The Numbers Behind the Colorado Rockies' Surprising Start

On Wednesday night, the Los Angeles Dodgers moved into sole possession of first place in the National League West.

The juggernaut Dodgers being on top of the division is hardly a surprise. What is surprising is the team they needed to jump was the Colorado Rockies, who still have the second-best record in the Senior Circuit.

The Rockies were a trendy wild card sleeper pick before the season, but it’s hard to imagine that anyone honestly thought they would be 47-27 in late June. If that wasn’t unexpected enough, it's Colorado’s pitching which has been the catalyst for its success.

They are eighth in baseball in runs allowed per game (4.42), and that does not even account for the Coors Field effect. The club’s park-adjusted ERA is third-best in baseball (behind fellow NL West surprise Arizona and the Dodgers), and its park-adjusted FIP is seventh. The Rockies’ 83 ERA-, strikeout rate (20.9%) and strikeout-minus-walk rate (12.1%) are all on pace to be the best marks in franchise history.

They presumably will not be able to sustain this pace, but thanks to the wins they already have in the bank, our projections give them an 85.0% chance to make the playoffs.

The Story So Far

While the Rockies probably aren't actually the second-best team in the National League -- our models have them as the ninth-best National League squad -- their underlying statistics do imply that they've played like a solid squad and one which should contend for a playoff spot.

Colorado has outperformed its 43-31 Pythagorean Record by excelling in one-run games, and its Pythagorean Record itself is inflated by sequencing (since Colorado’s hitters and pitchers have both done better with runners on-base than with the bases empty).

Still, the Rockies' BaseRuns record -- a “context-neutral run estimator” that strips away sequencing -- is 40-34, the eighth-best mark in baseball.

Pitching has carried the team as the Rockies' pitching ranks second in cumulative rWAR and seventh in fWAR. Its position players are 18th in WAR at Baseball-Reference and 20th at FanGraphs.

It certainly appears that Colorado’s pitchers have figured out the key to pitching at Coors Field. The Rockies have coupled their franchise-best strikeout rate (which, granted is just tied for 16th in MLB) with the second-highest groundball rate in club history (48.8%).

This has obviously helped the Rockies pitchers keep balls in the yard as they are tied for the ninth-lowest home run rate allowed (1.18 per nine innings), despite their massive park disadvantage.

They have also turned 77.6% of their grounders into outs, which is the fourth-highest rate in baseball. Some of this is presumably randomness, but soft contact (Rockies pitchers have allowed the second lowest Statcast-based expected BABIP on grounders) and solid infield defense seems to have also played a role.

Kyle Freeland (66 ERA-, 1.1 fWAR), Antonio Senzatela (80 ERA-, 1.0 fWAR) and Tyler Chatwood (79 ERA-, 0.9 fWAR) all have groundball rates of 48.0% or higher.

Jeff Hoffman has bucked this trend, sporting a groundball rate of just 38.0%, but he has managed to be effective regardless (last night’s disaster notwithstanding), putting up a 83 ERA-, 2.85 FIP (61 FIP-) and 3.87 xFIP (89 xFIP-) over 35.2 innings. He’s been worth 1.2 fWAR, despite a preseason projection of 0.1 WAR with a 4.38 FIP over 29 innings from FanGraphs.

Closer Greg Holland (1.0 fWAR, 32 ERA-) heads a bullpen that leads baseball in Win Probability Added.

They’ve needed the pitching to be on point as only four teams own a lower wRC+ than Colorado's (85). Charlie Blackmon (132 wRC+, 2.5 fWAR), Mark Reynolds (128 wRC+, 1.0 fWAR) and Nolan Arenado (120 wRC+) have been the team’s only above-average regulars.

Can it Continue?

Like Hoffman, the Rockies' staff as a whole has beaten its projections; updated projections at FanGraphs forecast Colorado to post a 4.39 FIP going forward while tying for 15th in pitcher WAR (Colorado currently has a 4.32 FIP and is 7th in the game in fWAR).

The projections have not budged all that much since the preseason -- FanGraphs had Colorado’s pitchers pegged to produce 14.6 WAR before the season, and if we prorate the current projections (8.1 WAR over 792 innings), we get 14.9 WAR. That pales behind their current level of production, which is currently at a rate of 19.6 WAR per 1,458 innings.

The latter figure is impressive in light of the preseason projections and the fact that virtually all of that production has come in Jon Gray’s absence. Gray has a career FIP- of 83 and xFIP- of 90, so his return should be a welcome one for a staff whose performance could regress. The Denver Post reports he will make his third rehab start on Saturday and could be up after that.

As for the rest of the Rockies' staff, it’s not inconceivable that they continue to pitch at a level close to how they’ve done so far, but is also easy to see it reverting to the merely average level the projections call for.

Freeland, for example, is projected to pitch to a 5.01 ERA going forward as he has a paltry 5.1% strikeout-minus-walk rate; his current ERA is propped up by a 79.4% strand rate and a home-run-per-fly-ball rate that is a percentage point below the league average.

Hoffman has a a 25.7% strikeout rate and 5.4% walk rate, making it tempting to be bullish on his projections, which forecast a 4.44 ERA going forward. Keep in mind, though, that, he had a higher walk rate in four of his five minor league stops, while his strikeout rate is higher than anything he posted in the minors.

Marquez’s is projected to post a 4.71 ERA, which is in line with his 4.60 xFIP. He has posted a 3.88 FIP so far, but that is mostly due to a 9.0% home-run-per-fly-ball rate (which is nearly 5% lower than the league average).

Offensively, the Rockies have actually slightly undershot their projections. They are currently posting a .332 wOBA but are projected to post a .341 wOBA going forward. Blackmon and Reynolds will presumably regress somewhat, but the team will count on improvements from Trevor Story and Carlos Gonzalez to offset this regression.

Story is expected to be close to average going forward, with a 95 wRC+ and 2.2 WAR/600 plate appearances projection from FanGraphs. This would be an improvement from his poor start, which has seen him post just a 77 wRC+, though his .291 Statcast-based Expected wOBA (xwOBA) is certainly concerning.

Gonzalez has a .224/.303/.353 (53 wRC+, .282 wOBA) slash line and has been worth -1.1 fWAR, making him the least valuable position player in the majors. Like Story, we can't really chalk this up to bad luck (despite a .262 BABIP) as his xwOBA is just .295. Still, the FanGraphs Depth Charts projections expect him to post a .347 wOBA and be worth 0.6 WAR over his next 342 plate appearances.

It still may not be enough as the Rockies' position players are only projected to rank 22nd in WAR going forward. Our numberFire projections have the Rockies going 43-45 the rest of the way, while FanGraphs has them going 44-44.

Their “true performance,” as measured by things like BaseRuns and Baseball Prospectus’ third-order winning percentage, has certainly been good, but it may also be outpacing their “true talent.”

Even if they do only play .500 baseball the rest of the way, that would give them 91 wins, meaning playoff baseball will almost certainly be played at Coors Field for the first time since 2009.