BP360 now on sale! A yearly subscription, '23 Annual & Futures Guide and t-shirt for one low price

James McDonald made his third start of the season on Monday night. It wasn't pretty. Although the Cardinals managed to run into two outs on the bases, McDonald exited after 1 2/3 innings, equalling the shortest start of his career. The quick trip to the showers came after a so-so start against the Diamondbacks and a strong season-opening effort versus the Cubs. 

Dizzying reversals of trends are nothing new for the Pirates' right-hander. Despite finishing the past two seasons with the same amount of innings and same ERA+, McDonald has been too inconsistent for comfort's sake. Take last season's odd tale of two halves. He looked on his way to a strong campaign after allowing 31 runs over 110 first-half innings. But something fell out of sort in the second half and caused McDonald to allow 54 runs over his final 61 innings. His strikeout-to-walk ratio folded in half and his ERA more than doubled.

Any Pirates fan unexcited about another season of McDonald's twists and turns can be excused. There are a few intriguing changes thus far that bear watching, however. The lanky right-hander is digging deeper into his arsenal than he has in previous seasons. Some knowledge about McDonald's fastball is required before understanding the potential significance of these developments.

McDonald's fastball is not fit to serve as the breadwinner of his arsenal. Rather it makes more sense in the role of bread sticks, which is to say a complementary piece of the package that you forget about afterward. Start with the velocity: McDonald may touch 94 on a hot gun, but sits in the high 80s to low 90s. Location is paramount, and yet fastball location is not a strength of McDonald's game. The problem stems from his mechanics. 

Though Doug Thorburn has written about McDonald's delivery before, the blemishes are worth repeating. At the release point McDonald's balance is out of whack. During the delivery itself he doesn't have great momentum toward the plate. The result is a less-than-stellar extension. Whereas Ian Kennedy's low-90s fastball plays up due to his release point, McDonald's does not. There is something to be said about the deception McDonald's long limbs create, but at the same time these mechanical inconsistencies by the reedy righty create a below-average command profile. His fastball is too loose in the zone and he pitches up a lot—a no-no for someone with his velocity. 

Bundled together these drawbacks mean McDonald has to pitch around his fastball. He seems to know it because this season he's taken to pitching backward more often. McDonald has this big, slow curveball that he can throw for strikes or bury. It's his best pitch and it gives his at-bats a different feel. Visualize an at-bat against McDonald with and without the threat of a curveball before a two-strike count. Without it anything up in the zone is likely a fastball. With it you'd have to give it a little more thought. That instance of hesitation can be the difference between a home run and a double; between an automatic run and the chance to escape unscathed. 

If James Shields is the poster boy for pitching backward then McDonald is doing a nice job mimicking him in certain regards. McDonald has thrown 40 percent of his first-pitch breaking balls for strikes this season; Shields is at 46 percent, according to Brooks Baseball. Both have thrown back-to-back curveballs to begin five at-bats. There are a few key differences between the two. One, at least thus far, has been fastball control. Roughly 60 percent of McDonald's first-pitch fastballs have been balls this year; Shields is at 37 percent.

Beyond the differences in strike-throwing abilities lie the contrasting quality and confidence exhibited in their changeups. Shields has one of the best in the league and he'll throw it to any batter at any point. McDonald's change, though hyped in the minors as a good pitch, has been seldom used in the majors: he's never thrown more than 11 percent in a single season, and has thrown it 10 percent of the time against left-handers for his career. So consider it curious that 21 percent of his pitches when trailing in the count to left-handed batters this season have been the changeup. Granted that only compares to Shields' rate against right-handed hitters but it would represent a career high for McDonald should it remain steady, not to mention that it signifies an increased confidence in the pitch during tight spots.

Of course not every pitcher that goes backward succeeds. Ivan Nova tried last season and had a topsy-turvy year of his own. Alas Nova's failures stemmed from batters hitting his secondary pitches around the park. At the time of that piece eight of his 13 home-runs allowed were on cutters or changeups. McDonald's weakest link has been his fastball, as the rest of his pitches have True Averages-against of sub-.200. If anything, McDonald's struggles reinforce the need to go backward. 

Even if McDonald does benefit from the new strategy, he'd still be a no. 4 starter. That's valuable—especially when it costs less than market value. The catch for the Pirates is they have plenty of talented young starters on the way. Yes, those guys named Gerrit Cole and Jameson Taillon, but potential middle-to-back-end starters like Kyle McPherson, too. With another round of arbitration promising McDonald a raise, it wouldn't be a surprise to see him traded at some point before the 2014 season. Especially if another team likes what it sees from McDonald's new approach. 

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
I wonder in this case if Russell Martin has any influence, if at all, on McDonald and Nova, a.k.a. pitching backwards.
Oh, that's an interesting connection. Larry Rothschild was supposedly behind the Nova change but it stands to reason Martin might have had some input on it, too.