Twenty one games into the 2015 season, the Houston Astros were 14-7. It was a delightful story because at the time even the Astros seemed to think they were a year or two away from real contention, and there they were seven games above .500 before the end of the first month of the season. At the time, I noted that their early season success put them in a situation where they were legitimately within a few moves of the postseason. After those strong opening weeks, their full season projection was around 83 wins, and getting to 90 was now doable with a few aggressive alterations to their roster.
As part of the exercise, I forbade myself from forecasting any breakouts. It would have been boring if I had just predicted MVP seasons from George Springer and Colby Rasmus in order to pad the win total. Instead, I opted for various call ups and trades that they could reasonably make in order to become real contenders. This included calling up Carlos Correa, trading for Andre Ethier, Gerardo Parra, Cole Hamels, Matt Garza, and Aroldis Chapman (or similar players).
A lot of people said it was an unrealistic move for the team, and fans who read the article said they preferred to ride out the good luck with an eye on the future which made total sense given the state of things in April. As I’m sure you know, however, the Astros kept playing well. So well, that over the last two weeks they traded for Scott Kazmir, Mike Fiers, and Carlos Gomez, and there were some rumors that they were after one of the un-traded relief aces until the deadline, as well.
As much as I would love to take credit for devising the Astros’ path to October, the club made those big trades because they kept playing well until late July and found themselves in the hunt. My plan was designed to get them to contention, but they pulled the trigger because a playoff berth was actually in sight and they are trying to close the deal.
If you go back to Opening Day, the FanGraphs projections, using ZiPS, Steamer, and manually coded playing time, believed the Astros would win 48.6% of their games. Obviously, projections are a messy business. It was possible we made mistakes in predicting who would get playing time and it’s also very likely that individual players would over- or under-perform their preseason projections. It happens every year.
But when the season began, our model had the Astros at a .486 winning percentage for the entire year. Today, it’s .537. Over 162 games, that’s an eight win difference. Keep in mind, I’m not talking about the Astros projected end-of-season forecast with their current .561 built in. This is no longer about the wins they have in the bank. On Opening Day, our model saw them as a .486 team. From today until the end of the season, our model now thinks they are a .537 team. Needless to say, that’s the biggest improvement by a very wide margin.
Part of that growth comes from the players they acquired at the deadline. In fact, on July 22 (the day before they got Kazmir), their rest-of-season winning percentage sat at .505. That’s still a hearty .019 increase, accounting for three wins over a fun season, but it’s certainly not the entire picture. This improvement is 60/40 in favor of trade deadline acquisitions to internal improvements. They filled holes with good players from the outside, but the guys in the organization when the season began have also helped. How did the Astros get better?
It starts with Correa. Before the season began, ZiPS projected a .300 wOBA for the young shortstop. While everyone is well aware of Correa’s terrific performance to date (.390 wOBA, 2.7 WAR in 209 PA), his projection has already spiked to .320 wOBA, helping his team’s rest-of-season forecast. While Correa has played like a superstar in his first two months, no projection system is going to go from .300 to .400 that quickly, so 20 points is a serious change.
The fact that Correa has handled major league pitching with such ease in his early days, however, has clearly changed things for the Astros above and beyond the value he’s already banked. Part of my plan to get them to 90 wins involved calling upon Correa, but I was utilizing the preseason projection. He’s blown that out of the water and you can probably expect it to keep climbing as he keeps hitting. It’s not just that Correa has helped them win so far, it’s that we were wrong about how much he was like to help when the season began.
The story is similar for Preston Tucker, who came into the season with a projection from ZiPS and Steamer around .300 wOBA, which has increased about 10 points thanks to his .340 wOBA in his first 249 MLB plate appearances. Back in April, he was barely on the radar as a potential difference maker for the team. Not only has his projection improved, but he has a larger role on the team. His impact doesn’t match Correa’s, but it’s meaningful nonetheless.
Dallas Keuchel came into the season with a projected FIP of about 3.65. That has since fallen to to 3.25 as the projection systems start to accept that a guy who typically sits under 90 can still generate a fair number of strikeouts. ZiPS and Steamer also didn’t expect much from Lance McCullers (and he didn’t factor heavily into the preseason forecast), offering a projected FIP above 5.00. The combined projection is now down to 4.29 and falling thanks to better than anticipated strikeout and walk rates so far, his rough start on Monday notwithstanding. He’s about to spend a little time in the minors do to workload concerns, so this could move the needle a bit depending on how they respond.
Retrospective analysis is relatively easy. Using Base Runs, the Astros have been the second best team in baseball so far this season, trailing only the Dodgers. The gap between the Astros and the Blue Jays, who are third, is massive. If you don’t like Base Runs, they’re third by Pythagorean, and sixth using good old fashioned winning percentage. They have played great baseball. That’s interesting and exciting, but perhaps the more compelling story here is that this is no longer an okay team playing above their heads. They have morphed into a legitimately good baseball team that should pose a real threat to the rest of the American League down the stretch.
Not only have the Astros added Kazmir, Fiers, and Gomez, but Correa, Tucker, and McCullers appear to be ahead of schedule developmentally, while Dallas Keuchel has truly become a great pitcher.
When the season started, we smiled because the Astros won more games than we expected them to win in April, but we’ve now reached the point where the story is very different. The Astros are a better team than we all thought they would be, in part because we undersold the abilities of some of their young players and in part because they added a few very good players for their stretch run. If you wanted to make the case that the Astros are the best team in the American League, you would have a decent set of facts to support that claim.
Should we have predicted this? I don’t think so, but I don’t mean that in a “we should never deviate from the projections sort of way.” None of the individual events we’ve discussed are unpredictable at all. The unlikely thing is the immediate rise of Correa, and to a lesser extent Tucker, on offense, coupled with a truly great Keuchel and a ready for prime time McCullers occurring all at once.
It’s one thing to project a big year from Keuchel or a breakout from Correa, but this is really the classic case of everything sort of breaking right for an okay team. We say it every year, but most teams are a few good breaks away from contention in the current MLB climate of parity, and the Astros combined a little of that with some good luck, putting themselves in position to be serious buyers in July.
The Astros weren’t supposed to be this good this soon, but once the hands of the clock started spinning more rapidly, the window opened and the team decided to climb through. None of us predicted the Astros would be a contender when the season began, but as it stands today, they have the third highest World Series odds in all of baseball. Along the way, they’ve played better than expected, but they’ve also legitimately become a better team.
Neil Weinberg is the Site Educator at FanGraphs and can be found writing enthusiastically about the Detroit Tigers at New English D. Follow and interact with him on Twitter @NeilWeinberg44.