Bold Deceivers

After glancing at one possible ’06 Minnesota lineup last time around, I pledged to dedicate some inches to the team’s leading run preventers. Before we go headlong into the projections, I should make a note about the nature of the projections. Due to the likelihood of a long-term injury, every player’s weighted-mean projection has to come down a bit. Since pitchers make less total appearances than position players, projecting a few less starts is particularly noticeable, making it look like every pitcher is about to lose a bunch of endurance. Think of how an insurance company spreads the risk; everyone pays a little so that if something really bad happens, those individuals are covered. Likewise, pitchers won’t have 10% of separated shoulder or Tommy John surgery, but the nature of projection requires that everyone take a little bit of the risk for those who throw out their arms in the first week. Also, there are projections for more players than who will actually pitch. For example, J.D. Durbin is predicted to pitch 113 innings. Those could come at AAA or the majors, and we have no way of knowing. More simply, projection metrics are generally flawed, but using the best available data, here is a brief profile of how the 2006 pitching staff may shake out:

Johan Santana (2005, 2.87 ERA, 73.0 VORP; 2006 projection, 3.00 ERA, 62.0 VORP): Santana’s diminished VORP may look like bad news, but it is still the best projection for any pitcher in baseball. PECOTA’s final projections are weighted means, or the most likely end result rather than the high or low ends. Santana’s breakout and improvement rates indicate that he stands about a 50/50 chance of being at least as dominant as he was during last year’s Cy Young level performance. Let’s appreciate him while we can, because a power lefty like Johan comes around very infrequently.

Brad Radke (2005, 4.04 ERA, 31.9 VORP; 2006, 4.17 ERA, 26.4 VORP): Radke had a good-not-great 2005, but with a permanently ailing shoulder, I’m not sure we can ever expect much more from him. A replication of that performance would justify his presence behind Santana at the front of the rotation, but continued slippage to go with the unhealthy projection for Silva could be problematic. Any team faces the nightmarish scenario of losing one of its top pitchers to long-term or chronic injury, but with Liriano, Baker and the load of arms behind them, the Twins are better prepared for that occurrence than most.

Carlos Silva (2005, 3.44 ERA, 38.9 VORP; 2006, 4.47 ERA, 18.2 VORP): Maybe some would say that there is a thin line between groundball pitchers and batting practice and Carlos Silva rides a unicycle down the middle of it. Other than the late-season injury, everything went right for Silva in 2005 with little bits of good timing/luck/what-have-you patching over his shortcomings. Well, one shortcoming really, he lets hitters put a ton of balls in play. Still, his microscopic walk rate, low HR rate and high groundball/flyball ratio kept him in games. On top of all of that, he had one of the best double play rates in baseball, which depends on strong fielders behind him rather than his own ability. Shirking responsibility like that angers PECOTA by confounding results, and he is punished for his democratic tendencies with a fat run of ERA plopped down on his ’05 results. Obviously, his groundball and baserunner friendly style leads to a higher double play rate than average. But here’s an open question: what effect does playing on turf have on the likelihood of double plays? Does a speedier surface help runners or fielders more? If we accept Bill James’ statement that turf helps slow runners more than fast ones in its tendency to get more balls through the IF (because fast runners would have beaten out the throws anyway), it would seem that turf would do more for fielders, since they gain the advantage on the balls they actually field versus fast or slow runners.

It is also interesting to think about the makeup of the top of the Twins rotation. A few years ago when Santana was still emerging from the bullpen, Radke looked worn down and Silva was a reliever in Philadelphia, it would be hard to imagine that they would compose a dominant rotational top-three. The Twins front office often gets caught up in pretty meaningless strategic games like avoiding back-to-back lefties in the lineup or alternating power and control pitchers in the rotation. Partly by coincidence and partly by design, they have ended up with three very different pitchers who will likely start their first series of ’06. Santana is a power lefty in the Steve Carlton mold, throwing nasty breaking pitches to go with a tough heater. Radke has always struck out just enough to get by, but makes his money by getting ahead of hitters and making them swing at bad pitches. Silva, as mentioned, gives up piles of hits, strikes out no one, but throws EVERYTHING with downward movement, abusing the bottom halves of lots of AL Central bats. Together, they form an interesting control group that reiterates the maxim that there are many different ways to win. They also outline something of a blueprint, collectively demonstrating the value of not walking anyone. Sure, batters earn walks, but pitchers with this much control make it exceedingly difficult to get anything for free.

