open the floor to comments and/or questions.
As a final note, you might have noticed that this article only focuses on position players. Don't worry, I didn't forget about those other guys, but I think this piece is long enough. Stay tuned for Part 2, in which I jump into the delightfully crazy world of evaluating pitching prospects on the 20-80 scale.
13 comments
|
2 recs |
Do you like this story?
Comments
Something about the 20-80 scale
is that it assumes each of the tools are equal to each other in importance, and arm strength really isn’t. It is a portion of the equation in fielding, but when you take the OFP it can be raised or lowered by arm strength grades that mean little. It has a hand in determining what position one will play in the future, and if someone has a special arm (in either direction) it can have an effect, but not nearly as much as having a special power bat (or a bat without any power at all). The 20-80 scale kind of makes me shrug my shoulders for that fact.
I prefer writeups, and I can’t wait to read yours on Cumberland, though. Always do good work
yeah, that is something to keep in mind
But there are definitely times where arm strength comes into play. When you watch a lot of major league games, you’re spoiled because the players with subpar arm strength tend to get weeded out. It definitely affects the projection on a guy like Cumberland.
But you do bring up a good point, and that’s why the OFP has such large ranges. The difference between an OFP 53 and an OFP 55 guy isn’t all that big, and in fact you might prefer the first guy because he has tools that appeal to you. The point here is that both project as major league regulars. When you start getting to differences of 5+, then we’re starting to talk about some dramatic differences that you should take into consideration. Is it a perfect system? Nope. Can we make a system that’s more comprehensive and explains more? Certainly. But the intuitive nature of this system makes it easy for us to work with, and gives us a foundation to go into more detail. Some may wish to weigh the bat-related tools more heavily, but that’s actually not the direction that a lot of teams appear to be headed in right now.
My example that I think proves the predicament well
We have Prospect A who is a SS, but projects a 50 fielding there, but has a 70 arm (I realize someone like this might move to 3B, but for the purpose of this exercise he sticks).
Prospect B is a 2B who projects for identical offensive tools, but a 60 fielder with a 40 arm.
At 2B, where arm strength is less of a necessity, the real important number should be the second baseman’s fielding tool, and despite the 5 run advantage we give to Prospect A for being a SS, this second baseman should total out to about the same major league defensive player, yet, in OFP, Prospect A is a better prospect (not by much, but by 10 in the defensive department, where it might not be warranted).
Bryce Harper has a 70-80 arm, Carl Crawford has a 40 arm. Crawford has an 80 fielding tool, lets say someone like Harper has a 55. By this system, Harper is clearly the better defender than Crawford, but that is absurd. Even at a position where arm strength supposedly matters (Corner OF) it seems to skew results too much for my liking.
Its less about Bat > Glove as much as Everything > Arm.
actually
After thinking about it a bit, I think I agree with you to the extent that OFP can be a little difficult to understand from the fan perspective. What I’m going to try to do is work on a similar but revised system that more closely matches major league outcomes, and until then rely more on just explaining what kind of player you can expect the prospect to be. So I may hold off on Part 2 of this article (depending on whether people want this or more reports/grades) until I decide where to go with that.
I suspect I’ll have a Cumberland report sometime in the area of a week from now after I’ve gotten more of a look. He’s a pretty good prospect but I did have some concerns about him both at the plate and in the field.
I know some people don't like them, but I really like comparisons. I feel it helps rationalize the grades a little bit
You can also articulate more objectively things that are difficult to explain such as the way you feel a player will age. Also, any projection that is ever made is only validated by past examples in which that outcome was achieved. This discussion on the 20-80 scale was never really meant to become a big thing, or a critique on the work in any way, just a point to be made.
oh, totally aware
I’m definitely thinking through my process anyways, as I’m trying to get a good feel for what people want in the future.
Comparisons are dangerous. They drive unfair expectations. I usually consider comparisons in terms of “who does this player remind me of,” not “well this guy was 21 and in High A while that guy was 22 and in Low A . . .” . . .it’s got little to do with numbers, or stage of development.
right. Its less about the path of development than where you feel the final result would be
if I feel someone like Jiovanni Mier will put up above average defense with low double digit power and speed numbers and compare him to Mike Aviles then it kind of contextualizes the prediction. I don’t actually think that is an accurate comp, mind you, but the process is there.
On the 20-80 thing
FYI, for those that don’t know, the 20-80 thing is actually an application of the normal distribution (as is the scoring of individual sections in the SAT). This style of scoring system is designed to mirror the fact that the vast majority (99.8%) of a population can be broken down into six standard deviations. 50 (or 500) was chosen as the mean because it is a logical “middle,” and each 10 (or 100 points for the SAT) represents a standard deviation. In the case of the 20-80 scale, 50 is supposed to represent Major League average.
Of course, the scouting scale is pretty subjective and I don’t think most scouts assign their scores based standard deviations and the normal curve. I seriously doubt any are thinking to themselves “this guy’s arm is better than approximately 84% of all MLB SS – which equates to a grade of 60.” But I do think it’s interesting to understand the theory behind the scoring system. FYI for those that are wondering, here is a general breakdown of what the scores SHOULD represent based on the normal distribution:
20 – Tool is better than approximately .1% of MLB players
30 – 2.2%
40 – 15.9%
50 – 50%
60 – 84.1%
70 – 97.8%
80 – 99.9%
right
The system might be founded in sound mathematics (very objective), but scouts tend to use it in a different way, at least to my impression.
We’ve definitely seen some grade inflation being floated about lately, too. Look at the Baseball America Top 100 list from last year, where there were tons of 70s and 80s being tossed about. The reality is that a lot of those grades were being grossly exaggerated. If those grades were accurate, we would have been looking at the best prospect class of all time. Something that people might not realize is that not only is it incredibly rare to find something that grades out as an 80 (current or projected), but it’s also pretty uncommon to find something that grades out as a 70. Even really good prospects often don’t have a 70 tool, and sometimes you find 70s in players who have absolutely nothing else to offer.
I think this is a great point
Quite frankly, the scale is meant to be relative to major leaguers, yet I see many many many more 50+ grades than lower than 50 grades. That shouldn’t happen, the distribution should be equal. Not every 2B has an arm that approaches “major league average” across all positions, that is preposterous.
Despite the fact that grades are usually inflated, I feel that if a grader is consistent in that 50s in one area mean the same projection time in and time out, then I’m ok with it, even if it is divergent from the original purpose of the scale.
arms
Lots of 2Bs are former SSs, so you see a wide range of arms there. Arm strength is tricky, because a player having a 45 arm doesn’t mean he won’t be capable of making a throw from SS to 1B . . .it just means that he won’t be able to do it as effortlessly or quite as effectively as a player with a 60 arm.
I agree, but what you're saying is reflected in the math above
Moving from 50 to 60 on the scale represents a huge jump from right at average to being better than 84%. The 20-80 scale is really a thing of beauty in that sense. If you think of 60 as being just 10 points higher than 50 it doesn’t really convey the difference. There may be some grade inflation from time to time but in general I think scouts are pretty stingy with above average grades.
Freude schöner Götterfunken, Tochter aus Elysium, Freude!

by 


















