I thought maybe I'd seen it all.

After all, I'm 36 years old and have been obsessed with baseball for half of those years (the second half). I don't doubt that baseball has many small surprises in store for me yet, but big surprises?

I thought not, until yesterday.

Yesterday I saw a hard knuckleball.

What's a hard knuckleball? It's a knuckleball that's faster than a slow knuckleball. And the slow knuckleball is the only knuckleball we've seen in the major leagues for I don't know how many years. Tim Wakefield throws the slow knuckleball, as does Steve Sparks, as did Tom Candiotti and Charlie Hough.

The fast knuckleball is to the regular (slow) knuckleball as the slider is to the curve. Sort of. It doesn't move as much but it moves quicker. Jared Fernandez throws it, along with the slower version. According to the Cincinnati Enquirer, Fernandez threw approximately 115 knuckleballs on Sunday at speeds ranging from 65 to 76 miles per hour. The former speed is typical for 21st-century knuckleball pitchers; the latter is not.

There's evidence that Hall of Famer Jesse Haines and Fat Fred Fitzsimmons both threw "hard" knuckleballs back in the 1920s and '30s, but theirs supposedly broke straight down like a spitball. In the 1940s and '50s, a lot of pitchers threw a knuckleball as a complementary pitch, and I have little doubt that some of them threw the hard knuckler, because it's a lot easier to disguise if you do throw other pitches. But sometime in the late 1960s or early '70s, it became Common Wisdom that if you're going to throw the knuckler, you have to throw it nearly all of the time. And the new breed threw it slo-o-o-w-w-w, rarely faster than 70 mph.

As it happens, Jared Fernandez does throw other pitches approximately 15 percent of the time. Well, sometimes. Yesterday he threw only four or five sliders and six fastballs (one of which was deposited over the fence by Julio Lugo for a two-run homer) while pitching seven innings to gain the first major-league victory of his career.

According to the Enquirer, it's not only the speed of Fernandez's knuckleball that makes it different. Every knuckleball pitcher for at least a generation hasn't actually used his "knuckles" to grip the ball. In fact, virtually every noted knuckleballer since 1908 has used his fingernails; for a while, the pitch was even called "fingernail ball" in some quarters.

But as Fernandez told the Enquirer, "I throw my knuckleball differently than they do. They dig their fingernails right in and do a push; I put my knuckles right on it and I throw it."

Because he's throwing rather than pushing, he can get more leverage on the ball, and thus more velocity.

Is Fernandez for real? I wouldn't bet against him. From the 1999 season through today, he's compiled a 3.96 ERA in 586 Triple-A innings. He's 30, but of course that's when most knuckleball pitchers are just getting their starts. Most successful knuckleballers of the last half-century didn't establish themselves in the major leagues until they were at least 30.

And if the Reds are going to be a factor in the NL Central race, they may need Fernandez to establish himself -- and soon. Because when your rotation is built atop a foundation of Chris Reitsma and Elmer Dessens, you've got problems. The Reds need Fernandez or Ryan Dempster -- or perhaps both -- to contribute down the stretch, or they're going to have trouble finishing in second place, let alone first.

It's suddenly become fashionable to discuss the Astros as contenders, despite their sub-.500 record. I would tell you that I saw it coming, but that's easy to say now so I won't bother (even though I did see it coming).

Last week, I ran the runs scored and allowed, along with the records we'd expect from those numbers, for both West divisions. Well, here's the same information for the NL Central (the only other division for which it's relevant):

          Scored Allowed   W-L
Cardinals   401   363     48-40 
Astros      425   409     47-43  2.0
Reds        392   413     43-48  6.5

In the real world, the Astros are actually six games behind the Cardinals and two-and-a-half behind the Reds. But it's difficult to reasonably argue that, their records notwithstanding, the Reds are fundamentally better than the Astros.

But are the Astros better than the Cardinals? Hard to say, though as Cardinals GM Walt Jocketty recently acknowledged, "We have several question marks (in the starting rotation.). ... I've spent most of the last few days on the phone talking to clubs and people are familiar with our situation. So they're not exactly bending over backwards to help us."

Best case? Woody Williams stays healthy, rookie Jason Simontacchi continues to bid for the Rookie of the Year award, and Bud Smith returns to his 2001 form. But best case rarely applies to young pitchers, which leaves room for hope if you're in Cincinnati or Houston. Particularly Houston, because the Astros have just as much talent as anybody in that division. So now there's just the small matter of making up six games in what might be two-and-a-half months ... or what might become significantly less time than that, if the billionaires and millionaires can't figure out how to divide their spoils.