Houston’s Offense Has Hit a Bump in the Road
The St. Louis Cardinals are having a terrible postseason at the plate. They were no-hit for seven and two-third innings by Aníbal Sánchez for goodness sake, and while their subsequent struggles against Max Scherzer and Stephen Strasburg made more sense, it’s fair to say the team has a problem. But St. Louis had the worst non-pitcher wRC+ among teams to qualify for the postseason, so it’s hardly a surprise to see its roster scuffle against playoff rotations.
The Cardinals’ recent three game stretch of futility leaves their team postseason slash line at .207/.264/.331, a truly grim line. No one is questioning their credentials as the worst postseason offense. The next team on the list, though, might surprise you: it’s the Houston Astros, who are hitting a collective .218/.281/.367 through seven games. A dramatic Carlos Correa home run evened their series with the Yankees 1-1, but they’ve still only produced three runs over two games of the ALCS.
Houston’s offensive ineptitude hasn’t yet caught up to them, but it’s still concerning. Heck, the team literally cracked the code in their shelling of Tyler Glasnow in the ALDS, and their offense has struggled mightily even after accounting for that. There’s some chance, however remote, that the Astros’ offense is doomed, that everyone turned into a pumpkin at once. There’s a higher chance that we should just completely ignore this result; during the regular season, the offense produced a 126 wRC+, easily the best in baseball. But rather than take either extreme course, let’s take a closer look at what has happened and see if we can find any takeaways.
One thing you’ll hear ad nauseam at FanGraphs is that context is important when it comes to looking at a team’s season-long stats. Postseason rosters can be constructed quite differently than a team’s regular-season squad, and looking at 162 games of fill-ins and getaway day lineups can obscure a team’s true talent level.
Of course, that doesn’t mean the Astros offense should be worse; in fact, all else being equal, the team’s postseason construction flatters them. Take a look at the percentage of the team’s plate appearances taken by each starter in the regular season and then the postseason:
| Player | Postseason PA% | Reg Season PA% |
|---|---|---|
| George Springer | 12.4% | 8.7% |
| Michael Brantley | 11.6% | 10.0% |
| Carlos Correa | 10.8% | 5.0% |
| José Altuve | 11.6% | 8.6% |
| Yuli Gurriel | 11.2% | 9.6% |
| Yordan Alvarez | 11.2% | 5.8% |
| Alex Bregman | 11.2% | 10.8% |
| Robinson Chirinos | 7.2% | 6.9% |
| Everyone Else | 12.4% | 34.6% |
The Astros have their A-squad on the field pretty much all the time, something that, all else equal, should improve their offense. If you kept everyone’s regular-season batting lines but apportioned playing time in the same ratio Houston has this postseason, the team would have a 142 wRC+ and .379 wOBA, which is patently insane. So playing time isn’t the issue.
Nor is it an issue of the Astros playing over their heads in the regular season before regressing in the playoffs. Take their postseason playing time allocation, but assign each batter their Depth Charts projections rather than their performance in 2019, and the team would project for a .350 wOBA and 122 wRC+ in the playoffs against neutral pitching.
Of course, “against neutral pitching” is a poor assumption when it comes to October baseball. The Astros have faced two teams with lights out bullpens and star-studded rotations, and they haven’t had the luxury of feasting on fifth starters or shuttle squad relievers. Take a look at the pitchers the Astros have faced so far:
| Pitcher | Batters Faced | Reg Season ERA | Reg Season wOBA | Proj ERA |
|---|---|---|---|---|
| Tyler Glasnow | 33 | 1.78 | .222 | 3.62 |
| Charlie Morton | 21 | 3.05 | .270 | 3.25 |
| Blake Snell | 20 | 4.29 | .301 | 3.24 |
| Masahiro Tanaka | 18 | 4.45 | .314 | 4.25 |
| Nick Anderson | 16 | 3.32 | .269 | 3.14 |
| Diego Castillo | 15 | 3.41 | .292 | 3.34 |
| Colin Poche | 15 | 4.70 | .278 | 3.45 |
| Emilio Pagán | 15 | 2.31 | .244 | 3.67 |
| Oliver Drake | 13 | 3.21 | .261 | 3.55 |
| James Paxton | 12 | 3.82 | .310 | 3.77 |
| Chaz Roe | 12 | 4.06 | .310 | 3.72 |
| Ryan Yarbrough | 11 | 4.13 | .273 | 4.24 |
| Adam Ottavino | 8 | 1.90 | .276 | 3.81 |
| Zack Britton | 8 | 1.91 | .249 | 3.39 |
| Tommy Kahnle | 7 | 3.67 | .272 | 3.55 |
| Chad Green | 6 | 4.17 | .311 | 3.06 |
| Brendan McKay | 6 | 5.14 | .333 | 3.91 |
| Jonathan Loaisiga | 5 | 4.55 | .342 | 4.09 |
| Aroldis Chapman | 4 | 2.21 | .242 | 2.67 |
| J.A. Happ | 3 | 4.91 | .329 | 4.56 |
| CC Sabathia | 1 | 4.95 | .351 | 4.40 |
That works out to a composite weighted ERA of 3.39 and a wOBA allowed of .277. Their projections would work out to a 3.58 ERA, nearly as good. That’s something like facing Tommy Kahnle every single time you come to bat; it’s a tough environment to hit in.
