I've been trying to figure out for a few weeks now The Point At Which The Season Turned Around for the Astros - something we'll come back to another time - but you could make a strong case that it was May 11.
After a 5-4 extra-innings loss to the Orioles on May 10, the Astros were 11-26 and had lost seven of their previous nine games, though they had managed to salvage a series split at Oakland and win two of three at Seattle from April 21-27. There were hints that they *could* play well, they just...uh...weren't (remember to follow Astros County on Twitter for more hard-hitting data-driven analysis).
That 11-26 record after the May 10 loss to Baltimore made them the worst team in the AL by five games, and the worst team in the Majors by 1.5 games. Things were bad. Then the Cosarts beat Baltimore 5-2. A loss to the Rangers preceded the Astros winning their first series against the Rangers since Ever. Then they won a series against the White Sox. A minor offensive blip saw four straight defeats on the road and then EXXPLOSIVE.
Since that May 10 loss at Baltimore the Astros are 20-11, one game behind Toronto for the best record in baseball and just one more loss than the Giants, who have the best record in baseball. It's worth noting that this stretch puts them 10.5 games better than the Rays, 7.5 better than the Red Sox, 6.5 better than the Tigers, and six better than the Rangers.
What's remarkable about this is that they've been winning with a shortstop who has been absolutely God-awful with the bat.
Jonathan Villar was Not Terrible when the Astros were terrible. Through May 10, Villar was hitting .232/.289/.464. When games wrapped up on May 4, Villar was hitting .255/.320/.532. That will play. Since then the bottom has fallen out. If we keep Villar's scope within the scope of the two unofficial halves of the season, since May 10 (79 PAs) Villar has hit .151/.215/.164 with 24K:5BB and one extra-base hit. Villar has one extra-base hit (a double) in his last 99 plate appearances.
Part of this has to do with his massive home/away splits. Fifteen of the Astros games played since May 10 have been on the road, and Villar is hitting .134/.168/.216 on the road this season (.273/.354/.489 at Minute Maid).
From Opening Day to May 10, Villar was hitting .250/.289/.472 vs LHP and .224/.289/.461 vs RHP - somewhat consistent, given the 40 AB difference between the splits. At Minute Maid he was hitting .327/.407/.673, and .150/.177/.283 on the road. His home splits were covering his road splits (Villar was a combined 1x30 at Detroit, Toronto, and Arlington).
But then the bottom fell out. From May 11-last night Villar is hitting .190/.261/.190 vs. LHP and .135/.196/.154 vs. RHP. In 39 PAs at Minute Maid in this time frame he's hitting .194/.275/.222 (with 16K:3BB) and in 39 PAs he's hitting .108/.154/.108 on the road. He has a .393 OPS vs starters and a .362 OPS vs. relievers with twelve strikeouts each.
The conclusion? Villar looked alright at the beginning of the season because he provided some pop. A .289 OBP is terrible, but at least he'd advance past first base once in a while, and that SLG was covering over some glaring issues. Now, though, it's not. Villar enjoyed a three-hit game on June 8, and a two-hit game on June 11, so maybe he's starting to recover, but it's remarkable to me the Astros have been able to overcome the lack of a bat at short.*
*Also, Left Field and Catcher, but let's save that for another time.