Opponent: Houston Astros (again)
Starting pitchers: Jordan Lyles vs. Bartolo Colon
First pitch: 7:05 PT

A's in the West: One game back of Texas
A's in the Wild Card: Leaders for first Wild Card
Baseball Prospectus playoff odds: 36% division, 44% Wild Card

You know that Flight of the Conchords song "Rap (Is Not Very Good)"?¹ Well, Jordan Lyles is Not Very Good. Of pitchers with a minimum of 400 batters faced, he is 111th out of 116 in ERA, 97th in strikeout percentage, and 55th in walk percentage. He doesn't miss bats and he doesn't throw the ball in the strike zone. He does actually get good downward action on his sinker, and he's in the top quarter of starting pitchers in terms of vertical break on his curve, but these pitches obviously aren't getting the job done for him.

He might just be on the wrong team. As a ground-ball pitcher who doesn't miss bats, he needs a strong defense behind him, but Houston has one of the worst defensive efficiency ratings in the game, and Baseball Prospectus's park-adjusted DE says that they turn about 1.5 percent fewer balls into outs than average. This doesn't sound like that much, but it's 29th in the league and applying linear weights to the numbers says that defense at this level should be expected to cost the Astros 30 or 35 runs over the course of the season, which makes things sound more dire in terms of how they're treating poor Mr. Lyles.

Compare Bartolo Colon, after all, who gives up even more balls in play than Lyles does, what with his 111th-ranked strikeout percentage. But, along with giving up more fly balls, Colon has a stellar defense behind him (though you wouldn't know it to watch Jed Lowrie), so his BABIP allowed is .287, a full 25 points lower than Lyles's. As yesterday and as always, if you can tease out how much of that 25 points is attributable to all the various relevant actors on a ball in play, then good for you, but given the other information we have, it seems reasonable to believe that the defense is playing a not-insubstantial role.

With a right-hander on the mound, we probably weren't going to see Derek Norris today whether he's hurting or not, so today's lineups likely won't give us any information about his back (though I'm sure the pregame reports will include something explicitly about how his back is feeling better and he's hoping to give it just one or at most two more days; or I guess he could go on the disabled list today). What we will find out is how Coco Crisp is feeling, because with Lyles on the mound, Chris Young should be nowhere near the lineup. If Crisp's wrist still isn't feeling up to snuff, though, Young to center it will be.

Alternatively, Young will play center because Bob Melvin believes in his career five hits in nine at-bats against Lyles, but I'd very much like to believe that that will not be the case.

Yoenis Cespedes since June 22nd (arbitrary!): .211/.282/.324. That the A's are one game out of the division lead and would host the Wild Card play-in game as of today with Cespedes hitting like that, with Josh Reddick batting .203/.287/.326 prior to his five-homer outburst, with Chris Young's batting average slipping under .200 on April 12th and rising back above the Mendoza Line just once (May 25th) since then (after the game of May 26th, he was back down to .198) is tremendous. And this isn't a low-quality team that's lucked its way into playoff position: by BP's third-order record (which builds an expected record based on the actual performances of the players and then adjusts for opponent quality), the A's "should" be the second Wild Card behind the Red Sox (with the Rays winning their division). That is, there's a way in which they've out-won their performance, and they are a .600 team in one-run games against .561 in all the others, but .561 is still damn good.

Prediction: A's win.