The newly confusing Roy Oswalt

Roy Oswalt was a prospect, then he was called up. He was good immediately, and he was good for a very long time. He had some injury problems, and then he wasn't the same. The end.

That's how the Roy Oswalt story looked like it was going to go. At the end of 2011, if you told someone that Oswalt was going to pitch 59 innings between the start of 2012 and June, 2013, you would have figured his arm exploded. It would have been a shame, but it would have been a typical end to a typically great career, if there is such a thing. There are great careers with ups and downs, stops and starts. And there are great careers that go from Point A to Point Retire without anything unexpected. Those are pretty rare. Oswalt was on his way to one of the latter careers.

Or, to put it another way, Oswalt has good company on this list of pitchers with 20-start seasons with a park-adjusted ERA better than the league average. His 11 seasons have him tied with Hall of Nearly Greaters like Orel Hershiser, Kevin Appier, and David Cone, which is just about right. Except Oswalt's seasons came consecutively, which is a rare feat. Even when he was hurt, he wasn't hurt that badly. He just kept chugging along, pitching better than most of his peers, throwing 200 innings more often than not.

For 11 years, Oswalt was the least enigmatic pitcher in baseball.

Except he didn't retire. He did a weird dance with interested teams before the 2012 season, trying to force his way into a golden situation (close to home, for a lot of money.) He eventually signed with the Rangers, and he gave them 57 bad innings. He had home-run problems, and he barely made the postseason roster. Suddenly the predictable starter was a little unpredictable. Maybe the layoff messed him up, maybe he was a poor fit for the Ballpark in Arlington, or maybe he had been declining steadily while with the Phillies and it just took a while for us to notice.

For the first time in his career, Oswalt was an enigma. And over the offseason, I'm guessing that teams treated him like an enigma, if not like someone with a communicable disease. Here's why I'm guessing there wasn't any interest: He signed an incentive-laden minor-league deal with the Colorado Rockies. Denver isn't especially close to Mississippi. Coors Field is the absolute worst park for a pitcher on a one-year deal. So while I don't want to assume Oswalt was desperate, I'll feel comfortable assuming he didn't have several good offers on the table.

And this is interesting now because Oswalt is close to returning to the majors:

Oswalt, a three-time All-Star with the Astros, gave up only five hits and two walks on 102 pitches, striking out seven. It was his strongest start in Double-A after he allowed just one run in seven innings Sunday.

For 11 years, Oswalt was the least enigmatic pitcher in baseball. And now he's the most enigmatic pitcher in baseball. The transition was swift and thorough. Oswalt is either still a quality pitcher who suffered through a stretch of bad luck in Texas, or he's a 35-year-old land mine who will take his homer-prone ways to one of the great homer-prone ballparks in history.

Maybe he's a little of both. Point being, is it possible to have any confidence at all in a prediction for Oswalt's Colorado tour? If you're convinced he's going to be awful, you're basing that on a 57-inning sample from last year (in which he posted the best xFIP since his rookie season). If you're convinced he's going to be good, though, you're basing that on a history of success that isn't that recent. Either way, there's no way to feel confident in whatever prediction he makes.

He looks the same as he did with his later years as a Phillie:

But he had flashes of brilliance with the Rangers last year, and they were usually followed by balls flying over the fence.

For 11 years, Oswalt was the least enigmatic pitcher in baseball. Now he's a freak. A fascinating freak who chose the most fascinating ballpark possible for his comeback. Make a prediction if you want, but know that you're probably going to be wrong.