With the news of Adam LaRoche‘s new deal with Washington, I remembered his well-known history of slow starts and hot finishes. Thanks to B-R’s new batting split finder (beta), we can compare his 1st-half/2nd-half splits to those of other players.

 

I ran a batting split search covering the past 10 years (2003-12) for each half, with minimums of 1,500 PAs in the 1st half and 1,200 PAs in the 2nd half. There are 219 players who appeared in both searches. I copied them into Excel and calculated the differences.

For the record, the composite numbers are just slightly better in the 2nd half — unweighted average of +.002 in BA and +.008 in OPS — so we can fairly talk about individual 1st-half/2nd-half splits without making an adjustment.

LaRoche’s OPS difference of +.118 (2nd) is the 5th largest in the group. The only larger gap among active players is Troy Tulowitzki‘s +.143 (2nd).

Here are the 20 biggest gaps in OPS by half:

  • Diff. / Half … Player
  • .143 / 2nd … Troy Tulowitzki
  • .132 / 2nd … Geoff Jenkins
  • .122 / 1st … Hank Blalock
  • .122 / 2nd … Carlos Delgado
  • .118 / 2nd … Adam LaRoche
  • .105 / 1st … Kevin Youkilis
  • .098 / 2nd … Adrian Beltre
  • .097 / 1st … Justin Morneau
  • .096 / 2nd … Ryan Howard
  • .092 / 1st … Eric Byrnes
  • .090 / 2nd … Ryan Zimmerman
  • .089 / 2nd … Aramis Ramirez
  • .087 / 1st … Sean Casey
  • .086 / 1st … Craig Biggio
  • .086 / 1st … Ivan Rodriguez
  • .085 / 2nd … Carlos Pena
  • .084 / 1st … Reed Johnson
  • .083 / 1st … Mike Lowell
  • .083 / 2nd … Jimmy Rollins
  • .081 / 2nd … Randy Winn

The top 20 OPS deltas for those better in the 1st half:

  • Diff. / Half … Player
  • .122 / 1st … Hank Blalock
  • .105 / 1st … Kevin Youkilis
  • .097 / 1st … Justin Morneau
  • .092 / 1st … Eric Byrnes
  • .087 / 1st … Sean Casey
  • .086 / 1st … Craig Biggio
  • .086 / 1st … Ivan Rodriguez
  • .084 / 1st … Reed Johnson
  • .083 / 1st … Mike Lowell
  • .081 / 1st … Craig Counsell
  • .080 / 1st … Scott Hatteberg
  • .072 / 1st … Chase Utley
  • .071 / 1st … Ian Kinsler
  • .066 / 1st … Jose Bautista
  • .063 / 1st … Jason Varitek
  • .062 / 1st … Ben Zobrist
  • .062 / 1st … Jose Guillen
  • .059 / 1st … Paul Lo Duca
  • .057 / 1st … Jonny Gomes
  • .056 / 1st … Brandon Inge

And the top 20 who are better in the 2nd half:

  • Diff. / Half … Player
  • .143 / 2nd … Troy Tulowitzki
  • .132 / 2nd … Geoff Jenkins
  • .122 / 2nd … Carlos Delgado
  • .118 / 2nd … Adam LaRoche
  • .098 / 2nd … Adrian Beltre
  • .096 / 2nd … Ryan Howard
  • .090 / 2nd … Ryan Zimmerman
  • .089 / 2nd … Aramis Ramirez
  • .085 / 2nd … Carlos Pena
  • .083 / 2nd … Jimmy Rollins
  • .081 / 2nd … Randy Winn
  • .075 / 2nd … Luke Scott
  • .074 / 2nd … Robinson Cano
  • .070 / 2nd … Juan Rivera
  • .069 / 2nd … Aubrey Huff
  • .068 / 2nd … Chase Headley
  • .068 / 2nd … Mark Ellis
  • .066 / 2nd … Jeff Francoeur
  • .064 / 2nd … Juan Uribe
  • .064 / 2nd … Eric Chavez

 

LaRoche in isolation

A detailed look at LaRoche’s splits by half turns up a bit of a puzzle:

Split G GS PA AB R H 2B 3B HR RBI BB SO BA OBP SLG OPS BAbip tOPS+
1st Half 666 607 2618 2312 307 572 156 6 94 365 272 579 .247 .326 .442 .768 .288 88
2nd Half 520 482 2051 1847 261 542 125 4 103 319 174 456 .293 .353 .533 .886 .335 115

(And here you can see the year-by-year splits for 1st half and 2nd half.)

But what’s actually going on? A large change in BA, in a sample of this size, usually reflects some improvement in strikeout rate. But his SO rate is the same in each half, 22% of PAs, 25% of ABs. The change in BA comes entirely from balls in play (BAbip). Given what we know about BAbip, it seems odd to see such a large in-season gap built up over nine seasons.

Now to the power: His increased HR rate in the 2nd half (5.6% of ABs vs. 4.1%) suggests that he’s driving the ball more as the year goes on. However, his doubles rate is exactly the same, 6.7% of ABs in each half. Doubles as a percentage of hits actually go down a bit (27% to 23%), as do the combined extra-base hits (45% to 43%).

And, for whatever it’s worth, his walk rate also declines in the 2nd half, from 10.4% to 8.5%. The combined walk rate of the study group held steady at 9%.

I don’t know what it all adds up to. LaRoche’s breakdowns don’t seem to tell a typical story of a significant change in results. Am I missing something?

Lastly, his splits by month:

Split G GS PA AB R H 2B 3B HR RBI BB SO BA OBP SLG OPS BAbip tOPS+
April/March 196 181 777 682 83 152 43 0 27 104 92 188 .223 .317 .405 .721 .267 78
May 226 205 887 771 107 190 60 3 26 124 99 188 .246 .330 .433 .763 .288 88
June 178 160 694 624 81 157 37 1 29 101 60 146 .252 .317 .454 .771 .281 88
July 176 165 695 629 86 183 38 2 34 99 54 143 .291 .344 .520 .864 .323 110
August 199 184 798 712 110 215 41 2 42 131 76 169 .302 .369 .542 .911 .341 122
Sept/Oct 211 194 818 741 101 217 62 2 39 125 65 201 .293 .348 .540 .887 .349 115

This table shows that measuring the “halves” as April-June and July-Sept. — which is closer to the actual midpoint than is the All-Star break — would yield an even bigger gain in the hot months: unweighted BA/OPS averages of .240/.752 for April-June, .295/.887 for July-Sept. That’s +.055 in BA and +.135 in OPS. And while the BA and OPS rise consistently from April through August, there’s a huge jump from June to July. All three months in the 2nd half are very good, while all three in the 1st half are mediocre to poor.

Is there a simple theme to these splits? For instance, if LaRoche simply preferred warmer weather, you’d expect to find some evidence in his ballpark splits and his day/night splits. But I can’t see it. So, let the speculation begin!

Subscribe to: RSS feed