How the Twins are proving the value of defense

ORGANIZED PROFESSIONAL BASEBALL is 148 years old, but you can still catch something new every day. Or, rather, Byron Buxton can. The 23-year-old center fielder made a diving backhanded grab to steal a hit from Alex Gordon in the third inning on Opening Day, and he's had a glove affair going with Twins fans ever since, even though he's hitting just .200. Indeed, Buxton is on pace to generate better than two wins above replacement in 2017, more than any other player who hit so poorly for a full season.

And in that stat line, you can see the latest strategy for rebuilding MLB teams. I can pinpoint for you the day defense first emerged as the next Moneyball. On July 31, 2004 -- three months before Theo Epstein's first title -- the Red Sox traded Nomar Garciaparra for two plus fielders (Orlando Cabrera and Doug Mientkiewicz), signaling a new direction in sabermetrics. Analysts began to quantify pitch-framing around 2008, the shift took off in 2012 and superior defense keyed the champion 2015 Royals and 2016 Cubs.

Even so, what the Twins are pulling off is special. Under new execs Derek Fulvey and Thad Levine, they have leaped from last to first by focusing heavily on fielding. The Twins replaced Kurt Suzuki, an awful pitch framer who threw out just 19 percent of runners attempting to steal last season, with Jason Castro, who's above average at getting borderline calls while nailing 43 percent of would-be thieves. As a left fielder, Robbie Grossman had the range of Minneapolis' Mary Tyler Moore statue, so the Twins made him mostly a DH. They shifted Miguel Sanó to third, which hasn't cost them in the field and certainly doesn't seem to have hurt the slugger's overall game. And they committed to Jorge Polanco at short, a big improvement over Eduardo Escobar, and to Buxton in center. The results: The Twins have rocketed from third worst in defensive runs saved last year (minus-58) to best this year (plus-31). They're allowing nearly half a run per game less than they did in 2016, even though their FIP, which measures pitching independent of fielding, is nearly half a run per game worse.

I see three reasons they could be a model for other clubs. For one, defense is cheap. The 10 players with the most fielding runs over the past three seasons generated 90.6 wins above replacement in that span, according to FanGraphs. They are earning a total of $70.4 million this year, or about $776,500 per WAR since 2015. The 10 best hitters have 143.8 WAR and are costing their teams $153.1 million in 2017, or 37 percent more per win. Those numbers are skewed, because some of MLB's best defenders haven't yet reached free agency. But Rob Arthur of FiveThirtyEight recently studied all open-market contracts for non-catchers since 2006 and found something even more drastic: Front offices paid more than twice as much for offense as for defense. His results were so stark that Arthur wondered whether the "metrics might just be wrong." But the publicly available stats seem reasonable, rating players like Kevin Kiermaier and Brandon Crawford as defensive wizards. I think it's more probable that most teams just don't yet pay for runs prevented the way they'll pay Nelson Cruz or Edwin Encarnacion for their runs created.

They may soon, though, because, among other obvious benefits, better D can give a boost to pitchers who aren't strikeout artists. Last year the Twins larded the field with subpar gloves while their staff had the highest contact rate in baseball. This season it's even higher (82 percent) -- and yet with Minnesota's defensive turnaround, Ervin Santana (2.20 ERA, 4.45 FIP) looks like an ace. If you allow a lot of balls in play, you can get better quickly by turning more of those balls into outs. Phillies and Braves, take note!

Then there's this: Defense buys you consistency. In 2009, analyst Sky Andrechek found that the standard error for MLB fielders -- the amount something could fluctuate just by chance -- is less than for hitters, and equal to 5.1 outs or about four runs per season. Fielding skills (and speed) seem to be subject to luck but not as much variability as hitting. Put it this way: I love Aaron Judge, but I have a lot more faith that Buxton's range will hold up than will Judge's .408 BABIP or his incredible percentage of fly balls leaving the yard (39.1).

The Twins were skeptical about, even hostile toward, analytics for years. It's pretty amazing how many lessons they are suddenly teaching the world now that they've become the right kind of defensive.