The Twins and More Hints of a Changing Market

While many of you probably spent the long holiday weekend with the family you already had, the Twins spent theirs building a new one, signing Ricky Nolasco for four years before signing Phil Hughes for three. While this post is being published much later than the reports, many haven’t been looking at their computers, and these transactions are too interesting to outright ignore. Without stepping on Mike Petriello’s toes, these moves are notable for a variety of reasons.

(1) The Twins aren’t good. Obviously, the Twins aren’t good, and while the Twins might become good in the near or less-near future, they’re not good now, and their division already has talented teams, and free-agent acquisitions are usually short-term improvements. You don’t usually expect lousy ballclubs to make multi-year commitments on the market. Not like this.

(2) The Twins are the Twins. They’ve never really been big spenders, especially when it comes to free-agent pitching, preferring to go cheap and/or build from within. It follows that this could signal either a change in approach, or a particularly strong belief in the two new arms.

(3) The Twins are the Twins. Their bullpen just racked up more strikeouts than their starting rotation. Famously, the Twins haven’t just downplayed the significance of strikeouts for starters — they’ve practically acted as if they’re a bad thing. Of course, there are the Francisco Liriano and Johan Santana exceptions, but most of the internal arms have been contact sorts, and the same goes for acquisitions. Kevin Correia doesn’t strike hitters out. Mike Pelfrey doesn’t strike hitters out. Jason Marquis doesn’t strike hitters out. Vance Worley got some strikeouts with the Phillies, but a bunch of them were called, and then you’ve got guys like Carl Pavano and Livan Hernandez and Ramon Ortiz and Sidney Ponson. Last year both Nolasco and Hughes posted strikeout rates that were above their respective league averages. They don’t quite fit the typical Twins mold.

(4) Nolasco and Hughes have some warts. Previously, one might not have expected them to be able to land seven guaranteed years, combined. They look better by peripheral numbers than they do by ordinary runs-allowed numbers. But now they have those commitments, and what’s more is that they’ve been given those commitments by what’s been considered an extremely old-school organization. Through a certain lens, you could say this is evidence of even the Twins buying into sabermetrics.

When the Giants signed Tim Lincecum earlier in the offseason, Dave saw it as a payment for FIP instead of ERA. Based on runs, it would’ve been impossible to justify the expense, even given what Lincecum has meant to the city and the team. We’re still accustomed to thinking players of the sort might be underrated on the market, but now we could have more evidence to the contrary.

Hughes is coming off a disastrous campaign in New York. Nolasco is coming off a year that saw him get some good exposure in Los Angeles, but for his career he’s been a lot worse than his peripheral numbers would suggest. Used to be that Javier Vazquez was the first guy to bring up when talking about guys who didn’t have matching FIPs and ERAs. Now you have Matt Cain in one direction and Nolasco in the other, as Nolasco owns a career 92 FIP- and a 108 ERA-. Even last season, he had a gap of 11 points. Look at Hughes and it’s easy to see runs allowed. Look at Nolasco and it’s easy to see the same. The Twins prefer to see a couple glasses half full.

An interpretation is that the Twins trust that Nolasco and Hughes will be able to strand more runners. The last three years, 79 starting pitchers have thrown at least 400 innings. Among them, Nolasco is tied for the worst left-on-base rate. Hughes, meanwhile, ranks eighth-worst. There’s a strong and obvious connection between LOB% and runs allowed, but then there’s this: since 2010, the year-to-year correlation in LOB% has been 0.18 for starters with at least 100 innings in consecutive seasons. In other words, it’s a skill, but only to a certain and limited extent, and a lot of it comes down to simple random distribution. The most cliche FanGraphs way I can put this: when it comes to LOB%, going forward you have to regress heavily to the mean. There’s way too much noise, and there are far more meaningful performance statistics.

Think too much about runs allowed, and you’ll see the problems that both Nolasco and Hughes have had. Dig into their issues and you’ll see some reasons for hope. In Nolasco’s case, this has been going on for a very long time, but then he pitched in front of some terrible Marlins defenses, and additionally a career still isn’t all that big of a sample size, to be perfectly honest. Nolasco still has to be regressed, and his FIP is still something of value.

With these two signings, it’s possible to see more evidence of a market paying pitchers for things other than the runs they’ve allowed. And both these signings were made by a team viewed as one of the last traditional-baseball holdouts. As Dave also wrote about with regard to the Jhonny Peralta contract, perhaps we need to re-evaluate what we think the market is going to underrate. It’s not 2003 anymore, and teams have gotten a whole hell of a lot smarter. They can see what we can see, and more than that.

Yet in the interest of considering all possibilities, for the Twins perhaps this isn’t about sabermetrics or sabermetrics-like thinking at all. Perhaps it’s a lot simpler. The last three years, Nolasco’s thrown just about 600 innings, and last season he achieved a sub-4 ERA for the first time since 2008. Maybe they just see him as a veteran workhorse. And with Hughes, it might be as simple as trying to get a talented guy into a better environment. Over his career, Hughes has allowed more than twice as many dingers at home as on the road. Yankee Stadium is a bandbox, while Target Field is more of a small national park. Maybe it’s a ballpark consideration, along with a memory of Hughes’ old prospect hype. If you’re insistent on seeing old-school thinking, old-school thinking can be seen here. This is by no means conclusive evidence that even the most traditional organizations are leaning analytical.

But, perhaps unwittingly or perhaps wittingly, the Twins are taking a chance on xFIP and regression. And they’re confident enough to have guaranteed three and four years to their newest starting pitchers, blemishes and all. No matter how this all works out, it’s plenty more interesting than getting a handful more Kevin Correias. Maybe they’re not the best moves the Twins could’ve made, but at least they could have the makings of a starting rotation people can actually watch.