Mike Trout: Reasons for Caution

A couple of days ago, I used comparable players to show how much regression should be expected from Mike Trout. Today, I am going to dive into Trout’s numbers from last season and see if any red flags exist.

Batting Average
2012 Value: .326

An AVG is driven by two components, BABIP and strikeouts. The 21-year-old’s BABIP was the third highest value in the majors at .383. By looking at his batted ball profile, his xBABIP stood at .356 or about 30 lower than his BABIP. Trout was a little lucky with balls in play over the course of last season.

Besides his full season luck, Trout saw less of the balls he put into play go for hits as the season progressed. Through July 31st, Trout had .353 AVG and a .402 BABIP. Over the rest of the season, he hit .287 with a .353 BABIP. The reason behind the drop in BABIP was that he quit hitting line drives and instead started hitting ground balls

The drop off in BAPIP, and therefore AVG, could be for several reasons. First, he could have tired as the season dragged on. He is young and it is long season.

Also, pitchers and fielders would have more information on him such as what pitches he struggles with (split finger fastballs for the few he saw) and location (inside pitches). Here is his Run Value heat map from the 2012 season (umpires perspective). The right-handed Trout is above or at league average on all pitch locations except inside pitches.

The interesting part is that pitchers didn’t attack the inside pitch as the season went on. Instead, pitchers pitched Trout higher and more outside.

Pitches in the top half of the strike zone
Mar through July: 49.9%
Aug through Oct: 54.8%

Pitches on outer half of the plate
Mar through July: 49.8%
Aug through Oct: 52.6%

Besides going up and outside to Trout, the pitches he saw changed. Downward breaking balls (change, curve and 2-finger fastball) were thrown to him less and cutters and four-finger fastballs were thrown to him more.

FT CU CH SL SI FS FC FF
Mar through July 16.4% 7.8% 6.8% 15.9% 9.7% 0.7% 4.6% 38.1%
Aug through Oct 11.0% 7.3% 6.5% 15.6% 9.8% 1.7% 7.0% 41.1%
Change -5.4% -0.5% -0.3% -0.3% 0.2% 1.0% 2.4% 3.0%

Usually downward breaking balls will induce more ground balls. This generalization was not the case with Trout because as the number of sinking pitches he saw decreased, the number of ground balls he hit increased.

Finally, he could have been hampered by a left knee contusion he suffered on July 29th. Usually, an injury brings down a player’s power, but it may have caused Trout to alter his approach and hit more ground balls. I looked to see if his swing changed from before the injury to after it. Here are 4 .gifs of his swing throughout the season.

His 1st Game in 2012

Game Before Injury

Game After Returning From Injury

Last day of the season

Looking at the videos at regular speed and frame-by-frame in a .gif viewer, I could not find any change in his swing because of the knee injury.

Besides the pitches having a possible effect on Trout’s batted balls, they also affected his plate discipline over the course of the season. His K% was 19.5% before the injury and then it jumped to 24.9%. On the other hand, his non-intentional walk rate dropped from 8.4% to 11.9%. A 5% increase in a strikeout rate would put a drag on anyone’s AVG.

Home Runs/Power
2012 Values 30 HR/.238 ISO

As whole, Mike Trout displayed more HR power as the season progressed when looking at his HR/FB%.

Month: HR/FB%
May: 15.6%
Jun: 12.5%
Jul: 31.3%
Aug: 29.2%
Sept/Oct: 20.0%

Besides the HR/FB%, his batted ball distance on fly balls and home runs increased over the last couple of months.

Mar through July: 280 ft
Aug through Oct: 288 ft

His ISO, on the other hand, went in the opposite direction

May: 0.231
Jun: 0.159
Jul: 0.412
Aug: 0.216
Sept/Oct: 0.211

The late season ISO drop should be expected considering how his batted ball distribution changed from fly balls and line drives to ground balls.

With the few number of fly balls Trout was able to hit over the last couple of months, more of them went for home runs. From the beginning of the season until the end of July, he hit a HR for every 20 PA. The value increased to 22 PA for every HR from the 1st of August onward. With an almost 10 percentage point drop in FB%, I was amazed he kept the HR rate up.

I don’t see as much HR variability as I saw with his AVG.

Stolen Bases
2012 Values 49 SB, 5 CS

Trout is still young, so I expect no decline from aging. Splitting out the season before and after the knee injury, he stole bases in 8.4% of his PA before the injury and 6.7% after it. Since he was on base less in the second half of the season, the SB numbers should be expect to drop some.

