• HUPGOOD
• PECOTA card
• Fangraphs card

If you're like me, you didn't know this guy existed prior to about 3pm.  That makes it difficult to know exactly what to make of the differences between this and this.  It would appear that we're only going to have actual data about half the pitches he throws. That's...suboptimal.  Moreover, I'm not sure why this happened.  He's been primarily a starter all year and it's pretty unlikely he was only throwing two pitches as a starter.  The cutter looks like it got incorporated into the fastball totals, since there's a slight disparity in velocity and a 2 inch difference in horizontal movement between the two.  I'm guessing if you averaged (appropriately weighted, of course) the player card fastball and the player card cutter, you'd get the tool fastball.  So keep that in mind when I talk about the results from the fastball.  It's actually the combined results of the fastball and cutter.  The change up apparently just disappeared.  He throws it 10 percent of the time to lefties and never to righties and throws it more frequently in counts that are not 0-2 or 1-2, so I doubt it's a big whiff pitch.  It helps him get into counts to use his curve, as the usage rate for the latter spikes in both 0-2 and 1-2 counts.

The curve appears to be a good pitch, but one that basically no one will offer at since he can't put it in the zone.  Against lefties, for example, only about 15% of curves in the database were offered at.*  The proper approach is apparently know league-wide at this point (otherwise, there'd be more swings on the curve).  To oversimplify using the international language of punctuation-derived wellness indicators:

TOPSPIN -->  : (

ALL ELSE --> : D

His straight stuff isn't tremendous and he doesn't like to put it in the zone that much.  Despite above average ball percentages, he still allows his fastball to be put in play with good frequency.  That and taking walks when sufficiently good looking pitches don't come (and there's no guarantee they will; his ISO numbers are decent, but his overall GB% is way down) is the appropriate approach.  Btw, if we aren't taking walks, then we're probably helping him out way too much.  Make Feldman give in and we're on our way to a hit parade with men on base.

*This reminds me that I need to re-configure whiff rate as it's listed in the Kalk database.  The "pure stuff" rating for a pitch would not be whiffs/pitch, but whiffs/swing.  I've been only working with the former and I think it's led to some misconceptions in previous BUOs, since I think I often conflated it to mean the latter.  The former incorporates the latter and adds something of an efficiency indicator, since, ultimately, you don't want to be stuck with merely good stuff.