Athletics breakout candidate: Ike Davis

If you look merely at Ike Davis's stat lines since his 32 home run effort in 2012, you will see a rather pedestrian performance for a first baseman: time lost to an oblique strain in 2013, and home run totals not even reaching the teens. Both major wins above replacement statistics made Davis out to be a replacement level player.

The market did not think much of him either. The Pittsburgh Pirates were on their way to cutting him from their roster before sending him to the Athletics for the mere right to avoid the MLB tax on acquiring young international players.

Available projections seem to express some optimism. Steamer thinks Davis will have the best wOBA on the A's of .338 (a wRC+ of 120) and if given 600 plate appearances would hit 20 home runs. ZiPS is less excited, projecting a .314 wOBA (and an OPS+ of 102) though still hitting 21 or 22 homers if given 600 plate appearances.

PECOTA, which was released by Baseball Prospectus on Wednesday, projects a True Average (TAv) of .281, with a .235/.331/.414 batting line with 17 home runs in 476 plate appearances (21 or 22 home runs in 600 plate appearances). Here's an imperfect scale of TAv against wRC+, using 2014 results:

There is, however, at least one measure that indicates that Ike Davis may exceed what the projections think of him. It begins with a simple question, how much did pitchers change their approach to Ike Davis as last season progressed?

Zone Distance Trend as a way to identify breakout candidates

Robert Arthur of Baseball Prospectus hypothesizes that when pitchers throw farther and farther away from the middle of the strike zone to a particular hitter, that's the clearest evidence that pitchers are, for whatever reason, afraid of the hitter. That the pitcher has reason "not to provoke a hitter and begin shading ever farther away from the zone's center." Generally, if pitchers suddenly begin deciding to pitch even farther from the middle of the zone, it means pitchers think you have a better chance of punishing them there:

Arthur identified the the players with the highest slope of the linear regression for zone distance in their 2012 seasons, that is those with the most dramatic change in pitcher zone distance. Arthur found that the top ten of those players outperformed their PECOTA projections by .0237 points of TAv. That's the difference between a Daniel Murphy (.277 TAv) and an Adrian Gonzalez (.300 TAv). Was this random luck? No, "The probability of randomly picking a group of 10 players who outperformed their PECOTA projections by that amount is something like .005, suggesting that changes in zone distance are statistically significant for predicting breakouts."

Did it repeat?

Arthur identified 12 more players with at least 1,500 pitches thrown to them in 2013 to predict results for 2014. Let's see how he did:

The average for this set of 12 is +.012 points above their projected TAv. The chance of a random set of 12 players performing as well is higher, though I'm afraid I don't have easy access to Arthur's data to determine the probability that this average was more than random chance. Still, that's two seasons of interesting results.

Ike Davis, breakout candidate

Using this method on September 9 (subscription required), Arthur identified Ike Davis as the player (min. 1500 pitches faced) with the greatest zone distance trend to that point in 2014:

Davis has flashed the raw ability necessary for elite play, and many of his plate discipline statistics are trending in promising directions. If he could also succeed in terrifying opposing pitchers into feeding him plenty of outside pitches, it seems to me possible that he could put together a truly superstar-class season. [His] change in zone distance is certainly sudden and severe.

Those plate discipline statistics include big drop-offs in the number of pitches swung at outside the zone, much more selectivity on pitches swung at inside the zone, and a much higher contact rate:

(from FanGraphs)

So what can we hope for from Ike Davis? A greater likelihood that he performs somewhere between .012 to .024 TAv points higher than his projected .281 TAv. A .293 TAv would be in line with Matt Holliday's 2014, and a .305 TAv in line with Adam Laroche's.

At any rate, Davis is already nearly projected by PECOTA to be as good as Brandon Moss is supposed to be next year, keeping in mind that Moss moves from the pitcher-friendly Coliseum to the neutral Progressive Field in 2015, which TAv adjusts for but the simple batting lines do not:

Ike Davis might be the second coming of Brandon Moss, or he might just be what Brandon Moss was going to be anyway. Let's hope Arthur's hypothesis continues to work for another season.