The big hit to Trout’s SB numbers looks to be from a possible move from the lead off spot to the 3rd spot in the Angels’ lineup. It seems intuitive that he would attempt less SB with the power of the lineup up next. I went and looked at a few players (*) that split a season hitting either 1st or 3rd. I equalized the plate appearances and found the move to the 3rd spot didn’t hurt the player’s SB values, but actually helped. Adjusted for 600 PA in the lead off spot, the players attempted 34.8 SB and were successful 25.9 times for a 74% success rate. When the players batted 3rd, they attempted 36.0 SB and were successful 29.4 times for a 81.6% success rate. The players don’t really get the brakes put on and since they aren’t expected to steal in the 3 hole. Also, their success rate increases.

Conclusions

The key for projecting Trout in 2013 comes down to possible reasons for the late season drop off. If a person believes the drop off was because the season wore on him (I don’t at all), then great. He should have more experience and conditioning and be ready for a fruitful 2013.

If the injury was the problem, he should be expected to perform better in 2013 than in 2012. Yes I said it, better. If he was able to maintain what he did in the first 3 months over the course of a season, he should be expected to have a better AVG and more HR and SB. If the knee was bothering him, Trout should be the slam dunk #1 pick.

I am not buying the knee injury. Usually power drops off because of an injury. After the injury, Trout hit fly balls which went further and for more home runs than before. He was able to maintain the same swing before and after the injury. He only missed one game. Even though the time of his drop off and injury coincide, I just don’t believe his decline came because of the injury.

I believe opposing team in the league adjusted to Trout. They pitched him so he would hit less line drives and fly balls. Also, they began to strike him out more. Defensive began to get better data on where to position their outfielders. He put less balls in play and those in play were more likely to be ground balls. He will still be a good hitter if he puts up numbers similar to his August to October slash line of .287/.383/.500. The second line put him in the range of hitters like David Wright, Billy Butler, and Carlos Gonzalez instead of with Miguel Cabrera and Ryan Braun. Good hitter, not great hitters.

Trout will need to adapt like the league did to him. Early in the season, I would watch his batted ball mix. Is it closer to 55%/28%/17% GB/FB/LD like he hit over the last couple of months or near 40%/35%/25% when he was hitting the leather off the ball? Also, where is his strikeout rate? Down near 20%. Good. Up near 25%. Bad.

Finally, maybe the best method is to roll the dice and go with his projection.

I hope the preceding was detailed enough. The biggest question mark for Trout is if he will hit like he did in his first 3 months, last 2 months or somewhere in between. In re-draft leagues, I would not take him ahead of Cabrera or Braun. Both have longer established careers and therefore less down side. Trout could carry a team, but he could also perform like players taken 3-4 rounds later.
.
.
.
* Hanley Rameriz (2007,2010), Alex Gordon (2012), Carlos Beltran (1999,2001), Andrew McCutchen (2010,2011), Ichiro Suzuki (2012), Carl Crawford (2006), Jose Reyes (2012)




Print This Post

Jeff writes for FanGraphs, The Hardball Times and Royals Review, as well as his own website, Baseball Heat Maps with his brother Darrell. In tandem with Bill Petti, he won the 2013 SABR Analytics Research Award for Contemporary Analysis. Follow him on Twitter @jeffwzimmerman.


21 Responses to “Mike Trout: Reasons for Caution”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Jon Roegele says:

    Really excellent, thorough work Jeff, thank you.

    I found the move from 1st-to-3rd in the order not affecting SB attempts a very surprising finding. Scoscia will no doubt have everyone run anyway, so this could be the same for Trout if he moves to 3rd also.

    I remember in August when Trout was above .330 AVG and .400 OBP with a K% above 20% looking at historical comparables, and there are not surprisingly a very small number of players who managed either of those percentages with that high of a K% in a season. Thanks for pointing out the drop over the season.

    Vote -1 Vote +1

  2. Yeah, great work here, Jeff.

    Vote -1 Vote +1

  3. JV says:

    Definitely enjoyed the article. Is his floor (let’s say .280, 18 homers, 35 steals plus bunches of runs) really comparable to a 3rd/4th rounder? I’d definitely take that over, say, Matt Holliday, a typical 3rd round type outfielder.

    Vote -1 Vote +1

  4. aidan says:

    Great Work Jeff, really enjoyed the article

    Vote -1 Vote +1

  5. Jon L. says:

    Just when we were thinking there was nothing new to say about Mike Trout!

    Vote -1 Vote +1

  6. Joe Cool says:

    batting average is driven by 3 things, babip, strikeouts, and non-balls in play hits (home runs)

    30 home runs can add 60 points to batting average in 500 ab. maybe less if some of them would have been hits. But surely some outs.

    I project Trout as a 245-280 hitter next year. I’m going a wide range, plz don’t criticize the extremes of the range, as that would not be productive.

