Last week, while attending a minor league game between Kane County and Beloit, someone noticed me leaning to the right after nearly every pitch to see what the radar gun held by the player tracking pitches said. “You care about how he’s pitching, or how fast he’s throwing?” asked a scout sitting near me, sarcastically. “Is there a difference?” I replied, trying to equal his snark. We both laughed, knowing that there was some merit to the question.
The next day I was talking to another scout, and we were talking about velocity when he just came out and said what we were both talking around. “I’ve more or less come to the conclusion that, as unsexy as it is, velocity means more or less everything,” he said. We talked about the rare exceptions, and how lefties aren’t always limited to same restrictions, but then we started to test our theory by naming right-handers in the big leagues who don’t have at least average velocity. Needless to say, it wasn’t a long discussion.
That conversation stuck with me for a while, and then I realized that we have the data to prove or disprove the assertion. The PITCHf/x system measures a variety of things, but at its most basic level, it is measure that core piece of scouting data when it comes to pitchers-velocity. So we now know the velocity of every pitch thrown in the big leagues, and when measuring the data against the basic theory of the scout, the numbers support his theory, and on a staggering level.
That said, obviously there are exceptions to the rule. Asserting that, if you throw hard, you will get to the big leagues is not an absolute truth; baseball has no absolutes. But the inverse-all those who get to the big leagues throw hard-that’s almost all the way there. Is command important? Yes. And secondary pitches? Absolutely important as well. But the leading statistical indicator for getting to the majors might not be measured by any spreadsheet or formula, but may instead be found on the radar gun.
To evaluate this data, just a quick step back to how scouts grade pitches. For fastballs, obviously location and movement can grade a pitch up or down, but velocity is the major factor. The standard 20-80 scouting scale, going on velocity only, looks like this:
Grade MPH 80 96+ 70 94-95 60 92-93 50 89-91 40 86-88 30 83-85 20 82-
Now, with that in mind, scouts do give half scores out, such as 55, so a quick extrapolation gets us:
Grade MPH 80 96+ 75 95 70 94 65 93 60 92 55 90.5 50 89 45 87.5 40 86 35 85.5 30 83 20 82-
OK, now let’s crunch some data. As of August 18, there were 230 right-handers who had thrown 300 or more fastballs in the big leagues. Taking their average velocity, how many have average velocity or better? Before crunching the numbers, my bet was at least two-thirds of them did. Afterwards, my mantra changed to “Throw Hard or Go Home.” Here’s the distribution:
Grade MPH RHPs 80 96+ 7 75 95 11 70 94 31 65 93 34 60 92 45 55 90.5 56 50 89 27 45 87.5 11 40 86 4 35 85.5 4 30 83 0 20 82- 0
That’s staggering. Nearly 92 percent of all right-handers have at least average velocity, 80 percent are above average, and well over half (55.7 percent) have true plus fastballs. If anything, it’s a cruel reminder that you can have as much pitchability or command as you want, but unless you are really unique, if not downright special, it’s just not going to matter unless you also throw one by a guy once in a while. For example, here are your five bottom right-handers in terms of average fastball velocity in the majors this year:
Pitcher Avg. MPH Brad Ziegler 84.6 Darren O'Day 85.2 Cla Meredith 85.3 Chris Young 85.9 Yusmeiro Petit 87.0
So, that’s three side-armers followed by Chris Young, whose height (6’11”) and deception makes him one of the most unique pitchers of the past decade, and then Petit, who has even more deception than Young. This suggests that standard-issue right-handers with over-the-top deliveries and three solid offerings have very little chance of making it without plus velocity or some kind of trickery, and this kind of data makes one a little leery of the upside of a prospect like Tim Alderson of Pittsburgh, who has put up impressive numbers in the minors while rarely getting clocked throwing anything outside of the 80s on the gun.
We can talk all we want about what kind of statistics one looks at when evaluating pitching prospects, and even delve deeper into things like BABIP, ground-ball ratios, and WARP, but when it comes right down to it, there is no evidence that we’ve found anything to replace the radar gun reading as an indicator for future big-league success.
A version of this story originally appeared on ESPN Insider .
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.
Subscribe now
If, as you say, "nearly 92 percent of all right-handers have at least average velocity, 80 percent are above average"
it would be helpful to define what you mean by average. Or at least define the set from which that average is derived. Major league pitchers? Professional pitchers? Humankind as a whole? Becaue my math says that if 80 percent are above average, the other 20% are throwing with negative velocity.
What should be of actual interest is what constitutes an average fastball, and based on this article the scouting scale does a dreadful job of doing that. If the scale says that 80% are above average, the scale needs to be re-defined.
The scouts world is populated with below-average baseball players, and his language needs the tools to describe them.
I will always remember Tanana for a great quote from about '77- when asked how he would celebrate his birthday(after pitching a gem) he replied, "I'll probably get stoned" Tongue in cheek, I'm sure...