Now that I've written most of a column in the Carlos Silva section, I suppose I ought to return to the theme. Any guesses who is my favorite Twin?

Kyle Lohse (2005, 4.18 ERA, 30.7 VORP; 2006, 4.67 ERA, 14.1 VORP): Color me surprised that Lohse is back for another go-round with Gardy. If we had an office pool on when the two will come to blows, I’d be taking a date in early-mid May. I’ve said before that I think Lohse has a high ceiling, but PECOTA disagrees, seemingly saying that he topped out last year with an ERA around 4.00. He also has a 6% Breakout Rate and finds himself highly comparable to Paul Quantrill. Maybe there is more to his problems than some obscure mental hang-up; maybe Lohse is just a league average starter, and there isn’t anything wrong with that for $3 million a year.

Scott Baker (2005, 3.35 ERA, 13.7 VORP; 2006, 4.49 ERA, 16.8 VORP): I’m making a pretty big step in anointing Baker the fifth starter before pitchers and catchers have even reported, but given the team’s track record for preferring any experience, I feel confident that Liriano whittle away a year of pre-arb production in the pre-rotation Santana role. I feel especially confident considering that Ryan et al. like a traditionally constructed bullpen and they sent their only decent lefty on his way to the Angels of the Angels. This projection slates Baker for only 150 IP, and I would not be surprised if he ramped that up enough to push his VORP over 20 runs with no improvement in his projected performance.

Joe Nathan (2005, 2.70 ERA, 22.9 VORP; 2006, 2.64 ERA, 22.7 VORP): It took a couple of years, but PECOTA finally got hot for Joe Nathan. A few prognosticators used to bemoan his flyballs, but his 22.7 VORP and 31% Breakout Rate give him one of the highest ceilings of any relief ace in the league. I know it is hard to watch A.J. hoist the World Championship trophy, sing with Steve Perry and cut promos for TNA Wrestling, but that trade sure worked out well for the Twins.

Juan Rincon (2005, 2.45 ERA, 23.4 VORP; 2006, 3.54 ERA, 15.3 VORP): By VORP, Rincon was the best reliever on the team last year even with his 10 game respite. He’s a 50/50 bet to collapse this year, which is a testament to the unreliability of relief pitchers. Once again, the organizational depth of live arms makes a collapse less frightening, except in the case that it continues all year with Gardenhire unwilling to pull the plug on his locked-in 8th inning guy. With Santana, Silva, Castro and Rincon (and past Twins Blanco and Guzman), I think it would be fun to see the Venezuela team post a strong showing in the WBC as long as it doesn’t cost Santana his health.

Jesse Crain (2005, 2.49 ERA, 23.2 VORP; 2006, 4.06 ERA, 9.3 VORP): If there is one candidate for a big regression from last season on this staff, Jesse Crain is that candidate. Sample sizes make it difficult to evaluate relief pitchers on a season to season basis, making flukes more likely in the positive and negative directions. Crain did lots of work with smoke and mirrors last year, ranking near the top of the league in BABIP (BA on balls in play), an indication of good fortune. PECOTA doesn’t look kindly on those who are more lucky than good, pegging Crain with a painfully bad 72% collapse rate (to give some context, Armando Benitez is coming off of a strong year that ended with a catastrophic injury, and he has a 68% collapse rate).

Fransisco Liriano (2005, 5.70 ERA, 0.3 VORP; 2006, 3.87 ERA, 27.1 VORP): That’s quite a breakout for Liriano, assuming he gets that kind of playing time. What’s not to like? He’s set to miss lots of bats, striking out about eight batters per 9IP while giving up only about that many hits. He has good control and projects to have the same HR Rate as ground-baller Derek Lowe. Last fall, I thought it was optimistic to think he could be a bullpen regular in 2006. Now, I think the team’s playoff hopes may rest on him securing a rotation spot early on.

A recurring theme last year (and years before that) was the frustration of management slotting in players who had better backups behind them. This year’s slim championship chances could hinge on precisely that issue. The defense looks better than last year, but if the offense struggles against multiple sub-.300 OBP deadweights and the Lohse or Baker struggles through starts that could have gone to Liriano, we could have a whole new set of Luis Rivases and Joe Mayses to lament.