2012 Reds / Reds - General

Will Mike Leake Succumb To The Verducci Effect?

Mike Leake (AP Photo/Chris Carlson)

Tom Verducci released his annual Red Flag List of young pitchers who are at higher risk for either injury or decreased performance for the upcoming season. His list is developed from the premise that inning workloads for young arms should be increased gradually from one season to the next. Pitchers whose workloads are increased by too much in a season over the previous year are more likely to suffer from injury or have a setback in their performance the following year.

The parameters for Verducci’s list are fairly simple. The pitcher must be 25 years of age or younger, have thrown over 30 more innings than the previous year, and be pitching at the major league level. Minor league and post-season innings thrown are included in a player’s total innings.

Mike Leake, 23 last year, came in 11th on Verducci’s 2012 list of 14 MLB pitchers by throwing 36 2/3 more innings in 2011 than the previous year.

Most teams have employed some sort of innings limitations on their youngest pitchers, including sometimes shutting down starters before the season ends. There is no hard-and-fast rule about what the cutoff is to avoid this “Year After Effect”, and the answer certainly varies a little from pitcher to pitcher.

Still, Verducci’s Red Flag List has been right more often than it has been wrong.

The Effect has become easy to see over the years. In just the past six years, for instance, I flagged 55 pitchers at risk for an injury or regression based on their workload in the previous season. Forty-six of them, or 84 percent, did get hurt or post a worse ERA in the Year After.

While I certainly don’t like seeing a Cincinnati Reds starter on this list, I think there are several reasons not to be too concerned about Mike Leake in 2012. Thirty innings is a “rule of thumb”, and Leake was just barely over that mark. Some of those innings were minor league innings, which Verducci acknowledges aren’t as stressful on the arm. Without his 7 minor league innings thrown, Leake is just under the 30+ inning increase benchmark. The Reds were careful with Leake’s pitch count during the 2011 season. Leake only threw more than 110 pitches in a game 4 times and never more than 114 pitches. He exceeded 100 pitches in just 8 of his 26 starts.

Bigger concerns should weigh on the pitching coaches of 2011 NL Central playoff teams. St. Louis Cardinals starter Jamie Garcia had the 3rd largest increase in inning workload last year with 57 more innings thrown. Injury prone throughout his professional career, 2010 was the first time Garcia had made more than 25 starts in a season, and it’s only second time he had made more than 20 starts. Including the post-season, Garcia made 37 starts last year, nine more starts than the previous year. Milwaukee’s Yovani Gallardo is on the list for the second time in his career, this time at 7th. The first time he made the list in 2008, he missed almost the entire season with knee injuries.

34 thoughts on “Will Mike Leake Succumb To The Verducci Effect?

  1. Innings pitched is perhaps a crude stat for measuring usage, and I think total pitches thrown may be better. In 2010, Leake threw 2131 pitches, per baseball-reference.com, and in 2011 he threw 2552 pitches (not including maybe 75 pitches in the minors). That translates to about 23% more pitches thrown.

    Leake doesn’t have the violence in his motion that a pitcher like Volquez does, so I wouldn’t put too much stock in the Verducci effect for him. He may well get hurt, because most pitchers do get hurt at some point, but I think the Reds have handled him pretty well overall.

  2. I tend to agree with Big Ed. The fact that Leake isn’t a max effort pitcher would seem to be a mark in his favor. His pitch counts were low all year, and I don’t think he has been mishandled in any way.

    In general, I think the so-called Verducci Effect is overblown. Pitchers are always a pitch away from injury, and betting on a young pitcher to regress somewhat or to get injured at some point during the year is probably a safe bet 70% of the time regardless of how many pitches the guy threw the year before.

  3. Ugh. There may actually be a Verducci effect, but I’ve never seen any of his studies actually prove a thing. For instance:

    Two out of the nine pitchers I red flagged last year actually stayed healthy or improved: Gio Gonzalez of Oakland (since traded to Washington) and Ivan Nova of the Yankees. More typical, though, were the regressions last year by David Price, Phil Hughes, Mat Latos and Brett Cecil, all of whom I red-flagged

    He takes credit for identifying Price and Latos correctly?
    David Price 2010 – 208.2 IP, 3.42 FIP, 3.82 xFIP, 2.72 ERA
    David Price 2011 – 224.1 IP, 3.32 FIP, 3.32 xFIP, 3.49 ERA

    Price pitched more innings, had a better K-rate and a lower BB-rate, but because his ERA regressed, Verducci says the workload from 2010 affected him?

    Similarly, Latos pitched more innings and while he did see a slight bump in his FIP (3.00 to 3.16) and xFIP (3.21 to 3.52), it is nowhere near something to be alarmed about. But again, because his ERA rose a half a run, the innings from the previous year must be affecting him.

    There is so much going on in ERA that is outside of the pitcher’s control that it’s a terrible stat to use to make such a claim as this. And every year when Verducci releases his list, I get all worked up again. Ugh.

