Identifying Struggling Pitchers with P.I.N.E.

I’ll be honest. P.I.N.E. is having some trouble. Pitchers in Need of Extraction, or P.I.N.E. to keep baseball jargon at an all-time high of acronym and buzz-word soup, is a simplistic model I developed to flag starting pitchers who have displayed some warning signs. P.I.N.E. flags statistics that have deviated from what that pitcher has done over a larger period, albeit in small data samples. That’s the most obvious issue. Taking small samples of volatile pitching data, in this case, the last 14 days or typically two starts per pitcher, and then comparing it to the last two years of a pitcher’s data, is tricky. Anything can happen in two starts and that’s why P.I.N.E. is not a model that should be automated to make your fantasy baseball decisions for you. Instead, it should present information that you otherwise may have ignored. Let’s use Max Fried as an edge-case example:

Had you seen Max Fried flagged by P.I.N.E. without some sort of explanation you would have scoffed, commented, tweeted, or done something within this virtual community to make sure you made your feelings known. After all, in Fried’s last outing, he threw a no-hitter through seven innings while striking out five. He did, however, walk three. Not atrocious, but still higher than what you would have liked as a fantasy manager, though you weren’t complaining. In an article that summarized this great outing, veteran writer Jay Jaffe also wrote about some warning signs in Fried’s game:

His whiff rates on his two main secondaries, his curve and changeup, are both down substantially relative to last season (from 39.1% to 32.7% for the former, and from 37.7% to 25% for the latter). His overall swinging strike rate has dropped from 11.7% to 9.3%, and his strikeout rate from 25.7% to 20.4%, his lowest since his rookie season. With his walk rate rising from 5.8% to 9.4%, his strikeout-to-walk differential has plummeted from 19.9% to 11%. All of that is at least a bit alarming.

But, then again, Jaffe also pointed out:

Set aside the ERA and FIP inflated by those early poundings, and it does seem to be working; of the 100 pitchers with at least 40 innings, his overall .185 batting average allowed is the seventh lowest, his .272 slugging percentage allowed the ninth lowest, and his .252 wOBA the 13th lowest. Whatever he’s doing, lately it’s working like a charm.

So, when P.I.N.E. flags Max Fried, it’s doing so because of some combination of the following has happened in the last 14 days:

• Fastball Stuff+ < 95
• O-Swing% is lower than two-year average
• Z-Contact% is higher than two-year average
• BB% higher than the league average plus an arbitrary two percentage points
• K% lower than the league average minus an arbitrary two percentage points
• BABIP is lower than the league average minus an arbitrary two-points
• Fastball velocity has decreased over the last three starts
• “Roster%” (PlayerRater) above 60%.

Here are the raw statistical comparisons that force Max Fried to be flagged by P.I.N.E.:

Layered in is the fact that Fried’s recent three-game velocity has dipped ever so slightly and you have all flags whipping in the wind. Yet, again, he just threw a seven-inning no-hitter. This statistical comparison is not telling us the full story, and as Jay Jaffe pointed out, Fried has had a lot of success with a sinker that he has begun using more and more to induce soft-contact on the ground. Z-Contact increases are fine if the ball that’s being put in the zone isn’t giving up hits. Perhaps the word “Extraction” in P.I.N.E. needs to be changed to “Examination”. Let’s examine all the starting pitchers flagged by P.I.N.E. this week:

Reynaldo López has been stellar in his short time as a starter for the Braves. He’s mostly being flagged in so many categories because of such a small sample of starts compared to his 14-day sample. We can place him in the edge case bucket.

If you’re skeptical, well, that makes two of us. This is an experiment. I drank my own Kool-Aid and benched Sean Manaea on Sunday, even though he pitched against the Marlins who had an MLB 26th-worst batting average of .214 against left-handed pitchers before the game. Manaea performed well, going five innings with four strikeouts and a win. I could have used that win. Still, he walked one and gave up a home run, but I would have traded that for the win any day.

As for Max Fried, he’ll likely line up against the Cubs at Wrigley Field this Wednesday, May 22nd. Will he ride your hypothetical fantasy pine? It’s a tough call to make. Will he be more like Max Fried against the Braves on May 11th (7.0 IP, 0 ER, 3 BB, 5 K, W) or Max Friend against the Padres on May 17th (4.1 IP, 3 ER, 3 BB, 2 K, L)? At least he’s been flagged.