The Rangers outscored their opponents by eight runs total all season, which is the type of performance we would expect from an 82-win team rather than a division leader. (AP Photo/Ross D. Franklin)

Texas won 95 games in 2016, earning it back-to-back AL West titles, but it appears the Rangers’ luck may run out this season.

Last season, Texas had the third-worst bullpen in the American League per wins above replacement (1.7 fWAR) and the second-worst per ERA (4.40), a below-average starting rotation (7.9 fWAR, fifth-worst in AL) and an average offense which created runs at a rate that was 2 percent below league average after adjusting for league and park effects (98 wRC+). As a result, it outscored its opponents by a mere eight runs total all season, which is the type of performance we would expect from an 82-win team rather than a division leader. Specifically, the Rangers benefited from what Joe Peta, a former Wall Street trader, termed “cluster luck” in his book, Trading Bases:

When a team’s batters cluster hits together to score more runs and a team’s pitchers spread hits apart to allow fewer runs, that’s cluster luck. Say a team tallies nine singles in one game. If all of those singles occur in the same inning, the team would likely score seven runs; if each single occurs in a different inning, however, it’d likely mean a shutout.

Here’s why the sequence is significant. In a July game against the Houston Astros, the Rangers managed just two runs on 15 hits. In that contest, Texas had just one inning with three consecutive hits without an out. Almost a month later against the Oakland Athletics, the Rangers again had 15 hits, although this time they managed to score seven runs, thanks in large part to two innings with at least three consecutive hits without an out. Same number of hits, but two very different outcomes.

FanGraphs runs this “cluster luck” into its BaseRuns formula, which calculates a total expected number of runs scored (or allowed) based primarily on hits, walks, hit by pitch and home runs, which are then adjusted for the league a team plays in. BaseRuns are preferable to just wins and losses because not all wins are created equal. For example, a team that beats its opponent by a score of 9-1 is much more dominating than a squad that squeaks by in a one-run game. Texas was the best team in the majors last season at winning one-run games, establishing a 36-11 record in these contests, six more wins than the next best team (Seattle Mariners). Based on run differential in one-run games, the Rangers should have won only 26 or 27.

The Rangers’ 13-game win differential over their expected record for the entire season was by far the largest in baseball last year. It was 14 games better than the Seattle Mariners, who finished second in the AL West by nine games, and 12 games better than the third-place Houston Astros, who were 11 games back. The Philadelphia Phillies were the next luckiest team in the majors behind Texas, winning nine games more than expected.

Since 2006, the first year MLB adopted leaguewide drug testing, there have been four teams that won 10 or more games more than expected based on their run differential: 2007 Arizona Diamondbacks (11), 2008 Los Angeles Angels of Anaheim (12), 2009 Seattle Mariners (10) and the 2012 Baltimore Orioles (11). All but Baltimore won fewer games the following season — it improved by four wins — with the Mariners going from 85 wins to 61 wins in the year following their fortunate season.

To be fair, Texas’s run differential took a major hit in July (it was outscored by 46 total runs for the month) and improved in August (plus-13) when it got Yu Darvish back in the rotation, traded for Jonathan Lucroy and added Carlos Gomez after he was released by the Astros. But they would get lucky again in September, winning 15 of 26 games despite being outscored, 134 to 125, for the month.

Plus, the Rangers’ offense is projected to lose some value to offseason moves. Ian Desmond, Mitch Moreland and Carlos Beltran all left for greener pastures and the replacements the team brought in, Mike Napoli, James Loney and Josh Hamilton, aren’t projected to contribute nearly as much in wins above replacement. In fact, FanGraphs’ depth chart projections, a combination of ZiPS and Steamer projections with playing time allocated by their staff, sees the net loss somewhere between one and two wins for 2017.

The pitching staff, despite the additions of Andrew Cashner, Tyson Ross and Dillon Gee making up for the loss of Derek Holland and Colby Lewis, is expected to be average — landing eighth in the AL per projected wins above replacement (15.9 fWAR).

This has the Rangers estimated to win 83 games in 2017, seven fewer than the Astros (90, most in AL West) and just four ahead of the Oakland Athletics who are expected to occupy the division’s basement. That’s a slim margin for Texas, who could easily drop from first to worst in one season.