But even acknowledging the opposition, the Astros hitters have underperformed. When a hitter the quality of the aggregate Astro lineup faces off against a pitcher the quality of their aggregate opponent, odds ratio gives the expected outcome as a .313 wOBA. Instead, they’ve produced a .278 wOBA, which is as bad as it sounds; that was Mallex Smith’s wOBA this year, and he hit .227/.300/.335 in the year of the home run.
As much as hitting like Mallex Smith is terrible, though, it’s not all that unlikely over a small sample. By back-of-the-envelope math, in fact, that outcome is only 1.1 standard deviations below the median outcome, using the rough rule that standard deviation of wOBA works out to .5/sqrt(plate appearances). In other words, the Astros should expect to put up a performance as bad as they have or worse roughly 13.5% of the time.
While we have this metric handy, we can work out how well or poorly each team has hit relative to expectations, given their hitters and the opposition they’ve faced:
| Team | wOBA | ewOBA | Difference | St Dev |
|---|---|---|---|---|
| Cardinals | .255 | .295 | -.040 | -1.4 |
| Astros | .278 | .313 | -.035 | -1.1 |
| Braves | .295 | .309 | -.014 | -0.4 |
| Twins | .299 | .309 | -.010 | -0.2 |
| Nationals | .299 | .301 | -.002 | -0.1 |
| Dodgers | .309 | .306 | .003 | 0.1 |
| Rays | .308 | .283 | .025 | 0.7 |
| Yankees | .354 | .323 | .031 | 0.9 |
These are just approximations: they ignore handedness matchups and home/road adjustments, as well as short rest modifiers for pitchers and pinch-hit penalties for batters. Even so, they paint a good outline of what has happened this postseason. The Rays and Yankees have hit better than expected, while the Astros and Cardinals have disappointed.
Most of that is obvious in the raw batting stats; of course the Yankees have done well after blitzing the Twins, and of course the Cardinals and Astros haven’t looked good. There are lessons to be learned, though. The Rays, for example, had a worse batting line than their overall season statistics, and they overperformed expectations to get there. The Astros pitching staff really is that good. The Twins didn’t hit well, but it wasn’t all down to their own players; the pitchers they faced were excellent, so much so that their projected batting line was no different than that of the Braves despite a far better regular season performance.
When it comes to the Astros offense, there’s only one conclusion to draw: they’ve been disappointing this postseason. Analyzing individual hitters is nearly pointless due to sample size issues, but only Altuve and Bregman have excelled so far, with the rest of the team struggling. Their road won’t get easier from here; the Yankees are going to keep throwing dominant relievers and short-stint starters at the wall to see what sticks, and the two National League teams both boast excellent pitching.
The Astros have a trump card: their own pitching is so good that they can overcome dry spells from the offense. But they could make the playoffs a lot easier for themselves by playing up to their projected level on offense. Even accounting for the imposing opposition they’ve faced, the Astros simply haven’t been up to snuff.
Ben is a writer at FanGraphs. He can be found on Bluesky @benclemens.
They actually did okay against Tampa. Not great but okay: .242/.294/.406
But in two games against the Yankees they’ve hit: .156/.250/.266
Which is pretty similar to what they hit against the Yankees in the 2017 ALCS: .187/.271/.294
This is what happens when hitters build their approach at the plate around juiced balls.
I’m open to that theory, but wouldn’t it also mean every team hits worse and not just the Astros?
Not necessarily. It depends on the specific spray chart. Aaron Judge probably doesn’t that much more help from the juiced balls. But Bregman hit 41 homers this year, and Springer hit 39, Altuve and Gurriel hit 31, and all those guys probably got the benefit of a bunch of balls that would have died on the warning track going for home runs, more so than other teams.
Good theory, it’d be interested to see it tested.
How would these theories be tested? The sample size isn’t small enough to get anything out of the postseason. It is absolutely true that players with big power don’t need juiced balls and the typical warning track power guy gets a major benefit. The proof is that you don’t see the 80 power guys hitting more HR. You see several guys with garden variety power battling hitting a bunch of HR. That’s anecdotal but I am sure it is true. I really do think that most studies are flawed enough that they are no better than anecdotal.
