The Best and Worst Fastballs, and Hard and Weak Flyballs

Justin Verlander has one of the best four-seamers in the American League.

In one of my last articles, I found that pitchers who got a lot of popups were pretty likely to outperform their xFIP by quite a bit. That isn't too strange, as xFIP counts all flyballs as bad, even though popups are easy outs and never go for home runs. So if a pitcher has a way to induce a lot of weak flyballs, xFIP will underrate him. But how does a pitcher induce weak outs in the air? The only pitch that seems like a good bet is the four-seam fastball, the one pitch that is meant to fool the batter by ending up higher than the batter expects. It's also the pitch that is used most often, and therefore put into play most often, for most pitchers. It also accounts for the highest percentage of flyballs among pitches, so logic seems to dictate that if there's a skill to inducing weak contact in the air, it's most likely to come from four-seam fastballs.

To test this theory, I used the same method that I used for the previous article on popups and line drives. I looked at the pitch values for pitchers who have pitched at least 700 innings since the start of pitch value data, which is in 2002. I then looked at the pitchers who had a fastball value of 0.5 runs per 100 pitches better than average, which gave me 18 pitchers with very good fastballs. I then looked at their ERA-, FIP- and xFIP-, which are all stats that have been corrected for league and park. We'll see if the values match, or if the advanced stats are off again, like was the case for pitchers who induce a lot of popups.

The Best Fastballs

Name wFB/c ERA- FIP- xFIP-
Roger Clemens 0.85 73 75 80
Chris Young 0.78 90 102 106
Johan Santana 0.78 67 78 80
Jason Schmidt 0.77 82 78 88
Curt Schilling 0.77 78 68 72
Clayton Kershaw 0.77 73 77 85
Roy Oswalt 0.75 77 79 84
Matt Cain 0.75 80 88 101
Cliff Lee 0.73 84 84 91
Jake Peavy 0.68 87 86 83
Tim Hudson 0.65 78 89 87
Randy Wolf 0.63 97 105 103
Carlos Zambrano 0.61 81 90 96
Chris Carpenter 0.58 78 82 81
Josh Johnson 0.58 70 73 82
Justin Verlander 0.56 80 80 88
Chien-Ming Wang 0.56 95 93 96
Pedro Martinez 0.50 75 73 78

Averages: ERA- 80.3, FIP- 83.3 , xFIP- 87.8

Analysis

The pitchers with the best fastballs surrended only 80% of the runs an average pitcher would allow in the same league and ballparks. In a run environment where an average ERA is, let's say, 4.30, these pitchers would be expected to have an average ERA of 3.45. However, xFIP thinks they should allow 88% of the runs an average pitcher would allow, which would mean a 3.78 ERA. So the xFIP formula underrates the fastball pitchers' ERA by 0.33 points on average, which is definitely significant enough to make us question the validity of xFIP in certain cases.

Perhaps you're now wondering why Curt Schilling and Jake Peavy would underperform their xFIP-s? Both have played in some extreme parks. Peavy pitched mostly in Petco (extremely hard to hit home runs in) and a bit in U.S. Cellular (very homer prone), and in addition he was from his best up until now in the latter. Schilling played most of his post-2002 period in Fenway, which usually suppresses home runs (turns them into doubles), but it can turn high flyballs, not even hit that hard, into home runs that would've been caught in a bigger park. In spacious Petco, pitchers - especially right-handers - don't have to induce weak flyballs to keep them in the park, which could mean that an ability like that is "wasted" somewhat in Petco. It's also possible that Petco Park simply boosted Peavy's fastball value a whole lot, and that he shouldn't be on the list. Now that he seems healthy, we'll see what his fastball can do in a much more difficult park.

Follow me after the jump, because there we'll look at the pitchers with the worst fastballs.

The Worst Fastballs

Name wFB/c ERA- FIP- xFIP-
Ramon Ortiz -1.36 116 120 109
Scott Olsen -1.23 113 112 105
Zach Duke -1.15 109 101 101
Kyle Davies -1.14 128 112 113
Josh Fogg -1.08 117 115 112
Jeff Weaver -0.91 110 102 103
Jason Jennings -0.90 105 99 107
Jeremy Bonderman -0.90 112 98 93
Brett Myers -0.87 101 104 92
Bronson Arroyo -0.82 96 105 105
Kyle Lohse -0.75 106 102 105
Carl Pavano -0.75 103 98 96
Ryan Dempster -0.75 105 96 97
Ian Snell -0.75 112 106 103
James Shields -0.70 95 94 85
Mike Maroth -0.67 112 113 106
Kelvim Escobar -0.66 82 81 90
Edwin Jackson -0.66 103 102 103
Ricky Nolasco -0.65 108 91 88

Averages: ERA- 107, FIP- 102.7, xFIP- 100.7

Analysis

Again, xFIP doesn't accurately predict the results for these pitchers. While the good fastballs were hit in the air less hard than normal fastballs, the worst fastballs were probably hit quite a lot harder. And while a crappy slider can be scrapped for better pitches, the fastball is quite hard to replace. Interesting to watch in this regard is James Shields, who dropped his fastball usage from 46% in 2010 (already low, around 60% is average) to 36% in 2011 to 26% in 4 starts so far this season. Bronson Arroyo is doing the same thing, with good results so far, after a horrible 2011 season. I wrote about both of them in December, when I first openly postulated that there may be a link between fastball quality and HR/FB%. Of course, from this table you can see that Arroyo has actually been the most notable exception to the trend, as Kelvim Escobar severely underperformed his FIP (no xFIP available for those years) in the years prior to 2002.

There are a number of factors that could lead to Arroyo's stubborn underperforming of advanced stats despite the bad fastball. One is his home park, which is a home run haven almost regardless of who is pitching. Second, Arroyo has a really, and I mean really, diverse arsenal which possibly includes a four-seamer, two-seamer, cutter, one or two changeups and two to three breaking balls. The third reason is that Arroyo often pitches "backwards", using his fastball more in strikeout situations rather than when behind in the count. And then there's the guy's funky leg kick, different release points and wacky horizontal curveball which gets flyballs instead of grounders. I think it's safe to say Arroyo isn't your typical bad fastball pitcher. He's an artist, and sometimes, a cook too.

Conclusion

While I´m hesitant to say this is `proof´ of a relation between fastball quality and giving up more or fewer home runs than you´d expect (using xFIP, anyway), I myself am pretty sure there is one. What this means for the Jays' pitchers is uncertain, as none currently in the majors have had clearly great or bad four-seamers. But that one guy currently struggling in New Hampshire, yes that's you Brett Cecil, has a big hurdle to overcome. Even before his velocity dropped even further this spring, he was already at a -0.82 runs per 100 fastballs, which would've tied him with Arroyo if he had pitched enough innings. I can't imagine that his fastball would've gotten better with the velocity drop.

Will Drabek, Hutchison and Alvarez have good enough fastballs to survive the AL East? Only time will tell.

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join Bluebird Banter

You must be a member of Bluebird Banter to participate.

We have our own Community Guidelines at Bluebird Banter. You should read them.

Join Bluebird Banter

You must be a member of Bluebird Banter to participate.

We have our own Community Guidelines at Bluebird Banter. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9351_tracker