Skip to main content
Menu
The Astros Are Making A Historic Turnaround

The Houston Astros came out of nowhere. Just two seasons after occupying the basement of the American League West, the Astros have returned to take the top spot. Their turnaround is so sizable and so swift that it’s historic.

In 2014, the Astros finished fourth in the division, with a correspondingly putrid winning percentage of .432. This year, the Astros are projected1 to end with a .530 winning percentage, sixth-best in the majors. Turnarounds like that don’t happen often, but they do happen: Since 1950, only 114 teams (7.2 percent of the total number) have managed to increase their winning percentage by .100 or more over the course of a single season.2 The 2013 World Series-winning Boston Red Sox managed it, as did the 2014 Los Angeles Angels.3 We’d expect to see a couple of teams every year bump their winning percentage by a similar margin.

The Astros’ turnaround becomes historic, however, when you look at how bad they were two years ago. In 2013, the Astros finished 51-111, good for a .315 winning percentage and the bottom of the division. The team was not felled by injuries or misfortune — it was genuinely terrible in every phase of the game. The Astros’ hitters racked up 1.4 wins above replacement (WAR) — 29th in the league — and their pitchers totaled 1.2 — 30th in the league. That’s what happens when your team is essentially replacement level. A team full of anonymous, AAA types who couldn’t make it in the major leagues would have been projected to finish at a winning percentage of .294, barely worse than the Astros’ actual performance.

If the projection for this season holds, the Astros will have increased their winning percentage by .200 over two years. Since 1950, that kind of reversal has happened a grand total of seven times — and when it has, it’s usually because a decent team has gotten radically better. The 2001 Seattle Mariners assembled a historic juggernaut of a team that won 71.6 percent of its games after winning 48.8 percent two years earlier. The team before the Mariners to accomplish this feat, the 1995 Cleveland Indians, became a 100-win team from a borderline contender. Only one team — the 1963 Philadelphia Phillies — had a starting winning percentage as poor as the Astros did at the end of their 2013 season.

It’s still early in the season, and the Astros likely won’t finish as well as they have started. That doesn’t mean, though, that the Astros’ current winning percentage (.625) is founded purely on luck. They have one of the best run differentials in the league, scoring 28 more runs than they’ve allowed.4 The Astros aren’t getting terribly lucky in terms of their batting average on balls in play (BABIP), either offensively (.278, good for 24th in the league) or defensively (.283, also 24th). They have been a little bit lucky in terms of clustering their hits, but even if we remove that, they’d have earned a .576 winning percentage so far, good for fourth in the league.

Most teams that accomplished turnarounds like the Astros’ did so on the basis of vastly improved play — both on offense and defense. On average, teams that saw their winning percentage improve by .100 year to year were helped by their offenses putting up .46 more runs per game and their defense allowing .54 fewer runs per game. The Astros are doing just as well. Their runs per game have improved by .55 relative to last year, and their runs allowed per game have declined by .61. In other words, this kind of improvement is no fluke.

The Astros have built their team on a combination of savvy trades (outfielder Jake Marisnick), high draft picks (right fielder George Springer) and an eye for talent disregarded by other teams (second baseman Jose Altuve and starting pitcher Collin McHugh). Guys like McHugh offer an insight into the front office’s analytics-heavy approach. McHugh was acquired not because of his results but because the spin on his curveball suggested that he could become a success.

As with any turnaround, however, luck does play a role. Indeed, projection systems are relatively unchanged in their opinion of the Astros despite their success. FanGraphs’ Steamer projection pegs them as a roughly .500 team going forward, and Baseball Prospectus’s PECOTA is even less optimistic.

But all those wins in the bank mean that even if they regress, the Astros stand a good chance at making the playoffs for the first time since 2005. Both projections tab them as better-than-even to get into the postseason, and their unexpectedly exceptional play may convince the front office to make further improvements to the roster.

One of the most optimistic parallels for the Astros’ recent success comes from the last team to increase its winning percentage by .200 in two years: the 2008 Tampa Bay Rays. Like the Astros, those Rays were coming from the basement, riding a wave of young talent gathered by a recently installed, sabermetrically advanced front office. For the Rays, 2008’s turnaround was the beginning of an impressive run that saw them make the playoffs in four of six years (getting all the way to the World Series in 2008).

If those Rays are any guide, we may be witnessing the rise of a new contender — one that will be competitive in the AL West for several years. This may be the last year the Astros sneak up on people.

Footnotes

  1. Using FanGraphs’ projected standings. I am using the projected standings to take into account that the Astros’ performance is likely to regress somewhat.

  2. I am using data from Sean Lahman’s database.

  3. Notably, both of these teams had been quite good two years before, implying that their improved winning percentages were returns to their expected level of play. That statement holds generally: The teams that improved their winning percentage by .100 or more had a .478 winning percentage two years prior, much better than the Astros’ woeful mark (.315).

  4. According to Russell Carleton’s work at Baseball Prospectus, run differential doesn’t stabilize until 70 games have been played. But such a strong differential does portend positive things for the Astros, even if the sample size is not yet large enough to be certain.

Rob Arthur is FiveThirtyEight’s baseball columnist and also writes about crime.

Comments