March and April Flashes of Power
Predictive?

.

Chad Young wrote an excellent article about hitters who showed early signs of "losing it" or of leaping a plateau.  He used batted-ball distance to ID five sleepers and five land mines:

Player 2013 distance as of Apr 23 2012 distance Change
Dexter Fowler 342 feet 279 +63 feet !
Todd Frazier 331 280 +51
Todd Helton 334 283 +51
Colby Rasmus 319 274 +45
Salvador Perez 331 285 +46
       
Matt Kemp 240 313 -73
Rauuulll 224 292 -68
Joey Votto 246 301 -55
Neil Walker 340 292 -48
Kendrys Morales 247 286 -39

.

Independently, Dr. D had been worrying about how hard Morales was hitting the ball, and this article didn't soothe his jangled nerves.   and who else is in that "losers" column?  Rauuuuull.  He seemed like an obvious candidate to pull a John Olerud.  (Oley, in his last year, was exactly the same hitter he'd always been, except his 370-footers started going 320 feet.)

But Morales has bounced back, with interest, and Rauuulll?  We know how that turned out.

So did that chart have any value, or did it have reverse value?!   I mean, the names also looked suspect ... why would Joey Votto be a big loser?  Why would Todd Helton, at this time of the day, suddenly gain a bunch of power?

On the other hand, we saw an article that did correlate huge Spring Training SLG (+200) gains with improved regular season performance.  Justin Smoak, IIRC, was in this category.  So was Morse.  (Smoak's OPS+ is indeed up, from 86 to 133.)  So maybe the early flashes of power are worth looking at.

Anyway, let's see how it turned out:

 

:

Player 2013 OPS+ 2012 OPS+ Change
Dexter Fowler 113 123 -10
Todd Frazier 97 97 0
Todd Helton 90 86 -4
Colby Rasmus 83 126 -43
Salvador Perez 113 88 +25
       
Matt Kemp 96 146 -50
Rauuulll 127 102 +25
Joey Votto 177 161 * no change
Neil Walker 111 111 0
Kendrys Morales 132 118 +14

.

For what it's worth ... roto players who bet on this 10-player list didn't win.  They steered clear of Matt Kemp, and they cashed in on Salvador Perez.  But they missed out on Rauuuul and Kendrys, and the other six players were pushes.

Looking a level deeper ... Dexter Fowler seemed like a player who could be interesting here.  But his BIP distance was way up in April ... his SLG is way down now, from .474 to .437.  Granted, his ISO is unchanged.

The 10-player "sample" (sic) isn't conclusive, obviously.   But combining this list with the scout's eye there is a light bulb to be had.

Don't worry about it too much if a veteran is taking short swings at the ball in his first month.  He could be just getting his bearings.

On the other hand, if a player is suddenly showing much-improved power in March and April, that very often IS a sign of things to come.  It's worthy of consideration.  In Justin Smoak's case, the March SLG and the entire history arc did signal a big move forward.

 

 

Comments

1

Honestly, when the typical MLB "streak" or "slump" tends to max out at SIX (6) weeks ... any single month stats have to be taken with a truckload of salt.
And if you're looking at the extreme ends based on April stats, you've got a tailor made situation where the most likely results are going to be regression to the mean.
This doesn't mean monthlies are useless. But, they are not useful tools (IMHO) "predictively". I tend to use them (over time) to note if/when a player goes significantly beyond "their" normal up/down boundaries.
Take Smoak. Back in 2010, in the 3 months where he played 20+ games what were his OPS totals?
May - .589
June - .832
July - .414
In 2011, his 20+ games in a month OPS totals were:
April - .527
May - .750
June - .737
July - .399
Sept - .793
In 2012?
April - .572
May - .754
June - .458
Sept - .1005
If one looks ONLY at September of 2011, one could conclude a great 2012. But, then his 2012 September was his best month ever. Is that "predictive" that he's a different hitter?
IMO, you have to at least be aware of ALL the data. That 1005 month came just 2 months after that .458 month. It's a NEW top-end - and the potentially has some significance, (but he hit 1001 back in 2010 ... but only in 14 games, so it didn't make the cut for examination).
So, what about 2013?
April - .643
May - .780* (only 18 games)
June - .888* (only 10 games)
July - .843
Finally, in 2013, we see REAL departure from the previous pattern. But note - it is *NOT* tied to a new upper limit. It is tied to a new BOTTOM. Every previous season, he would go a month with an OPS under .500. His bottom in 2013 currently stands at .643. *THAT*, (for me), is the "tell" of a player who is adapting and improving. And even in the partial months for 2013, Smoak's numbers are MUCH steadier than his previous years.
That said ... age is a huge factor, and when you see Raul's power numbers swoon, it would be perfectly reasonable to combine the numbers and his age and conclude "he's done". Obviously, that conclusion was wrong in hindsight. There are NO "perfect" tells. My own tell for a player aging out is to watch his BABIP plunge below normal variance levels. But, that plunge has to be sustained to be a valid tell. Monthly splits just don't contain enough data.
I think one of the big problems for SABEs who utilize monthly data is that some have failed to study the data to understand what "normal" variance is. To me, you have to be grounded in the basics - 300 points monthly swings in OPS are so common as to be trivial. As noted above, Smoak had "routine" 400 point OPS swings over multiple seasons. As you note, guys like Shandler and James have poured over so much data, they "know" these things even if they don't "know" these things.

Add comment

Filtered HTML

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd><p><br>
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.

shout_filter

  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • Web page addresses and e-mail addresses turn into links automatically.