Decoding ZiPS, Part 2: the Colorado Rockies Lineup

facebooktwitterreddit

Credit: Chris Humphreys-USA TODAY Sports

Last week I wrote about how that the Colorado Rockies starting rotation out-performed its 2013 ZiPS projections and explained why it is slated for regression this coming year. But I also suggested that because the rotation’s collective forecast for 2014 is more optimistic than 2013, the rotation is at a better starting point and should provide more value in 2014 with a healthy Brett Anderson, even accounting for regression from Jorge De La Rosa, Jhoulys Chacin, and Tyler Chatwood. Not to suggest that health is an easy accomplishment. It’s elusive to be sure, especially for pitchers. But for Rockies fans, counting on healthy pitchers that might lead to success is more settling than hoping for ERAs under 6.00 as a path to mediocrity. Such is the tantalizing environment in which we find ourselves in 2014. Today, I want to go back to the 2013 ZiPS projections to examine offensive forecasts and compare them to team production in 2013, and then consider what we can expect from the position players in 2014 based on the new estimations.

Position players need to be evaluated in terms of offensive and defensive worth, as well as the total value he offers when considering both. In addition to including each player’s triple slash, I included his weighted on base average (wOBA). If you are unfamiliar with the statistic, it takes into account all forms of offensive production (singles, doubles, walks, etc.), weighs them according to value in terms of run production (a home run is more valuable than a single), and the final number is weighted against league average. While league average fluctuates year to year, it has been hovering around .315 for the past few years, so that’s the baseline. Conveniently, wOBA is scaled similar to on base percentage, so if the wOBA would make an excellent OBP, such as Troy Tulowitzki’s, then it is likewise an excellent wOBA. If it suggests a poor OBP, then it is also a poor wOBA—think DJ LaMahieu. Finally, a great wOBA might not necessarily translate to an excellent total value, as measured by WAR, because WAR for position players takes defense into account in order to calculate the total value of a single player—think Michael Cuddyer. It’s important to keep defensive value in mind. The 2013 ZiPS projections:

PAAvg/OBP/SlgwOBAzWAR
Wilin Rosario462.263/.301/.470.3271.5
Todd Helton330.264/.355/.426.3380.8
Second Base
    Josh Rutledge670.280/.355/.445.3302.4
    LaMahieu598.295/.328/.379.3040.8
Nolan Arenado644.280/.319/.433.3201.6
Tulowitzki461.294/.362/.535.3784.6
Carlos Gonzalez629.291/.360/.531.3734.0
Dexter Fowler620.280/.316/.445.3522.4
Cuddyer509.284/.340/.499.3581.3

Similar to the rotation’s forecast, ZiPS was not terribly high on the line-up as a whole, and this is even after inserting Arenado into Chris Nelson’s projected WAR of roughly zero. Nelson was forecasted to be a replacement level player, no more valuable than a warm body from AAA. Arenado, Cuddyer and Helton were projected to produce just under two WAR, which is generally considered to be league average. Rutledge, Fowler, and Rosario (plus his back-up) were each estimated to produce just above average. Finally, ZiPS projected Tulowitzki and Gonzalez to be the two truly exceptional players for the Rockies. I included both Rutledge and LaMahieu to demonstrate how difficult it is to project players with minimal playing time in the Major Leagues. After his excellent showing late in 2012, many Rockies fans were excited to see Rutledge occupy second base permanently. ZiPS was also high on him, and is somewhat optimistic for 2014, as we will see. We know, of course, that Rutledge’s early struggles led to LaMahieu assuming the starting role for most of the season. The projected totals for the daily lineup were 18.6 WAR with a full season Rutledge, and 17 with LaMahieu.

This is how things turned out:

PAAvg/OBP/SlgwOBAfWAR
Rosario466.292/.315/.486.3482.2
Helton442.249/.314/.423.322-0.8
Second Base
    Rutledge314.235/.294/.337.281-0.4
    LaMahieu434.280/.311/.361.2950.8
Arenado514.267/.301/.405.3082.7
Tulowitzki512.312/.391/.540.4005.6
Gonzalez436.302/.367/.591.4084.9
Fowler492.263/.369/.407.3472.2
Cuddyer540.331/.389/.530.3962.5

