Game 33, Rays at Mariners
Wade Miley vs. Drew Smyly, 7:10pm
Sorry for the lack of posts here – just returned from vacation to watch Felix hold it together enough to pitch 7 strong against the Rays, and for Ketel Marte to have his best game as a professional. Let me bid all of you a happy belated Felix Day; I was encouraged by what I saw, though it’s still pretty clear Felix doesn’t yet have his royal command at his disposal. The two HBPs were particularly ugly, but Felix consistently missed his spots, especially with 2 strikes. That said, he was good enough to limit the Rays to 2 solo HRs, and the M”s defense ably handled the balls in play Felix allowed.
Last night felt like the first, clear, unambiguous, win for the M’s shift-happy ways. The M’s heavily shifted a number of Rays hitters, including guys pretty far from the ol’ David Ortiz/Adam Dunn template of “shiftable” hitters. Kevin Kiermaier was a great example – a guy who’s fast and doesn’t have a sky-high ground ball rate. It’s actually lower than average. Still, the combination of Kiermaier and Felix was one the M’s felt they could exploit, and they did: he grounded out in his first two trips against Felix, and hit a tailor-made DP ball the 2nd time that the M’s bungled into a FC. I’ve been thinking about the shift a lot recently, as Manny Acta’s tweeted a couple of times about how successful it’s been for the M’s. At the same time, though, analysts are starting to question why we’re not seeing more evidence of its utility now that teams are shifting 10X more than they did 5 years ago, or whatever the actual multiplier is. It’s gone from a very rare play, probably done by only a couple of teams, to a league-wide phenomenon. And yet BABIP is essentially unchanged. Teams wouldn’t DO it if it didn’t work, but, uh, how do we know it works?
Russell Carleton at BP had an interesting article a week ago arguing that it simply doesn’t work, and that a possible explanation is that pitchers pitch differently when they know their defense is shifted behind them. As a result, it’s possible that they give up, say, more home runs – they may be able to induce more GBs overall, but by sticking to smaller segments of the plate, or using specific pitches in specific locations, any “miss” may end up getting punished more severely than it would if they were pitching normally (read: more unpredictably). What we can see at a league-wide level says that BABIP is unchanged, and actually moving higher recently. If shifts don’t save enough hits to make a dent in BABIP, as this Rob Neyer tweet indicates, then…what? It seems to clearly limit singles to a certain set of players, but other than that, it all kind of washes out? I doubt that, given how enthusiastically teams have been adopting it, but I’d love to know more about how the M’s measure success. The most stunning thing I’ve learned in looking into this was that there’s no publicly available data about home runs when the defense is shifted. Fangraphs shows some stats with and without the shift, but the denominator is *balls in play*. Does K rate tank with the shift on? Do batters walk more? Fewer HRs, or nah? I have no idea. That seems like an easily addressed issue, but until then, what do these shift numbers mean?
Hey, so there’s a game today. Drew Smyly’s someone I’ve had an eye on for many years, since he was a promising hurler at AA for the Detroit Tigers org. Why did M’s fans suddenly care about Drew Smyly? Because the M’s and Tigers had just agreed on a deal that sent Doug Fister to Detroit in exchange for a package including Charlie Furbush and Caspar Wells, and a mystery prospect who’d be revealed later. Nick Castellanos was the big name, but most people assumed there was no chance he’d be thrown in. It came down to two pitchers: a reliever named Chance Ruffin, and the over-the-top starter, Drew Smyly. I have no idea if the M’s made the choice, or if Detroit decided to protect Smyly from poaching, but the M’s ended up with Ruffin, who retired a few years back, and the Tigers were able to use Smyly to land David Price.
Drew Smyly has missed a lot of time due to shoulder and other ailments, but when he’s toed the rubber for Tampa, he’s been an exceptional pitcher. As this Jeff Sullivan post details, Smyly’s racking up strikeouts at an impressive pace, especially for a guy with below-average velocity. Smyly struck out a lot of minor league hitters, but his ceiling always seemed limited both by his lack of top-shelf velocity and the fact that he didn’t have a big breaking ball. As I noted way back when he made his first start against Seattle in 2012 (holy crap I’ve been doing this a long time), he throws a cutter that features pretty much no discernable horizontal movement. Not 0″ of break, but it doesn’t look that different from his (straight) four-seam fastball. He has a slurvy curve/slider thing that likewise features basically no curving action whatsoever – his fastball gets 4+” of armside run, and his “curve” gets 2″ of armside run. I don’t think I’ve ever seen a curve that curves less, or slides less, depending on what you want to call the thing.