1. There are NO absolutes, but the data is very compelling (at least it was to me).
2. Not all guys who throw hard get there, but the inverse is almost true.
One other question. If fastballs are rated mostly on velocity, how are curveballs ranked? Break? velocity? deception? feel? what!?!
Don't know why feliz isn't on the leaderboard, his fangraphs page says his average FB is 96.4.
Zumaya, Joel
Broxton, Jonathan
Bard, Daniel
Wilson, Brian
Lindstrom, Matt
Lowe, Mark
Jepsen, Kevin
Great analysis. This is definitely a cool area of discussion.
That said, he has to be in the "exception" category. I am guessing if players can set up their pitches against each other so that it confuses hitters, and can do it consistently, then OK. But those must be the exception.
Still true.
Can you make the majors as a RHP with a FB less than 89 mph? Yes - 8% means it is possible. Can you have success as a RHP with a FB less than 89 mph? Again, yes. And simply because one throws 95+ does not mean they are an elite pitcher.
So when evaluating RHP, if 92% average 89 mph or faster, then I'd say that is a beneficial characteristic to examine in the evaluation of which pitchers will make the majors and which might have the ability to stay there once they get there. It isn't a hard-and-fast rule and it isn't the only factor to success. But it is a good start.
But it is very strange to read an article on a statistically-oriented site that says something as absurd as 80 percent of major leaguers are above the major league average in velocity. Simply saying "that's the scale so accept it" is contrary to the thinking that BP applies to other aspects of the game, right?
However, I wouldn't be too quick to dismiss the concerns that your readers are raising regarding semantics of "above average."
If scouts consider a 50 rating on a fastball to be "average," and they grade a 89 MPH fastball as a 50, then its time to slide that scale. It may have been the case at some point in the past that a 89MPH fastball was 'average' for a major leaguer. However, your study demonstrates very clearly and conclusively that this is no longer the case. If you're not sitting at 92 with your fastball, then you're 'below average' at the major league level.
That is also an important conclusion to draw.
The real point is: maybe some prospect lists might have a guy ranked high where KG might have him ranked lower due to velo concerns.
Again, I'm talking about prospects.
So I think an even more interesting way to look at this would be to examine the actual 2009 results of the pitchers in each of these groups, and see if a trend emerges which justifies the bias for rostering harder throwers. While I think in general you're likely to find a strong correlation in support of your theory, I also think -- based on the groups of five "20's" and seven "80's" you identified -- that you might find some anomalies at the extremes. That is, at the top end of the scale there may be a slight dip in performance, in that the group (exclusive as it is) may include a disproportionate number of arms who are in the bigs merely because they can light up the radar gun, even though they have no idea where the pitch is going (read: Lindstrom). Whereas a 60 or a 65 heater is very good, but not so good that you can coast to the show on it without developing your control, command, and secondary offerings. Similarly, at the low end of the scale you may be more likely to find only those pitchers with true mastery of control, command, and secondary offerings.
And deception and extreme height, too, yes. But I think something like deception should be analyzed as an almost perfect substitute for velocity, or perhaps simply as a component of it. That is, the *reason* velocity is important is that it gives the hitter less time to read and react to the pitch. And deception is important for the very same reason. All that matters is that the hitter has X.X seconds to see and swing. It doesn't matter if he has this amount of time because the ball is on its way faster (Neftali), stays hidden longer (Ziegler), or is released from closer to the plate (Chris Young, Danny Almonte, Jennie Finch). All accomplish the same result. In fact, one might even argue that accomplishing it through deception is superior to doing so through sheer oomph, since all things -- read/react time for the hitter included -- being equal, harder contact (and with it, presumably a higher HR rate and BABIP) will be generated off of a 99 mph fastball thrown without deception than off of an 87 mph one thrown with it.
In any event, definitely food for thought. Though I'm worried that the more you talk about this, the more you are taking away my secret weapon for drafting pitchers in my fantasy leagues. First stop is PECOTA, but the second is fangraphs for the velo data!
I think that the next step in this line of thinking is to examine velocity of pitchers across levels and see if fastball velo is a predictor across other levels. Although I'm not sure if Pitch F/X data is available at all levels.
"[W]e have the data to prove or disprove the assertion [that 'velocity means more or less everything']."
"[W]hen it comes right down to it, there is no evidence that we've found anything to replace the radar gun reading as an indicator for future big-league success."
Except those two are actually from the article.
I'm not sure why there have not been more studies into this considering all the video resources available now to help determine the time a hitter has to react to a pitch for every different pitcher. Having that information would allow us to properly compare guys like Ziegler/Young/Feliz and it'd be a great tool for scouting since I'm sure there are guys with 93mph fastballs that play down to 90 because of their delivery and guys with slower pitches that perform better because of their deception or height.
That's why I think everyone should be rated on a major league scale. Why should I care if a guy is an above average Rookie League shortstop, I want to know how he compares to major leaguers.
It was interesting to see that velocity histogram Kevin - thanks for the piece.
Future success is predicated on a number of other factors, many of which have already been alluded to in this chat.