Evan Gattis will be really good in 2016

Every off-season I try to make at least one "hawt take" -- my feeble attempt to fit in better with national sports media.  In 2013, I boldly proclaimed that Dallas Keuchel would be the Astros' best pitcher in 2014.  Win for me.  In 2014, I dared suggest that Brett Oberholtzer would have a breakout season in 2015.  Obie got injured, tried to bean Alex Rodriguez, and then got demoted.  If all that hadn't happened, I'd be two-for-two, and you can't prove differently.

My Hawt Take of the 2015-16 offseason is this:

Evan Gattis will be really good in 2016.

To qualify that statement up front, I wish to make clear that I am not saying that Evan Gattis will be "2015 Bryce Harper good," or "2004 Barry Bonds good," or even, "in a non-hilarious conversation for 10th-place in the MVP race good."  Rather, I mean that he will be merely really good.

Good grief, I feel uncomfortable, going out on a limb like this.  Maybe that's why I only do it once a year.

*note - there's a short glossary of the stats mentioned in this article at the bottom of the page

There are several reasons I can point to, not the least that a guy who hit 27 home runs in 2015 and knocked in 88 runs can't be doing all bad.  He's not a strikeout machine the way some of the other Astros' regulars are.  He's got great bat control, and hey, if he chases a few (read: a lot) of bad balls, he makes pretty good contact on them, which is something you can't say for fellow Astros slugger Chris Carter.  Gattis' batted ball velocity on flies and liners ranked within the top 15% of all batters last season (100 PA min).

During 2015 while playing as the Astros' designated hitter, Gattis hit .246/.285/.463 (meh...) with 27 home runs, 88 RBI, and 66 runs scored.  He managed a strikeout rate slightly under 20%, just about exactly a league-average rate, and he walked 5% of the time compared to an 8% league average.  Earning the free pass always has been Gattis' biggest weakness.  But very few players have no weaknesses.  Overall, Gattis amassed a 99 wRC+ last season, indicating he was approximately 1% worse than the average hitter in 2015.  Considering the average DH last season scored a wRC+ of 115, that's not a great showing.

But my main reason for boldly predicting 2016 goodness for Gattis is the fact that Gattis was already a lot better in 2015 than you think he was.

Under the hood

But Gattis' 2015 counting stats don't tell the whole story.  Baseball researchers have known for a long time that random variation can play havoc with a batter's stats over the short term, and yes, one season can be classified as "the short term."

Take Gattis' .264 batting average on balls in play (BABIP) in 2015, for example.  During the season, as with all players, his BABIP fluctuated and corresponded to his hot and cold streaks.

The point of the chart above is only to show that a player's performance is definitely affected by his BABIP, something largely out of the batter's control.

I told you that to tell you this: overall, Gattis' BABIP was way lower in 2015 than it should have been.  The easy thing to do would be to point to his .270 career BABIP and say, "it was only a little low last season".  But that's disingenuous because since he amassed 604 plate appearances in 2015 compared to 783 during 2013 and 2014 combined, his 2015 BABIP would have a disproportionate effect on figuring his total career value.

Luckily, we have xBABIP, or expected BABIP.  I promise not to dig deeply into the nuts and bolts.

Warning: xBABIP explanation ahead!

xBABIP uses stats that generally do reflect things the batter can control as a proxy for what his BABIP should have been if that darn ol' luck thing hadn't gotten in the way.  xBABIP is constantly tweaked by different folks always seeking an equation that better correlates to BABIP.  The most recent (and most accurate) one that I've found was developed by Alex Chamberlain over at Fangraphs and discussed back in May.  This version is interesting compared to previous ones because instead of using batted ball percentages alone, it digs a little deeper and includes a players' Speed score (something missing in previous versions that definitely affects BABIP), batted ball location (how often a player hits to the opposite field, in this case), and how hard the batter hits the ball.

End xBABIP explanation.

I dusted off Excelibur and crunched Evan Gattis' xBABIP for 2015.  And lookie - it stands at a nice .301, meaning that he should have reached base 30 out of every 100 times he put a ball in play, as opposed to the 26 out of 100 that he actually did.

That matters.  A lot.  That would have added 16 hits to his 2015 total.  I had a little fun with math, and based on his actual HR/H, R/H, and RBI/H rates, I came up with an approximation for his performance had his BABIP actually been .301 instead of .264.

Folks, that's how much difference BABIP can make.  Would anybody be talking about trading (or, ridiculously, DFA'ing) Evan Gattis if he had hit .274 with 30 home runs and 98 RBI?  Of course not.

But again, this is a little disingenuous because it only considers 2015 data.  I re-calculated Gattis' xBABIP based on his career batted ball statistics.  I wondered if perhaps his 2015 xBABIP was not representative of his true talent -- was it a little too high?

Turns out, yes.  Over his career, Gattis' xBABIP maths out to .294.  That's still a good chunk higher than his actual values.  So here's that same chart again.

That's still pretty good.

Really?

Well, not necessarily.  But probably.  xBABIP is a relatively new and still evolving metric.  Generally speaking, it is a fair predictor of a player's true BABIP expectancy.  But there may still be a piece of the puzzle missing that could theoretically make Gattis' xBABIP drop like a rock, down to around his actual career levels.

But odds are very good, based on what we know, that Gattis' true talent and expected performance level is quite a bit better than what we saw in 2015.  He's still the same guy who managed .263/.317/.493 (126 wRC+) in 2014 during an injury-plagued season.  It's possible, or even likely, that 2014 was a truer measuring stick for his long-term expectations than 2015 was.

For that reason, I boldly predict that Evan Gattis will have a very strong season in 2016, and will stand proudly among the better DH's in the American League.  2016 slogan:  Goodness for Gattis!

Last-minute edit:  There's also this link, which I just found after writing this post.

* * *

Glossary

BABIP - Batting Average on Balls in Play:  For every ball put in play, the chance that the ball falls for a hit.  Heavily influenced by opponent defense and luck.

xBABIP - expected BABIP, calculated various ways, most recently with batted ball rates, player speed scores, and hard hit ball percentage.

wRC+ - a measure that uses linear wweights of a batter's overall offense that adjusts for park factors so that batters can be compared as neutrally as possible.  100 represents an average offensive player.  112 would be a player 12% better than an average batter.