    I imagine the rest of his list from last year is littered with guys who he claims were good picks, but really don’t pass any real analysis standard. But then Verducci gets that:

    Bear in mind this is not a highly scientific study. It is a rule of thumb used to gauge what by now has become a kind of industry standard of trying to keep young arms healthy.

    In other words, it’s not a very good study, but for some reason people follow it anyway.

  4. Went back and looked at his predictive list before 2011 season. Here are the names:

    Better FIP in 2011
    Madison Bumgarner, SF
    Gio Gonzalez, Oak.
    David Price, TB

    No change (or within the error bars of FIP)
    Mat Latos, SD

    Fell off in 2011
    Phil Hughes, NYY
    Brett Cecil, Tor.
    Travis Wood, Cin.

    Hard to tell since they spent most of 2010 in minors
    Dillon Gee, NYM
    Brandon Beachy, Atl.
    Ivan Nova, NYY
    Alex Sanabia, Fla.

    So, by my count, 4 of the pitchers pitched well enough to either be considered to be improved or at least the same. Three pitchers clearly pitched worse (and Cecil was helped in ’10 by a much better HR/FB than his career avg). And who knows on the other 4.

    Here’s the thing too, you could include Bumgarner and Wood in that last section too since both of them pitched nearly half of their innings in the minors in 2010. How is that a sufficient sample to be able to tell whether or not a pitcher improved or fell off?

  5. @Joel: Yeah, nice post.

    Correlation is not causation. A pitcher who pitched exceptionally well in a given year is likely to pitch more innings than he did the year before. A pitcher would pitched exceptionally well in a given year is also likely to regress some the next year. His argument that he’s seen at least some regression in many of his picks in past years isn’t very meaningful. There would need to be a much more thorough study done to show that Verducci is doing anything more than picking out young pitchers who pitched exceptionally well last year, and are therefore likely to pitch worse the next.

  6. Reds pitchers most likely to get injured from increased workload:
    1. Chapman
    2. Bailey
    3. Cueto (I’m not convinced he’s healthy right now from the shoulder injury. I think it was worse than they let on).

  7. “injury” and “regression” are fuzzy enough terms that it makes it easy to be right. TJ surgery is a heck of a lot different than missing three starts in July because of shoulder fatigue or inflamation. Same with regression. Verducci’s effect only applies to guys 25 years or younger which are the same types of pitchers who tend to have a wider variance in performance from year to year.

  8. Agreed his operational definitions could use some real work, but there’s merit to the overall idea. Not too many here are saying “let Chapman throw as many innings as he can get” or they wouldn’t be talking about trading him because it will take too long to get him up to starting shape to be useful to the Reds.

  9. How did major league pitchers survive and pitch so many innings back in the golden without all this information we are overloaed with today on pitch counts and innings pitched and such…

  10. If you get more granular and look at pitches rather than inning, Leake and Jaime Garcia are pretty even as being the most efficient when looking at pitches per batter and pitches per inning. Considering that and what i believe will be improving control, I don’t foresee this “curse” being an issue.

  11. @Joel: The Verducci Effect also ignores pitching mechanics. Leake doesn’t have any of the dreaded “inverted W/V/Ls,” and is a soft-tosser so I don’t expect him to be an issue.

    The interesting thing is that Gio Gonzales is someone that has had some pitcher abuse in the past and somewhat poor mechanics. I’m glad the Reds got Latos!

  12. @dn4192: I’d like to see some data on the average career arcs of players back then. There are going to be outliers of course but it seems to me the pitchers today have a much smaller edge than before (smaller mounds, small stadiums, better athletes playing). WE NEED MORE DATA!!!! 😀

  13. 1. I found it ridiculous that Verducci humblebrags, “[An effect] I call the Year After Effect, and which some places, including internal metrics used by at least one organization, referred to as the Verducci Effect.”

    2. Leake actually pitched more innings his last year at ASU (143 IP) than he did during his rookie year. His 2011 was actually only 32 IP more than he’d ever done before. We’re only talking about 3 innings’ difference, and the whole thing is imprecise anyway, but it’s worth mentioning.

  14. I agree that his yearly SI article is pretty worthless, but I also agree with Matt that there’s clearly something to the idea.

    Does anyone know of a study of a “Verducci-like Effect” that is done in any meaningful way? Like, with a control group say?

    What is clear is that Verducci, at least in these articles, is guilty of some pretty bad cherry picking of stats.

    For example, he says Leake is a risk to break down because of his frame, and cites his W-L in the second half? When did wins and losses come back in style? Here are Leake’s ERAs by month:

    Apr: 4.40
    May: 6.94
    Jun: 2.63
    Jul: 3.90
    Aug: 4.60
    Sep: 2.74

    Does that look like a guy who was consistantly wearing down as the season wore on? Seems pretty evenly distributed to me actually.

