How Houston’s pitchers beat Minnesota’s hitters

Many factors can reasonably be pointed at as the cause of the playoff sweep that provided another disappointing end to a Twins season. Here’s a sampling of my favorites: Rocco Baldelli overmanaged Game 1. Josh Donaldson’s injury. Byron Buxton’s injury. Jorge Polanco’s error in the 9^th inning of game 1. An over-reliance on a suddenly leaky bullpen. Bad luck on Eddie Rosario’s lineout with the bases loaded in the first inning of Game 1. You likely have your favorite among these and no doubt have several others to add to the stack. Despite the wealth of potential choices, perhaps no factor was as impactful to ensuring the Twins exit to the Astros than the collective disappearing act performed by the Minnesota offense. The weak attack from the bats yielded just two runs scored on a grand total of seven hits over the 18 playoff innings. Twins’ hitters went a combined 7 for 59, drawing 10 walks against 14 strikeouts. Only two of the hits went for extra bases (Nelson Cruz doubles) in the two games and 28 baserunners were stranded. No matter what else had happened on the mound, in the field, or on the bases, it would have been very difficult to win with that level of offensive production.

Offensive production proved to be Minnesota’s weakest link (surprisingly) all throughout the 2020 season. The Bomba Squad of 2019, despite the adulation of the homeruns, was also a reasonably balanced and well-rounded offensive attack. The 2020 Twins still bopped a lot of homeruns but struggled to score runs without them and entered the playoffs as the American League team most dependent on homeruns to score. After leading the majors in numerous statistics measuring quality of batted ball contact in 2019, the 2020 club was middle of the pack at best – despite mostly the same players. In some ways, perhaps we should not have been too surprised the offense would be the team’s downfall in the postseason. But that doesn’t make it any less frustrating for those associated with or interested in the Minnesota Twins to watch the team’s postseason aspirations evaporate with each feeble swing, disputed called strike three, or unlucky at ‘em ball.

Now that a few days have passed and it’s a little easier for me to be rational and objective as an analyst, I needed to break down the series. Specifically, I needed to look at the Astros’ pitching approach against the Twins hitters. What happened to the offense? Was there something clear and obvious in the Astros’ approach that might explain the results?

Houston’s Strategy

Of the 18 innings covered by Houston hurlers, 16 and a third of them were handled by four piggy-backing starting pitchers – Zack Greinke, Framber Valdez, Jose Urquidy, and Cristian Javier. The remaining five outs were handled by matchup lefty Brooks Raley and closer Ryan Pressly. Houston manager Dusty Baker has often been derided for being stuck in his old school ways, but his handling of his pitching staff earlier this week implies that he has paid attention and learned a few things from his contemporaries the past few postseasons. Following the philosophy used by Cleveland’s Terry Francona, Los Angeles’ Dave Roberts, and Washington’s Davey Martinez (to name just a few) in recent tournaments, Baker entrusted as many innings as he could to his best pitchers regardless of their regular season roles. This is an approach that makes a lot of sense in a short series and when considering Baker possessed a young, inexperienced bullpen that contained only one player with more than a single year of MLB service time (Pressly).

Houston’s Tactics

Thanks to the availability of Statcast data on baseballsavant.mlb.com we can analyze the two games and the pitchers with pitch by pitch granularity. In game 1, starter Greinke dodged big trouble when he escaped the Twins’ bases loaded rally in the first inning. Greinke settled after that and added three more solid innings, giving up one run with two hits and three walks in his four-inning outing. Greinke threw 79 pitches – relying mostly on his four-seam fastball (38) and changeup (19). The remaining 22 offerings were breaking pitches, equally split between curveballs (11) and sliders (11).

Greinke mostly worked down in and below the strike zone. There is a clear pattern in the chart above that shows Greinke threw his fastball for strikes (red) and his other pitches outside the strike zone, especially his changeup (green) and slider (yellow) down below. Greinke averaged 88.7 mph with his fastball in this game, topping out at 91.3 mph. Despite the lack of big velocity, Greinke was aggressive going after Twins hitters with his primary pitch.

