A clear-eyed assessment of Ryan Dempster's trade value
Of the deadline-related deals floating around in the buzz-o-sphere, the most imminent one appears to be the trade of Ryan Dempster, who happens to be the hottest pitcher in baseball by dint of a 33-inning scoreless streak. As many as 10 teams were said to be interested in the 35-year-old Cubs righty over the weekend, and since then, we've learned that the Dodgers are the leading suitor and that the Red Sox have dropped out. The Cubs are eager to move him sooner rather than later, not only to take advantage of the hot streak but because they want to turn their attention to trading fellow righty Matt Garza, who's also on the block.
How much of an impact will Dempster have? That answer depends upon a realistic assessment of his current ability. Right now, he's leading the majors with a 1.86 ERA, but it's hardly a given that he'll continue to exhibit Cy Young-caliber form. That ERA is less than half his career mark of 4.30, and right around half of the 3.61 mark he's posted since moving back to the rotation in 2008. Dempster's doing this despite striking out fewer hitters per nine than he has since his return to the rotation, but he's also trimmed his walk and home run rates:
Dempster is also getting a tremendous amount of help from his defense. His batting average on balls in play this year is .244, down from .331 last year, and from a more normal .304 from 2008-2011. His Fielding Independent Pitching, an estimate of his ERA based upon his strikeout, walk, hit-by-pitch and homer rates that is, in smaller sample sizes, a better predictor of a pitcher's ERA than his actual ERA, is a more realistic 3.17. That's down from 3.87 last year (his actual ERA was a swollen 4.80) and 3.78 from 2008-2011.
In smaller sample sizes, FIP is a better predictor of a pitcher's ERA going forward than his actual ERA, so we can use that figure for some back-of-the-envelope estimates to compare the impact of trading for Dempster versus that of trading for a bigger name such as the Phillies' Cole Hamels. For illustrative purposes, let's look the incremental gains to be had by the Dodgers, who are said to have proposed a deal. While Los Angeles' offense ranks second-to-last in the league in scoring (3.77 runs per game) and its run prevention ranks third (3.72 per game), it now has two starting pitchers on the disabled list in Ted Lilly, who's been out since late May due to shoulder inflammation and didn't resume throwing until just before the All-Star break, and Chad Billingsley, who went down this week due to elbow inflammation. While general manager Ned Colletti is still in the market for a big bat or two — a thought that makes Dodger fans shudder given the current lineup's sinkholes at first base, third base and leftfield that bear his stamp — preventing the other team from scoring is just as important when it comes to winning games.
The comparison to be made is for adding Dempster or Hamels versus keeping the status quo (assuming Billingsley's quick return to health) by replacing fifth starter Nate Eovaldi. If we assume the following:
• that each pitcher will continue to average the same number of innings per start over 13 starts (the number of full turns remaining for the Dodgers, who have played 93 games)
• that each pitcher's current FIP accurately reflects his true talent level
• that the quality of the Dodger bullpen is constant at 3.54 FIP (based upon their year-to-date peripherals), even though the relievers who pitch in the sixth inning in relief of the fifth starter are probably worse than those who pitch in the seventh or eighth inning
...then this is what we get:
|Pitcher||IP/GS||IP remaining||FIP||Runs||Bullpen IP||FIP||Bullpen Runs||Total Runs|
Based upon their 2012 work, Dempster actually grades out as slightly better than Hamels, whose 3.07 ERA and 3.33 FIP are quite closely matched, and whose 2008-2011 FIP (3.52) isn't that far off of Dempster's 3.78. Both are about 10 runs better than Eovaldi, perhaps a couple more if we were to more accurately model the bullpen. Given the rule of thumb that the marginal decrease of 10 runs saved equals one extra win, that's about one win added for the name-brand pitchers over the minor league callup, and basically a wash between the Phillies' co-ace and the Cubs' lightning-in-a-bottle mid-rotation guy.
If we instead use their 2008-2012 ERAs — above 400 innings, the advantage of using an estimator disappears – and innings per start (3.61 ERA in 6.3 innings per start for Dempster, 3.26 in 6.6 innings per start for Hamels), the difference swings to Hamels, but only by 3.3 runs (43.4 versus 46.7), with Eovaldi (4.38 FIP in a career of 86.2 innings) coming in at 52.8 runs. So the advantage between Hamels and Eovaldi is 9.4 runs (around one win), and between Demspter and Eovaldi is 6.1 runs (a little more than half a win).
Those are pretty small gains, ones which make it hard to justify the difference in terms of talent Dempster and Hamels will fetch — the former maybe two B-grade prospects, the latter at least one blue-chipper and probably at least two other prospects — versus that of standing pat. All of these estimates are somewhat complicated by the injuries in the Dodger rotation. Eovaldi may at some point be replaced by Lilly (3.93 FIP this year, 3.68 ERA from 2008-2012) or Billingsley (3.43 FIP this year, 3.80 ERA from 2008-2012), but neither may be as durable or effective as they were before the injury. Furthermore, Eovaldi isn't likely to start a playoff game, should the Dodgers make it that far, whereas the battle-tested Hamels might even start Game 1 ahead of resident ace Clayton Kershaw, and Dempster might be positioned as a Game 2 starter ahead of reclamation projects Chris Capuano and Aaron Harang.
The Dodgers are 2 1/2 games out of first place in the NL West and two games back in the NL Wild Card race (with a 23.8 percent chance of reaching the postseason according to Baseball Prospectus' Playoff Odds). Given that, any upgrade from Eovaldi to either Hamels or Dempster will not by itself be enough to push the Dodgers into the playoffs.