It hasn’t mattered at all. As Jeff’s article points out, the Rays got Smyly to throw his extreme-riseball four-seamer up in the zone, and batters have had trouble not only with *it*, but with everything else he throws, too. He gets whiff per swing rates of over 30% on his fastball, his cutter AND his curve. I said four years ago that he was going to need to develop a change, and while he has one, his entire career has been a rebuke to my by-the-book prediction of future problems with opposite-handed-hitters. Instead, his fastball’s turned into one of the game’s least likely weapons. It’s not just that his whiff rate is so high; the corollary’s actually more shocking to me. Batters have a hell of a time putting Smyly’s 90-91mph fastball in play. This year, batters have offered at 45% of his four-seam fastballs. They’ve taken another 38% for called strikes. When they swing, they come up empty on 32% of them. They foul off another 42%. Put it all together, and batters are putting 12% of his fastballs into play. Since the beginning of 2015, that figure is just 13%.
For some context on that, batters put 16% of Noah Syndergaard’s four-seamer in play (and over 20% on his sinker). Stephen Strasburg? About 18%. Andrew Miller? 15% last year, so hey, pretty close. The only pitcher I can find with a modicum of effort that has a lower BIP rate is Aroldis Chapman. That’s kind of insane. For starters who are in the same basic ballpark, the guy who’s the closest parallel is probably Marco Estrada, who’s at around 14% this year and was in the 16% range last season. Like Smyly, he’s got a four-seam with a ton of rise, and he’s not afraid to pitch up in the zone. While Estrada’s not getting quite as many whiffs with his heater, he’s getting even more foul balls this year, and he drew a ton of fouls last year as well.
An obvious question is: how to teams attack this? Can you pick up a high spin fastball on video and subtly adjust your swing in response? Or is adjusting your swing in response to each opposing starter a sure recipe for messed-up mechanics and a self-imposed slump? The M’s looked troubled at times by Matt Moore’s extreme *horizontal* movement – movement that seemingly allowed Moore to sneak strikes when non-Marte M’s took pitches middle-middle. Dae Ho Lee looked confused by a first-pitch center-cut fastball, and I think he did it to Kyle Seager as well. If Smyly leaves any middle-middle pitches hanging around the M’s need to attack them. They also have to be mindful of the fact that Smyly thrives by getting hitters to expand the strike zone upwards; limit the swings on pitches at eye level, fellas.
1: Marte, SS
2: Gutierrez, LF
3: Cano, 2B
4: Cruz, RF
5: Iannetta, C
6: Seager, DH
7: Lee, 1B
8: Sardinas, 3B
9: Martin, CF
SP: Miley
Tacoma’s Stefen Romero and Chris Taylor have continued their hot streaks, which makes up for Mike Zunino going a bit cold over the past couple of weeks. The M’s split a four-game series at Albuquerque and now take on Fresno tonight behind Cody Martin.
Jackson finished up their own four-gamer with Jacksonville, and they too split the series. Jacksonville won an 18-inning contest on Friday. The Generals are in Birmingham now, and have Dylan Unsworth on the hill tonight. The South African righty’s started off the season well, limiting hits allowed and posting an ERA under 1 over his first five starts.
Bakersfield suffered the ignominy of being swept in a series to Inland Empire, the Angels’ affiliate and a team that came in 7-21. Inland Empire “features” Cuban 2B Roberto Baldoquin, the player that’s been Jerry Dipoto’s biggest miss in the international market. Baldoquin signed an $8m bonus a few years back, one of the highest bonuses ever for a bonus-pool player. Playing in the Cal League last year and this year, Baldoquin has compiled a .290 slugging percentage and a .262 OBP. Bakersfield hosts San Jose tonight.
Clinton won their series with Lake County 2-1, and won the opener of their series against ex-affiliate Wisconsin yesterday. They’ll face off with the Brewers affiliate today, with Art Warren on the mound opposite Dominican teen Marcos Diplan.
Speaking of the minors, the News Tribune’s Bob Dutton linked to a Baseball America report on minor league park factors that found Cheney Stadium was the most extreme home run park in the PCL. To say this is a counterintuitive finding is quite an understatement. For years, Tacoma’s been the preeminent pitcher’s park in the hitter-friendly PCL. Last year, though, teams hit more HRs in Tacoma than in any other park in AAA, and hit 39 more at home than were hit on the road. That’s…remarkable, and so strains what we know about places like Albuquerque and Las Vegas that I wonder what they did to account for schedule and players. Statcorner’s park factors show a very distinct gap between the factors for lefties (home run heaven) and righties (average-to-a-bit-below). Interestingly, Tacoma still grades out as a pitcher’s park because run scoring – despite the maybe a fluke, maybe not dingerfest last year – is still much lower there than in other PCL parks. At this point, it’s apparent that the remodel had a much bigger impact on HR/FB than I ever would’ve thought possible. I’m still betting the under on the park factor BA reported, and by a lot, but it’s clear Tacoma’s much more like an average park (and may in fact be aiding HRs now) than it was in the good ol’ days.