Worley:Looking::Beachy:Swinging
The Phillies entered the season with the makings of an historical rotation. With Roy Halladay, Cliff Lee, Cole Hamels and Roy Oswalt, their starting pitching was covered. The Braves entered the season with ridiculous pitching depth. They had a rotation filled with good major leaguers, another solid starter recovering from surgery, and some of the most renowned pitching prospects in the game waiting to make a dent.
It’s funny, then, that the performances this season from Vance Worley and Brandon Beachy — two prospects without much minor league fanfare — have helped elevate each rotation to another level.
Worley and Beachy weren’t exactly afterthoughts, but neither was expected to be a key contributor this season. In fact, Worley started the season in the minors and spent even more time on the farm when both Oswalt and Joe Blanton returned to the rotation. Fast-forward to early September and both pitchers have tallied 2.2 WAR in under 130 innings. Both pitchers are also making a big case for their inclusion in the playoff rotation.
Despite these similarities, the major difference in their production makes a comparison interesting. Worley relies on the called strike while Beachy has become a master of the whiff. Though each is fairly inexperienced as the final month of the season pushes on, their different approaches invites a discussion on the sustainability and predictive value of called and swinging strikeouts.
First, some numbers to set the scene. Worley has thrown 110 1/3 innings this season with a 7.8 K/9, 2.9 BB/9 and 2.85 ERA. Given his decent groundball rate and low rate of homers allowed, he has a solid 3.58 SIERA. Worley has also shown tremendous improvement throughout the season, with xFIP splits decreasing successively from April-August: 4.80 to 4.36 to 4.12 to 3.46 to 2.90.
After two September starts he furthers his improvement with a 3.30 xFIP. Something clicked right around the midway point resulting in more strikeouts and fewer walks. He walked 18 in 42 innings over the first half and 18 in 68 1/3 innings since then. His K/9 rose all the way to 9.9 in August and remains around that mark this month.
The cause of his improvement is a dynamic two-seam fastball that catcher Brian Schneider decided to start calling. Schneider has essentially become Worley’s personal catcher and the two-seamer was rarely thrown prior to their battery marriage. Worley seldom threw the pitch in the minors and not until Schneider decided it worth throwing did the rookie develop the pitch. Given the nature of two-seam fastballs, a high percentage of Worley’s strikes thrown and strikeouts are of the ‘looking’ variety. Measured as a percentage of total strikes thrown — not total pitches — Worley has a swinging/looking breakdown of 9 percent/34 percent. The league average swinging strikes/total strikes rate is 15 percent, with called strikes/total strikes at 28 percent.
Contrast that with Beachy, who has emerged as a strikeout machine with a wicked swinging strike rate. He also has one of the biggest gaps between his xFIP and SIERA — xFIP has him at 3.25 against his 2.86 SIERA. Beachy’s called strike percentage is right in line with the league but he has performed well above average in terms of swings and misses. His overall strikeout rate — 10.1 per nine — is better than Worley’s for sure, but his command of four different pitches results in a much different approach on the mound. Worley relies on the movement and deception of one plus-plus pitch still in its infancy while Beachy gets called strikes but also fools hitters into empty swings.
While swinging strikes are seemingly more telling of pitcher success, they actually don’t predict future strikeout rate any better than called strike percentage. A study from Matt Swartz confirms this. Swartz looked to see whether swinging strike percentage provided more predictive value than strikeout rate alone and found the result lacking in significance. Further, he found that ‘called’ strike percentage predicted future strikeout rate in practically the same manner. ‘Swinging’ strikes had a slight advantage, one worth noting but not indicative of the idea that Worley’s success is less sustainable because batters don’t swing and miss as frequently.
Both pitchers will combine to serve as an interesting case study as their careers progress. Can Beachy sustain his gaudy strikeout rate without a real plus-plus pitch? Will the league catch up to Worley’s two-seamer? These are questions to keep in mind even after acknowledging Swartz’s study on the predictive value of strikeouts. Sometimes looking beyond overall strikeout rate paints a more accurate picture of a pitcher’s success, and this is one of those times. Both Worley and Beachy are having great seasons with good strikeout rates and different approaches.
Eric is an accountant and statistical analyst from Philadelphia. He also covers the Phillies at Phillies Nation and can be found here on Twitter.
I’m lucky to own both in my deep dynasty N.L. only league.
You certainly are! Great foresight!
ZZZZZZZZZZZZZZzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
Banality-triggered narcolepsy?
I am guessing this might be Mike Newman of Scouting the Sally. Both Worley and Beachy pitched in the South Atlantic league briefly, in 2008 and 2009, respectively. Before they were on most prospect service radars.
You’re awesome.
You’re snide.
That makes two of us. Who do you guys think has a higher upside? Either a potential frontline guy (1 or 2)?
Beachy has higher upside. Neither is a frontline guy, though.
“an historical”? Questionable, Seidman, very questionable.
Great piece, though. Swinging v. looking strikes me as one of the few areas that has yet to have trained on it the sabrmetric klieg.
Actually, as an historian who also majored in English, I can tell you that there are times when words like “historian” and “historical” properly take “an” rather than “a.” It has to do with stress patterns in the sentence.
You might very well be a historian, but “an” is not used in that sense. Now, if you wait an hour, you’ll find that I am an honorable person, and will listen to an honest reply; as you may use “an” before a soft/silent H, but not a hard H.
