Talk of a "mid-90s fastball" gets tossed around a lot in describing hard-throwing pitchers, but thanks to Fan Graphs and Pitch-F/X we know that there are very few guys who consistently work at that velocity. In fact, none of the 99 pitchers who logged at least 150 innings last season had an average fastball that cracked 95 miles per hour. Even lowering the cutoff to 50 innings so that relievers are included results in just 11 of 316 qualified pitchers averaging at least 95 miles per hour with their fastball in 2008.

Most big-league pitchers work at 88 to 92 mph, an excellent fastball is 93 or94, and a handful of relievers in each league consistently crack 95. All of which is a long way of saying that "mid-90s fastball" likely gets used too often when describing guys who occasionally reach 94 or 95 while actually working at 92 or 93. Juan Morillo is among the rare pitchers for whom "mid-90s fastball" is not only an accurate description but perhaps even a bit of an understatement.

Prior to being claimed off waivers by the Twins this weekend, Morillo had thrown 135 fastballs in the big leagues, all as a member of the Rockies, and they averaged 96.5 mph. Morillo made his Twins debut Saturday night, working the final inning of a 9-2 victory over the Angels and throwing 13 pitches. One was an 88-mph slider that struck out Mike Napoli, and the other 12 were fastballs clocked by Pitch-F/X at 95, 96, 97, 97, 97, 97, 98, 97, 96, 97, 97 and 97 mph.

Save for perhaps a healthy Joel Zumaya, he may throw consistently harder than any pitcher in baseball. Not surprisingly, he racked up 126 strikeouts over 126 innings between Double-A and Triple-A after becoming a full-time reliever in 2007. So why was this strikeout machine with a rare, legitimate mid-90s fastball available for absolutely nothing on waivers? Because he can't throw the ball over the plate. He walked 87 batters in those 126 innings, including an astounding 7.9 walks per nine innings at Triple-A.

For the better part of a decade, the Twins have built an organization full of strike-throwing machines with high-80s or low-90s fastballs and strong off-speed pitches, but Morillo is as far from fitting that mold as it gets. He's shown no ability to command what is an overpowering fastball and his "off-speed pitches" consist of a little-used changeup and a slider that could pass for a fastball. And that's exactly why the Twins were smart to essentially trade Philip Humber (who was designated for assignment) for Morillo.

Since he came over in the Johan Santana trade, many Twins fans have simply assumed that Humber is a good prospect because he was the No. 3 overall pick in the 2004 draft after a brilliant college career and once received a ton of hype coming up through the Mets system. However, his stuff hasn't been the same since undergoing Tommy John elbow surgery in July 2005, and little in Humber's post-surgery performance suggests that he's capable of becoming more than a fifth starter or long reliever.

Humber certainly still has some value and keeping him around to soak up low-leverage relief innings and perhaps make a spot start or two would have been just fine, but the payoff is minimal, given that he's already 26 and the Twins develop pitchers in such a way that they will rarely be lacking in back-of-the-rotation starters or long relievers. Morillo won't even have Humber's limited application if he can't find the plate, but simply going from "awful" to "bad" control would give him much more upside.

Whether pitching coach Rick Anderson can work his magic on Morillo remains to be seen and likely qualifies as improbable, but he's certainly in the right place when it comes to throwing strikes and the downside for the Twins is minimal as long they keep him in a low-leverage role. If they can get Morillo to harness his extraordinary fastball, they will have stumbled upon a late-inning relief option unlike nearly every other pitcher in the system. And if not, they can throw a rock and hit another guy like Humber.