The total WAR is 19.7, which counts both Rutledge and LaMahieu, as they generally contributed to the second base production for 2013. The lineup, like the starting rotation, out-produced its projections, although only by about a win. The comparison is instructive. First, Todd Helton was really quite bad in his final year. He had a triple slash that was below average for all measures and an average wOBA. His negative WAR was a result of poor defense. Helton was actually worth negative value for the Rockies in 2013, whether your preferred WAR metric comes from FanGraphs (-0.8), Baseball Prospectus (-0.6), or Baseball Reference (-0.4). One season won’t sully Helton’s extraordinary career profile, but under-performing a below average forecast is a sure sign that it’s time to retire. The other player who saw a significant differential between projection and performance based on defense played across the diamond from Helton. The accolades Nolan Arenado received while in the minor leagues had mostly to do with his bat. In his first major league season, however, Arenado hit worse than his mediocre projections. But his defense was so good that it resulted in a nearly three win season. Finally, a cursory glance at the above table confirms that Tulowitzki and Gonzalez are, indeed, the superstars of the Colorado Rockies, which is news to nobody and is not expected to change in 2014, as noted in the following table:

PAAvg/OBP/SlgwOBAzWAR
Rosario474.265/.295/.458.3251.5
Justin Morneau484.280/.343/.457.3440.8
Second Base
    Josh Rutledge587.271/.317/.416.3221.3
    DJ LaMahieu609.286/.320/.376.3020.8
Arenado619.284/.319/.440.3262.7
Tulowitzki529.296/.369/.522.3795.2
Gonzalez553.286/.353/.530.3773.1
Left Field
    Corey Dickerson556.267/.315/.472.3360.5
    Charlie Blackmon497.266/.316/.413.3190.0
    Drew Stubbs546.253/.321/.396.3201.2
Cuddyer503.290/.348/.489.3590.8

Of every component of the Rockies 2014 roster, ZiPS is most unkind to the value of the lineup. Even if the Rockies gave full seasons to the three possible left fielders—which could only happen if we fielded four outfielders, and that’s against the rules—the estimated WAR total is only 17.9. Or, about two wins fewer than the actual performance of 2013 (personnel changes duly acknowledged). The primary lesson learned from the table above is that the 2014 Rockies might be one of the more difficult teams to project. The starters in left field and second base are unclear right now and a topic of debate. Additionally, left field and first base will be subject to quite a few splits, as Morneau should not be trusted to hit against left-handers and Drew Stubbs should sit against right-handed starters. And that is without even mentioning Brandon Barnes and Charlie Culberson as two guys who might challenge for a roster spot. The projections above also reveal how valuable defense is. Cuddyer is expected to produce quite well offensively, but his defense in right field is expected to be even more of a liability in 2014 than it was in 2013, thus limiting his overall value.

But defense can also be made an overall strength rather than a potential weakness. While no offseason transaction should be considered in isolation, some were still better than others. I think that the best move the team made this offseason will contribute to defensive run prevention and the worst transaction will detract from it. In my mind, the best deal was trading Josh Outman for Drew Stubbs. Stubbs likely will not be the everyday left-fielder, but coming off of the bench he can provide defensive value in left field late in games, while also neutralizing left-handed specialist relievers that might be called on to face Dickerson or Blackmon. ZiPS projects him to be as good a defender as Carlos Gonzalez in center field, which translates to even higher value in left field. Additionally, the move only cost the team a very replaceable left-handed reliever, who has in fact already been replaced twice-over by Boone Logan and Franklin Morales.

I think that the worst transaction was the Morneau signing because of the ripple effect it has. Without Morneau, the most likely candidate to start at first base would have been Cuddyer, which would have been a better option than right field. First, Cuddyer is not going to reproduce his offensive production from 2013. He was immensely helped by a BABIP of .382, which he won’t be able to replicate on purpose in 2014. Second, Cuddyer is actually projected to improve his defense in 2014, and it’s not because he spent the offseason watching Tom Emanski VHS tapes. He’s a 35 year old outfielder due for a pretty big offensive regression. So the Morneau signing blocked what I thought was an essential and intuitive lineup change. Of course, as fellow Rox Piler Lev Cohen notes, Morneau might see his bat revived in Denver, but I’m not optimistic that Morneau will produce enough value to justify the signing—I think the ZiPS projection looks about right.

I read the forecast for the lineup in a similar fashion as the rotation. Given the personnel, I can envision a lot of moving parts over the course of the season and within games. Walt Weiss can take advantage of severe splits depending on the opposing starting pitcher and neutralize left-handed specialists. Additionally, he can take advantage of the quality defenders that will be found on the bench that can serve as in-game platoon partners, especially in the outfield. While the projected WAR of every outfielder is slightly diminished because ZiPS takes into account Coors Field’s expansiveness, I ultimately think that in terms of contributed value on a game-by-game basis, the team’s newfound depth can lead to more wins. Whether it adds up to enough wins for a .500 record is another matter. One that can be addressed once baseball games are actually being played and depth charts reordered due to on field performance—and that time can’t come soon enough.