Any idea on how many batters faced it takes before strike rates are predictive of future success?
I’m guessing, from just a quick assay of it, that the N was so high as to approach a limit that they will never have significance. At a certain point N approaches infinity.
That seems counter intuitive. Because N is so high shouldn’t that mean you have more data to use as a way to predict future results based on past history?
That’s correct, Domingoes. Which is why after more than 100 years of baseball, bookmakers in Vegas are able to predict the winner of the World Series every year in spring training.
That isn’t how probability works.
I meant more the difference in significance between statistics, say the, using the difference between swinging and looking as a predictive statistic. At a certain point, you can’t iron out bad F’s, z’s, or t’s by increasing your N.
I suspect that one average pitcher talent level typically changes before strike out rates have predictive value.
Ahem, that’s “on average,”
Very interesting story, I’m intrigued by both pitchers. FWIW, it might have been worthwhile to include some of Beachy’s stats to make the direct comparison – we’ve got all the pertinent numbers on Worley, but the only actual stat we get on Beachy is K/9 with a vague reference to his swinging/looking strike percentages.
Can you link to the source where you mention that throwing the two seamer was Schneider’s idea and that he was the one who helped Worley develop confidence in the pitch? I had not heard that before and would like to read more on it.
He won’t link the source because there is no source.
It’s made up information.
Worley wasn’t throwing the two seamer more, but it has nothing to do with “it being Schneider’s idea and he helped Worley develop it”
Once again, another case of an author talking out of his arse.
Must be nice to be so sure of oneself. You’re wrong of course, but I’m curious how you were so sure of yourself you felt it was worth posting such an obnoxious comment.
“All six of the looking strikeouts were on the two-seam fastball, or sinker, a pitch Worley said he would throw “maybe four times” a game in the minors. Brian Schneider, his personal catcher, has called the pitch more often with lefthanders at the plate. It has created success.
“I think Schneider has a way of setting it up where we’ll go four-seamers into lefties to get them thinking it’s straight,” Worley said. “Then I’ll throw the two-seamer at them.”
–Matt Gelb, Philadelphia Inquirer (9/1/11)
The source is the Phillies broadcast. Announcers Tom McCarthy and Chris Wheeler have mentioned it in each of Worley’s last two starts. Schneider noticed the movement in a warmup session and decided to start calling it more often.
What’s the pitchFX data say, especially on the relative movement on Worley’s fastball. Also are there studies showing a correlation in ERA, strikeout rates, swinging strikes, called strikes, FIP, with fastball movement?
Eric is spot on about that. Schneider has been Worley’s designated catcher most of the year. This pitch was not something I saw Worley throw last year. It is a key weapon for him this year… relying on it more and more as the season has progressed.
When he is getting the call. It’s filthy. Knee high, just off the plate, then late break and tails back over the corner. Strike. Inside corner to LHB. outside corner for RHB.
Something about Worley feels unsustainable to me though. I keep waiting for the BB rate to spike up a bit, and especially for the HR rate blow up on him. But it hasn’t happened.
He has located his pitches well all season. Good command. Bulldog temperament, but fun. Easy to root for the guy.
I still feel like Beachy has more upside. He has ace level strikeout rates already. If he can get a little better control over the HR rate, he could be very solid. Already though, he has similar results to Worley, and has some area where he might improve some. I just always have the feeling that Worley is maxed, and is walking a tightrope… and one misstep and he’s gonna take a tumble.
SAT flashback.
What does that mean?
The title of the post ‘Worley:Looking::Beachy:Swinging’
includes the analogy marks (the colons and double colon)
which in SAT parlance means Worley (is to) Looking (as) Beachy (is to) Swinging
perhaps can look at both Worley and fellow called-strike machine, Doug Fister in a future article as well
Use K% and BB% instead of X/9.
A guy on Phillies blog thegoodphight.com has examined Worley’s in-game patterns and found that he tends to get hit pretty hard the 3rd time through the lineup – it’s part of the reason he’s pretty much a 6-inning pitcher at this point. I have no idea whether this anecdote means anything, but it seems plausible to me that this might be typical of pitchers who depend heavily on backwards Ks.
He has upped his ip/gs to 5.9 this season, from 5.0 last year. I think its more of pitches thrown issue. He has gone 100+ pitches in 11 of 19 starts this season. 80-99 in 6. and less than 80 only twice. In 2 starts last year, he was less than 80 pitches in both.
Beachy is actual quite similar, though a tad less than Worley. 100+ in 8 of 22. 80-99 pitches in 13. less than 80, once. He has averages 5.7 ip/gs this season, after averaging 5.0 in 3 starts last year. (and less than 80 pitches in all 3)
So, again, very similar.
Just wondering where you get some of your data from.
For example, % swinging strikes & looking strikes as a % of total strikes not pitches. Where do you get this info?
Regards.
Not sure where the author got his info, but it is available on BBref.
This is the NL 2011:
http://www.baseball-reference.com/leagues/NL/2011-pitches-pitching.shtml
Worley actually is leading NL qualifiers in looking strikes/strikes, while Beachy leads all NL qualifiers in swinging strikes/strikes.
I have been following these two all season. They are an interesting comparison.
Good article Eric. Thanks.