    Vote -1 Vote +1

    • JeffMathisCera says:

      Things that also aren’t productive: predicting things with that hilarious range and then asking for no criticism. What’s the point?

      Vote -1 Vote +1

    • BJsWorld says:

      I’m predicting that Trout will play between 80-162 games. His average will be in the 230 – 360 range. He’ll easily hit between 2 and 40 HR’s. I have a 75% comfort level with these numbers. Now, if he gets injured that could effect his numbers.

      Plz don’t criticize the extremes of the range, as that would not be productive.

      +29 Vote -1 Vote +1

    • ralph says:

      Batting average is also driven by BB%, though to a much lesser extent. If Player A and Player B have the exact same BABIP, K%, HR/PA, but Player A has a 15% BB% and Player B has a 5% BB%, Player B will actually have about a 4-point higher AVG.

      Vote -1 Vote +1

  7. Kevin says:

    Now that the season has ended, and the sick two months midseason he had, we can look at the big picture of Trout. He still will be an exciting player next year I think, just not take the world by storm as he did.

    Hopefully he will be spending a lot of time over the winter watching film and using that mild California weather to keep practicing.

    Vote -1 Vote +1

  8. philosofool says:

    This article was fun, but it does help me plan my fantasy baseball teams.

    Articles that examine obvious top picks in excurciating detail don’t help fantasy owners. I mean really, did you think someone reading this thought he was going to be available with a 5th overall draft pick? It’s fun to specualte about his 2013 performance, but the practical reality is simply that you should draft him if you can, perhaps unless Ryan Braun is also available, and we readers benefit more from article about guys about whom there’s uncertainty.

    Vote -1 Vote +1

    • AWS says:

      Auction leagues are a popular format now, in which all managers have the potential to select from the entire player population. So this article is very useful for people in those formats, particularly because the author goes well beyond the usual rote analysis for top-10 overall players. Thanks, Jeff.

      Vote -1 Vote +1

  9. Daniel says:

    So if Trout’s numbers come down to earth a bit into the Wright/Cargo territory, but he is able maintain a similar stolen base total, then isn’t he still very worthy of a top 3 pick? If Cargo or Billy Butler swiped 50 bags, either would be a sure fire first round pick right?

    Vote -1 Vote +1

  10. Nate says:

    Pfff, math.

    Mike Trout will eat your soul’s soul. Draft him before he drafts you!

    Vote -1 Vote +1

  11. FGreader says:

    Great article.
    So after clicking the xBABIP link and looking a little into that statistic more than I have before, I have a few general questions-
    What exactly is the difference between xBABIP and BABIP? It’s just a different formula, correct? Is xBABIP in a nutshell just a more complicated but more accurate portrayal of what a players avg. should have been? or, essentially just using LD, GB, FB rates etc. instead of standard stats.
    Also, what is a normal xBABIP? is it a good measure to look at a player and decide how much luck affected him or is it like the normal BABIP in that it is very fallible and it’s true purpose (while commonly mistaken) isn’t to portray luck?

    Sorry if some of those questions are confusing or worded strange but I can’t really find a whole lot of information on the subject and just wanted to look into that interesting measure a little more.

    Vote -1 Vote +1

    • Jeff Zimmerman says:

      What exactly is the difference between xBABIP and BABIP?

      xBABIP is the expected BABIP knowing the player batted ball mix (line drives, ground balls, etc). Some player may be lucky or unlucky on the balls they put in play?

      It’s just a different formula, correct?

      Yes

      Is xBABIP in a nutshell just a more complicated but more accurate portrayal of what a players avg. should have been? or, essentially just using LD, GB, FB rates etc. instead of standard stats?

      The second explanation.

      Also, what is a normal xBABIP?

      Same as a normal BABIP, around .290-.300

      is it a good measure to look at a player and decide how much luck affected him or is it like the normal BABIP in that it is very fallible and it’s true purpose (while commonly mistaken) isn’t to portray luck?

      Yes on seeing how much luck they had.

      Let me know if you have any more questions are need more thorough answers.

      Vote -1 Vote +1

  12. Scott says:

    It looks like his swing also became a little longer in the last two. He wasn’t as quick to the ball.

    Vote -1 Vote +1

  13. Swfc-dan says:

    Great stuff Jeff, just great work.

    My thoughts on Trout too, with the analysis to back it up! Always felt he’d be too much of a gamble to draft him no1 hoping he puts up another amazing year. Guys like Miggy and Braun have the track record and you know what you are getting from them. I can see Trout performing as a late 1st/ early 2nd rounder next year.

    Vote -1 Vote +1

  14. Ruki Motomiya says:

    Could the knee injury have made him slower and thus beat out less balls and affect his BABIP?

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>