Joc Pederson had a bizarre rookie season, but it was ultimately a successful one.

The Dodgers' outfielder struck out in over 29% of his plate appearances and only hit .210, suggesting he struggled in the bigs. In reality though, he had a good year at the plate, with a 15.7% walk rate and .206 isolated power rate, pulling up his on-base and slugging percentages.

Pederson slashed .210/.346/.417 with 26 home runs, producing a 115 wRC+ after park and league adjustments, (meaning he was 15% better than the average hitter) Overall, he was worth 2.8 fWAR, 2.3 rWAR and 2.3 WARP.

This success, despite the poor batting average, is not really what made his season so strange, though.

The weirdest part of his year was the drastic split between his first and second halves at the plate. During the first half, he had an .851 OPS and his 137 wRC+ was tied for 23rd in the Majors.

He struggled in the second half, though, posting a .617 OPS and 79 wRC+.

All things being equal, we should probably give the most weight to his full season stats, rather than look at either split, when trying to forecast his upcoming season; a 585-plate-appearance sample is superior to a 219 one.

That said, the split here is so drastic it does warrant some closer attention.

What Went Wrong

Pederson’s problems in the second half were caused entirely by balls in play.

His strikeout and walk rates in the second half were almost identical to his first half rates, as his walk rate went from 15.8% to 15.5% and his strikeout percentage moved from 29.2% to 29.8%.

His batting average on balls in play plummeted though, from .282 to .232, and it certainly seems like there was more here than mere random variation.

Infield fly rate and line drive rate both correlate well with in-season BABIP, and both fell for Pederson in the second half. His infield flies per fly ball more than doubled, going from 10.1% to 22.6%, while his line drive rate fell from 17.1% to 13.6% (for non-pitchers in 2015, the MLB average in-field fly rate was 9.5%, while the average line drive rate was 21.0%).

Also, while exit-velocity and hard-hit-ball rates actually do not correlate well with BABIP, Pederson did decline in both areas, giving further support to the implication he was not making quality contact in the second half.

His average exit velocity went from 93.5 miles per hour in the first half to 89.3 in the second half (according to Baseball Savant) while his hard hit ball rate fell from 41.5% to 30.3%. Perhaps more distressing was the increase in soft contact rate, which rose from 14.9% to 29.4% (per Baseball Info Solutions, via FanGraphs; for non-pitchers, the average hard hit ball rate was 29.0%, and 18.3% for soft contact).

Though these quality contact stats do not correlate with BABIP, they do have a strong relationship with power stats (in particular ISO), which also dipped for Pederson.

His ISO dropped from .257 in the first half to .122 in the second, while the MLB non-pitcher average was .153. His total bases per hit rate (“Power Factor”) also fell, from 2.1 to 1.7.

Should We Worry About It?

While Pederson’s second-half slump is hardly encouraging, it is also probably not predictive.

As mentioned, his walk and strikeout rates barely moved from one half to the next, and his groundball rate was consistent as well (42.0% in the first, 41.5% in the second).

These stats are among the most stable in small sample sizes, while BABIP and line drive rates are notoriously volatile (Russell Carleton of Baseball Prospectus wrote that line drive rate needs about 600 balls in play to become reliable, while BABIP needs more than 820; Pederson put 311 balls in play last year).

The decline in power and contact quality, in addition to the high pop-up rate implies that there was a real performance issue here, rather than mere random variation. However, it would be silly to chalk this up to a drop in true talent and thus invalidate his first half.

Craig Edwards’ research over at FanGraphs is probably instructive here. Edwards found there were six other players since 2002 who were above average hitters in the first halves in their rookie seasons and below average in the second half.

Not only were all six players above average hitters in their second big league seasons, they all did “the same or better over the next two seasons relative to their rookie years.”

As for their second years themselves, the players’ overall rookie wRC+ correlated with their second-year wRC+ a higher rate than that of the second half of their rookie year.

This should not be a surprise, given this is also true for offensive stats in general, as Jeff Zimmerman found.

This research lends further credence to the idea Pederson’s overall rookie season, instead of his late decline, will be most predictive of his 2016 season. And just in case you needed more proof, have a look at what the projection models have in store.

These include our projections here at numberFire, which are now live (but happen to be more pessimistic about Pederson than Steamer, ZiPS and PECOTA).

The low batting average limits his fantasy value in standard leagues, but his power still makes him an intriguing option.

And if you’re in an OBP league, Pederson should be even more attractive.