We Good, Pham

Playing with the wonderful new splits leaderboard that was just rolled out on these very pages has led me down a Tommy Pham-shaped rabbit hole.

Tommy Pham has a stat line that is currently boggling my mind.

.214 ISO, 10.9% BB%, .342 BABIP, and 9 HR in 183 PA…good to excellent offensive numbers, in my opinion. Yet despite all of these good to excellent offensive numbers, he sported a major-league-high 38.8% strikeout rate (min 100 PA) that dragged his wRC+ to a barely-above-average 105. This deserves some digging into.

Looking at this 15-game rolling K%, there were times this past season that his rolling K-rate was down to 20.8% (on August 12). The AMAZING thing happens the further right you look on that graph — he begins striking out at a rate that makes Bartolo Colon look patient, hitting a high of 66.7% in the middle of September. From the beginning of the season to August 12, Pham had a wRC+ of 126 — higher than the full-season numbers of Carlos Beltran, Nolan Arenado, and Jose Bautista. After August 12, his wRC+ was 40. 40! FOUR ZERO. That’s behind nine pitchers (min. 40 PA).

AND SOMEHOW

SOMEHOW

He managed to have a higher BABIP when he was walking through life in a strikeout-induced haze. After August 12, he ran a BABIP of .417 with a K% of 59.1%, meaning he didn’t put the ball in play much, but when he did, it was finding the holes. BABIP and wRC+ have an R^2 correlation of 0.23, so you’d sort of expect them to move up and down together. However, before he started striking out like he was afraid someone was going to outlaw strikeouts so he was getting them all in while he could, his BABIP was 89 points lower — .328.

That’s not just lower. That’s much lower. That’s the difference between Dexter Fowler and Albert Pujols. And somehow an 89-point difference in BABIP resulted in an 86-point difference in wRC+ in the wrong direction.

You’d think running a much higher BABIP would be the result of hitting more line drives. After all — that is the variety of batted ball that lands for a hit more often than any type.

BUT. IT. GETS. WEIRDER. He hit line drives 28.0% of the time up to and including August 12. After August 12, he hit line drives ONLY 7.7% OF THE TIME. So with a 28.0% line drive rate, he ran a .328 BABIP, but his 7.7% line-drive rate resulted in a .417 BABIP. WHAT KIND OF MAGIC IS THIS?

Well, you know. The magic of small samples. 183 plate appearances falls nearly 70 short of being half of a qualified season’s plate appearances. Weird things are going to happen when you are looking at smaller samples. Weird things are always happening in baseball; that’s part of its charm. We just don’t always notice because over the course of a season, some weird things will balance out other weird things and we’ll forget how weird things were at some point. That’s why it’s worth it to dive into the numbers — to remind yourself that fun things are always happening in baseball. You may even find yourself surprised with how interesting you find Tommy Pham at the end of it all.





David Palardy is a father, husband, software developer, amateur sabermetrician, and lifelong Phillies fan. More of his writing can be found at palardelphia.com, which is mostly but not exclusively Phillies related.

2 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Rylan Edwardsmember
7 years ago

Nice article! Small sample size, yes but maybe more to it. I looked into Pham last season and found he became more selective (despite seemingly losing all contact abilities), started killing braking pitches (exit velo and line drives), and hit to all fields (helping the babip). Not sure where his contact went but a crazy evolution for sure.

http://www.fangraphs.com/fantasy/a-tommy-pham-fantasy-focus-on-fangraphs/

YKnotDisco
7 years ago

That magical .417 BABIP you refer to came from 13 AB. I know you mention SSS, but 13 AB? He hit the ball hard 38.5% in those 13 AB. He hit the ball on the ground 61.5% of the time in those 13 AB. Those grounders found holes when he hit up the middle (38.5%) and to his pull side (38.5%) in the 13 AB sample. You don’t need a high LD% in 13 AB to run a .417 BABIP.