  15. I’ve always kind of thought the Verducci Effect was an odd predictor, but over time it seems to have proven itself. Others have tried to tie it to pitches throw during a season, or ‘stressful innings’, or used some graded scale. But nothing seems to have as high a success rate and the basic old Verducci formula.

    Spooky, like a water witching or a wigi board if you ask me.

  16. @TC: Where are you getting that the Verducci Effect has been proven?

    Everything I’ve read on backtesting it has been negative. Samples:

    http://www.sabernomics.com/sabernomics/index.php/2010/02/testing-the-verducci-effect/

    http://www.hardballtimes.com/main/article/the-year-after-effect/

    http://baseballanalysts.com/archives/2010/02/verducci_effect.php

    I honestly just wonder if the Verducci Effect is just another example of “experts” making guesses and us subjectively unable to separate their “hits” and “misses”.

  17. @TC: I’m wondering what you’ve read that has backtested the so-called effect with positive results?

    Everything I’ve seen at attempts to backtest the so-called effect has been negative on the subject. I have a comment awaiting moderation with links to at least 3 articles which take a scientific approach to the subject and find it most likely does not exist. The Verducci Effect strikes me as something that sounds good, looks good, but is far too simplistic to give accurate results over a decent sample size because of the multitude of variables it doesn’t take into account. (Similar to the Dogs of the Dow approach to investing or various other gimmicky-type stuff).

    It just seems odd that MLB teams and fans have been so hesitant to adopt sabermetrics (something with years and years of objective data supporting it) but so willing to accept something so arbitrary as going beyond 30 innings/year is bad for young pitchers!

    • It just seems odd that MLB teams and fans have been so hesitant to adopt sabermetrics (something with years and years of objective data supporting it) but so willing to accept something so arbitrary as going beyond 30 innings/year is bad for young pitchers!

      Probably because it has a cooler name than wOBA and xFIP.😀 It’s all about spin.

      But again, we can see many ways in which his measures are poor… but then if one debunks the Verducci Effect are we saying there is nothing to limiting innings at all? Or just different reasons for coming to the same conclusion?

      More importantly, if the Verducci Effect doesn’t exist, do we have to forfeit the ability to call Dusty an arm busting old timer? I wouldn’t give that up for anything.

      • More importantly, if the Verducci Effect doesn’t exist, do we have to forfeit the ability to call Dusty an arm busting old timer? I wouldn’t give that up for anything.

        😆

  18. @CP: Let me look at the list I keep of every website I’ve ever looked at. 😉

    Actually, we have this discussion every year when a Reds pitcher finds his way on it. Last year it was Bailey. The year before it was Cueto. The year before that it was Matt Belisle. I read it during this threads sister thread within the past few years.

    Actually, the whole things sounds so unscientific it wouldn’t surprise me if it is bunk. I’m pretty sure water witching and wigi boards are bunk too.

    • I’m pretty sure water witching and wigi boards are bunk too.

      I don’t know… Albert’s out of the NL and Braun is suspended. I’m just sayin’ maybe somebody knows somebody with some of that juice. Don’t ask around.

  19. @dn4192: @CP: Notably, there have been studies done showing that it takes a lot more pitches to get through a game today than it used to. This has a lot to do with increased HR rates and more selective hitters.

  20. Pingback: Leake on ‘Verducci Effect’ list | Cincinnati Reds

  21. It’s good that the Reds didn’t trade for the pitcher formerly known as Fausto Carmona.

    • It’s good that the Reds didn’t trade for the pitcher formerly known as Fausto Carmona.

      Ouch!

  22. I appreciate all the comments. I will click through to your links, CP. Thanks for those!

    I don’t think Verducci’s benchmarks are in any way a hard-and-fast rule, but certainly there is merit to the premise, no? Most teams do control both innings and pitch counts of their young arms, including Cincinnati. Is the inning watch a waste, and they should focus on pitches? Or are each component overblown?

    Interesting note on Garcia and Gallardo. The Cardinals are very obviously watching Garcia’s pitch count. He never threw over 110 pitches in a game last year. In the first 2 months of the season (11 starts) he threw more than 100 pitches just 5 times…4 of those times were 102 or 103. Over the next 4 months (21 starts) he threw more than 100 pitches just three times. He also never threw more than 100 in the playoffs.

    Gallardo, on the other hand, threw 100 or more pitches in 12 of his first 13 starts. He threw 100 or more pitches 26 times in 33 starts. He topped 120 pitches twice.

  23. Mikey really strenghtened his arm last year which is why he could pitch that many more innings without getting tired. He’s been working out even more this year. I’ll predict he’ll be pitching the whole year and he’ll have a better record than last year if the Reds can put up more than three runs a game.

  24. @Greg Dafler:
    I think pitches per inning and batters faced per start should be looked at in conjunction with total pitches and the difference between Gallardo and Garcia is negligible(.3 pitches/inning and .2 batters/start)

Comments are closed.