We played a community projection game at VeB for Kip Wells. The idea is for a bunch of people to make blind guesses about how well Wells will perform in 2007 and hope that the wisdom of the masses will give you a reasonable estimate of what he’ll actually do by averaging the guesses. It’s a decent enough way to kill time while waiting for Spring Training games to begin. As is describing in excrutiating detail how I made my guess…
Here’s the line I submitted:
190IP 198H 69BB 140K 18HR 16-9 3.85
You might think that I pulled those numbers straight out of thin air, but it’s less direct than that. I assumed he’d have his first healthy season since 2003, when he pitched in 197 1/3 innings over 31 starts. I assumed that he’d pitch about the same numbers as Jeff Suppan did last season—and used the exact number as Suppan’s 2006, 190.
Once I had assumed how many innings he’d pitch, I could make what I think are pretty good guesses about the rates at which he’ll allow the hits, walks, homers, and how frequently he’ll strike out batters.
Kip Wells has allowed 9.32 hits per nine innings over his MLB career, skewed slightly higher by last season, when he was injured and pitched terribly. In 2003, he allowed 7.8 H/9, as a peak benchmark. I figured he’d be encouraged to induce groundballs—to be aggressive early in the count, to pitch low in the strikezone, to prefer the batter to put a weak ball into play rather than pussy-foot around the zone and end up walking the batter. Inducing groundballs means more hits squirting through the infield, so I made a reasonably conservative rate of 9.4 hits per nine innings, or 1.044 BB/IP. Multiply that walk per inning rate by 190 innings, and you get the 198 hits total.
Wells has historically walked far too many batters. Over his career, he’s walked 4.09 batters per nine innings, compared to Suppan’s 2.95 or Carpenter’s 2.75 BB/9. I figured that if Wells buys into Duncan’s program described in the previous paragraph—and all indications are that he chose to sign with the Cardinals to get with his program—he’ll be working ahead of the count more frequently than in the past and should trust his defense enough to avoid nibbling when he’s behind in the count. I guessed he’d walk significantly fewer batters while recognizing that Wells won’t likely have the same control as Suppan does—I gave him a rate of 3.25 BB/9 or .36 BB/IP. That’s pretty optimistic, that in an average six inning outing, he’d only walk two batters. That works out to 69 walks, given the 190 IP assumption.
Wells is a pretty good strikeout pitcher, much better than Suppan. Between 2002-2004, he struck batters out at rates of 6.08, 6.70, and 7.55 per nine. Kip Wells’ third closest comparable pitcher through age 28 is Jason Schmidt. (I used age 28 instead of 29 since Wells has been injured the last two seasons.) Schmidt’s a better pitcher than Wells, but the similarity is remarkable, in many ways. Schmidt’s peak season was 2003, his age 30 year, and I optimistically guessed that Wells would similarly have a strong year at that age. The rate I guessed was 6.65/9, approximately in the middle of those early years that I believe show his true talent level, if it’s not higher. That worked out to the 140 strikeouts.
Allowing home runs hasn’t been a serious problem for Kip—his career rate is 1.02/9 innings. If he held that rate, he’d allow 22 HR over a 190 IP season. With Busch III being a pitcher’s park and assuming that Wells looks to keep the ball on the ground, I guessed a rate of 0.85 HR/9. That’s Chris Carpenter’s 2006 rate, and in retrospect, that’s too optimistic. I’d be surprised, though, if he didn’t beat that career rate by ten points or so, a difference of 2 HR. Of course, he’ll be helped by not having to face the Cardinals lineup.
I arbitrarily made up the win-loss numbers since I don’t put much significance in pitcher wins. That’d be the same record as Jeff Suppan had in his first season with St. Louis—you may have noticed that I don’t think there’s much difference between Wells and Suppan. Soup has better control, Wells has better stuff.
For the ERA guesstimate, I used a bastardized version of Tangotiger’s FIP formula, using 3.00 as the normalizing term instead of 3.20, to arbitrarily optimistically credit the Cardinals defense.
FIP = (13HR + 3BB - 2K)/IP + 3.00 = (13*18 + 3*69 - 2*140)/190 + 3.00 = 3.847
So yes, I just made those numbers up, but I tried to make reasonably educated guesses based on clearly stated assumptions. It’ll be fun to see whether they were at all close to decent at predicting his actual performance. Hopefully, they do.