:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/45630280/usa-today-8153668.0.jpg)
The Miami Marlins are still keeping tabs on the James Shields market. While the team may still be a long shot to sign the righty starter, the fact that the team is supposedly involved in the race is a positive sign. Miami should at least consider Shields in the right deal, because at this point, he represents an upgrade over the roster's current final starting pitcher options.
This was probably not the same situation just a few months ago. The Fish had Nathan Eovaldi and had a plethora of marginal but still worthwhile pitching prospects who were close to the majors, including top prospect Andrew Heaney, Anthony DeSclafani, and Justin Nicolino. Fast forward a few months and the Fish are up a Mat Latos but down Eovaldi, Heaney, and DeSclafani. With the uncertainty surrounding Dan Haren and whether he may or may not retire even if he shows up for Spring Training, the Fish are now in a more difficult spot in terms of filling in their rotation before Jose Fernandez arrives by midseason. The team's fifth starter options are marginal, and if an injury occurs or if ineffectiveness from questionable starters like Tom Koehler or Jarred Cosart strike, the team could be down a significant amount of talent.
The situation has changed since we deemed Shields a bad idea at the start of the offseason. But how much of an impact will he make? We will use a few different methodologies based on projections to make that determination.
What follows is the team's current Steamer projections-based FanGraphs depth chart for the Marlins' starters.
SP Marlins
Name IP K/9 BB/9 HR/9 BABIP LOB% ERA FIP WAR Jose Fernandez 
122.0 10.4 2.9 0.7 .305 74.6 % 2.94 2.82 2.7 Mat Latos 179.0 6.6 2.6 1.0 .302 70.0 % 4.11 4.04 1.8 Dan Haren 162.0 6.8 1.9 1.1 .302 71.3 % 3.92 3.92 1.6 Henderson Alvarez 160.0 5.6 2.1 0.7 .310 68.8 % 3.90 3.70 1.7 Jarred Cosart 102.0 6.4 4.0 0.7 .309 68.0 % 4.44 4.09 0.6 Tom Koehler 65.0 7.0 3.5 0.9 .306 69.1 % 4.41 4.20 0.3 David Phelps 66.0 7.9 3.2 0.8 .305 71.7 % 3.79 3.78 0.7 Andre Rienzo 66.0 7.2 3.8 0.8 .308 68.7 % 4.36 4.06 0.4 Brad Hand 9.0 7.5 3.6 0.8 .306 71.0 % 3.93 3.83 0.1 Justin Nicolino 9.0 4.8 2.3 1.0 .303 67.8 % 4.53 4.40 0.0 Total 939.0 7.1 2.8 0.8 .305 70.2 % 3.94 3.80 10.0
If we just assume this configuration of innings and FanGraphs Wins Above Replacement provided by the above pitchers, we can easily carve out a spot of innings for Shields. The current depth charts have Koehler, David Phelps, and Andre Rienzo (acquired for reliever Dan Jennings from the Chicago White Sox) taking up the vast majority of the innings not occupied by the top four starters and the returning Fernandez. The total innings indicated for those pitchers was 215 innings. If we replace those with Shields's projected 201 innings and tack on fractions of a win, the Marlins should improve by about 1.6 wins.
You can compare the two groups of pitchers by conglomerate stats, both rated at 201 innings.
| Player, 2015 Steamer Proj | K/9 | BB/9 | ERA | FIP |
|---|---|---|---|---|
| James Shields | 7.8 | 2.2 | 3.63 | 3.57 |
| Group | 7.3 | 3.5 | 4.19 | 4.02 |
It's pretty clear Shields is significantly better than the ragtag group as per Steamer projections. But the depth chart does not accurately reflect that these ERAs for guys like Rienzo and Phelps are based on split starting and relief roles. Naturally, both players will be better and lower their ERAs in relief, and they are not likely to pitch as well as their overall ERA as starters. If we look at their player pages, you can see that the projections are drastically different. Let's use a prorated projection for each guy based on his WAR listed on the player page and the number of innings he's expected to start. You get a better sense of how good those guys actually are.
| Player, 2015 Steamer Proj | Total IP | Starter IP | Starter WAR |
|---|---|---|---|
| David Phelps | 92 | 66 | 0.3 |
| Andre Rienzo | 97 | 66 | 0.0 |
Using these totals, you would get an advantage of 2.3 wins switching from that group to Shields. That's more than two wins, which would make a huge difference in the Marlins' chances for a playoff berth. By FanGraphs and Steamer's projections, the Fish are currently the seventh-ranked team in the National League, just behind the defending champion San Francisco Giants at a projected 83 wins. Pick up two wins and the Marlins are right in the thick of what may be a tight race for those final two Wild Card spots.
How do other projection systems stack up these players? Baseball Prospectus's PECOTA system does not like a vast majority of the Marlins' starters, projecting only Mat Latos to be worth more than one win in the rotation. If we replace the lowest-ranked pitcher in the rotation and knock off Phelps's starter innings, we can come up with a reasonably close estimate of Shields's expected 193 innings. Shields is projected to put up 1.9 wins by PECOTA, and the two players he would be replacing are expected to produce 0.1 wins, giving Miami almost a two-win edge.
Looking at multiple systems and multiple assumptions, we can reasonably come up with a two-win improvement in signing Shields in 2015. The question the Marlins have to answer is how well they think he will age and approach his 200 innings and whether that two-win improvement is worth the likely $50 million-plus commitment over multiple seasons.