On July 10, 2014, Dave Cameron at Fangraphs posted an article explaining why the Rockies were making a mistake by keeping Jorge de la Rosa at the trade deadline and wanting to resign him that offseason. They did, at $25M over two years and it's safe to say that Cameron was correct.
To start with - was the contract worth it? The going average cost of a win that offseason was around $6M. With one start left this season that will probably not move the WAR needle, DLR has posted 2.7 WAR (B-Ref) combined the last two years (a ghastly -0.1 this season). That comes out to $16.2, so while not a gross overpay in terms of dollars, the Rockies overpaid by 54%. Fangraphs pegs him at 2.2 WAR total, which looks even worse. Teams like the Dodgers can afford to overpay without feeling an impact, but small and mid-market teams cannot.
Side note: I'm also confused by the 2.8 WAR he put up in 2015. He was credited with 2.4 WAR in 2014, even though his 2015 numbers were inferior in almost every category, most notably that he pitched 35 fewer innings in 2015 (just 149 total). The only thing that improved was K/9 while every other number was either flat or worse (BB/9 was notably worse). To be fair, Fangraphs only gave him an increase of 0.1 WAR in 2015 over 2014.
But the most important notes from that article revolve around innings pitched. Cameron pointed out that DLR had only eclipsed 150 IP twice, though he did pull it off again in 2014. However, through a combination of injuries and ineffectiveness, DLR failed to reach 150 in each of the last two seasons, even being demoted to the bullpen for part of 2016. This leads to the most important reason why resigning DLR was a mistake - the impact on the bullpen, as noted in the column:
"Currently, de la Rosa could best be described as a below average starting pitcher. He’s never thrown 200 innings in a season — he’s only thrown as many as 150 innings twice, in fact — so you can’t even make the innings-eater argument, or suggest that he’s helping to provide necessary rest for an overworked bullpen. He doesn’t pitch a large quantity of innings, nor does he perform all that well when he is on the mound. And he’s firmly entrenched in the decline-phase of his career. With a couple more years of aging, he probably won’t even be a viable big league starter."
In 2015, DLR averaged 5.7 IP per start, which means the bullpen was absorbing more than 3 innings every time DLR left the game. Remember, this was a bullpen whose unofficial mascot (on this site) was Gassy the Gascan. It's not unreasonable to believe that the extra work contributed to a terrible bullpen performance. It would be fascinating to see someone figure out the negative WAR impact to the bullpen.
The resigning also showed poor long-term planning. Very few people believed the Rockies to be contenders in 2015 or 2016, so why waste money trying to pick up, at most, a couple of wins that wouldn't make a difference in the standings? I don't know the full set of rules regarding money, but it seems like saving that $25M for a 2017 signing would have been prudent. Given how DLR has pitched, giving those starts to Christian Bergman or Yohan Flande at league minimum would not have made much of a difference in the results the past two years, but those guys are expendable when the young prospects are ready for their call ups. As we have seen with Parra and DLR, the Rockies are loathe to eat contracts when the players aren't accused of domestic abuse and those guys end up taking playing time away from the young players expected to contribute in the future who should be getting valuable experience.
Like the Cameron article, this isn't meant to be a slight against DLR. DLR is a fan favorite and has been one of the best pitchers the franchise has employed. That's worth something. But if keeping DLR ends up costing the Rockies a much needed free agent this offseason because of that money they didn't save (think back to the Jose Abreu rumors), it will further prove right the folks who continually point out the Rockies failure to accurately assess themselves.