On reflection, though, it’s an alarming thought. Why would mlb have decided to deaden the balls for the postseason? If it helps some teams more than others, one might think…
If they *did* change the ball, my guess is that it would probably be because the MLB is worried that the average fan won’t like the product from the 2019 Regular Season. It would be for ratings, not giving a particular team an advantage.
I don’t know why they would do it. Can the same organization that coulnd’t determine that something had changed really manage baseballs? They have proved incompetent so far. It seems like a bad idea and I don’t know why they would de-juice for the playoffs. If anything I would think they would super-juice the balls for the playoffs.
I think it is just the opposite. The person who doesn’t follow baseball closely but watches the playoffs is not going to enjoy most of the games featuring very little offense and in today’s game offense is home runs.
When was the last time MLB did something smart? They denied this whole thing the whole time… nailed their own coffin shut by introducing juiced balls to AAA… who knows what they would do next? My best guess would be whatever the worst choice is.
It would mean that the team benefiting the most from juiced baseballs gets hurt the most. I think it might be hard to argue that HOU is that team – I am not sure what it would be based on. I may have heard a stat that HOU hit 100 more HR this year than last at home… something like that… but that isn’t conclusive proof of anything.
Is there a way to find out what percentage of the Astros’ home runs at Minute Maid Park would have been home runs at other parks?
This is true as well. If they ever de-juice the balls you will see that LA is not the evolution of hitting but the adaptation to juiced balls. This will be hailed as the evolution of pitching and defense.
Except LA still helps with ball de-juiced. Maybe the optimal LA drops 2-4 degrees, but hitting grounders is a recipe for disaster.
Astros have the second largest xwOBA-wOBA this offseason which takes a slump and magnifies it. Pitchers have been controlling the zone against them (wOBA on Ks and BBs basically half what Astros did in the regular season). Batters getting down 0-1 a lot more in playoffs. Some great defensive plays by outfielders have helped keep Astros from taking an extra 90 feet (off the wall singles, runners getting held up at 3rd on a double or thrown out at home).
I expect Astros will get luckier on contact (xwOBA-wOBA) going forward, but that shouldn’t affect that pitchers are being stingy with balls, and defense of opposing teams.
Yeah, just eye-balling the gamefeeds on Baseball Savant: ALCS Game 1 xBA: 0.244, actual BA: 0.111; ALCS Game 2 xBA: 0.309, actual BA: 0.189.
So across those 2 games, they have a xBA of 0.281 and an actual of 0.156. It would be difficult for me to accept that this difference of results on contact is entirely “real”.
Yep… most of this can be explained well hit baseball finding gloves
At the time the author posted this article
Odds ratio calculated ewOBA – 0.313
Contact quality calculated xwOBA – 0.315
If Houston’s offense has hit a bump in the road, what has St Louis’s offense hit?
A tree on the side of the road.
I think the Cards offense has gone more full on Wile E. Coyote against a mountain.
And the Nationals pitching staff painted on it to look like a road. True Rembrandts.
All this analysis is great, but it doesn’t help much if you can’t watch the game. 3:08 PM Central time start?!? One of the most anticipated Championship Series match ups in some time and it’ll be over by the time I get home from work. WTF MLB? or is it GEICO? or FOX/TBS?
It’s because there are two games and MLB doesn’t want them going head to head.
If you look tomorrow, it is the reverse (assuming there are games tomorrow).
As someone who watched the Astros all year, its actually very simple and doesn’t require heavy calculator use. The Astros as a team would slump for games at a time, multiple times, throughout the season. The thing that usually got them back on track was facing a terrible Rangers , Mariners, or similar bad team pitcher. The plate approach and swings in the postseason haven’t been great either. Springer, Altuve, Bregman, and Alvarez have been swinging out of their shoes trying to launch the long ball. Simple level swings would have likely produced a much higher percentage of hits and in the gap doubles. In Game 5 of the ALDS, they knew what pitch was coming and you could notice an entirely different level of confidence and control in the swings. Other than that inning, its been all about trying to mash. The pitchers know it and take advantage.
If the Astros get back under control instead of trying to mash, they will score 5+ per game. If they don’t adjust, we will see the same thing and they will likely be bounced.
Yeah they did have the advantage of 3 of the 6 worst pitching staffs by RA/9 in their division. So 57 games with them.
That sounds like the perspective of a person that watches baseball regularly. It takes a distant perspective to not understand how streaky every team is, which is what you would have if you were only looking at averages. Is the view that many people who talk baseball on the Internet have. I think HOU has hit a lot of atom balls in this series – they haven’t been as bad as I think you think they have been. Yeah, that sentence could use some work!
Has any analysis been done of pitch susceptibility? To the eye looks as if the Astros, and Cardinals for that matter, are helpless against the changeup.