Greinke was relieved by left-hander Framber Valdez to begin the fifth. Valdez would complete the game, throwing five scoreless innings, allowing two hits and two walks, while striking out five. Primarily a two-pitch pitcher to begin with, Valdez threw 66 pitches – 36 sinking fastballs and 30 curveballs.

There wasn’t anything tricky or new about Valdez’s approach. His 55% rate of sinkers was right in line with his season average, and he scrapped his changeup (clearly his third best pitch) in favor of a few additional curveballs. He challenged the Twins with his best stuff and won. Like Greinke’s chart above, there is a clear pattern of challenging hitters inside the strike zone with his fastball and working at the bottom of or below the zone with the breaking ball.

Game 2 was started by right-hander Jose Urquidy. Like Greinke a day earlier, Urquidy escaped major damage when he left the bases full of Twins in the bottom of the first. From there, he would work into the fifth inning, allowing a run on two hits and two walks. Urquidy delivered 76 pitches – 49 four-seam fastballs, 12 curveballs, 9 sliders, and 6 changeups.

Again, Urquidy followed what seems to have been the Astros’ gameplan for the Twins. Most of those fastballs found their way into the strike zone, with the offspeed and breaking pitches either used early in counts to get easy strikes or down and away as chase pitches.

After left-hander Brooks Raley finished the fifth with his cutters and sliders, right-hander Cristian Javier took over for the sixth, seventh, and eighth. Javier allowed just two walks in his three innings of work. Of his 54 pitches, a full 47 (!) of them were four-seam fastballs. Five sliders and two curveballs completed his offerings.

True to the script, Javier pounded the strike zone with his fastball. He mainly worked that heater to his glove-side against right-handed hitters (away from them) and to his arm side against left-handed hitters (also away).

Altogether, Houston hurlers offered 302 pitches to Minnesota’s hitters. 184 (60.9%) of them were fastballs or sinkers. Greinke, Urquidy, and Javier all threw their fastballs noticeably more often in the Wild Card games than they had throughout the season.

Greinke had only one start this season where he threw his fastball more than the 48 percent he did against Minnesota. Urquidy’s 64% was his highest rate of the season. Javier was the most dependent on his fastball of the three during the season and he kicked that into overdrive with his 87% fraction on Wednesday (also his season high). Pressly is perhaps the clearest signal the Astros were committed to challenging the Twins with fastballs. For the season, he threw nearly two breaking balls for every fastball. In his Game 2 appearance 10 of his 13 pitches were four-seam fastballs.

While two games is a very small sample and ripe with opportunity to overreact to randomness, the Astros rate of fastballs for the season was a much different 50.8%. It doesn’t seem like a random blip. It seems their game plan was to challenge Minnesota with fastballs.

The Outcomes

I wrote a couple of weeks ago about the Twins dramatic regression in performance against fastballs this season. After destroying the baseball universe’s best fastballs in 2019 (115.2 wFB), Twins hitters in 2020 finished just 17^th against fastballs (0.8 wFB). Using a per 100 pitch basis to adjust for the short 2020 season, the decline was from 0.16 to 0.02.

The Astros targeted this weakness and the Twins’ struggles continued. Against the 184 pitches classified as four-seam fastballs or sinkers, 92 (exactly 50%) of them resulted in strikes (53 called or swing and miss) or foul balls (39) against Twins hitters. 64 were taken for balls. The remaining 28 were put in play, mostly unsuccessfully. Just ten were classified as hard hit (95+ mph exit velocity) – four of those by Miguel Sano alone – and only eight had even an expected batting average over .300. Altogether, four the Twins seven hits in the series came against the hard stuff, while 24 fastballs put in play resulted in outs.

Going Forward

The challenge of the 2020 offseason for teams and those of us who analyze and write about baseball will be discerning how much stock to place in the 2020 campaign. 60 games is much different than the usual 162 and they were played in unusual conditions that could have impacted outcomes in uncountable ways. Further complicating the issue is the weird 2020 season came when we still had not completely sorted out the impacts of the juiced baseball that showed up in 2019 (and maybe is still around in 2020?). At any rate, the last two years’ worth of data are more uncertain than usual. This is especially impactful to a club like the Twins, who saw numerous young players post career-bests in 2019’s hitter friendly conditions, while seeing some of those same players crash back down to earth in the challenge of 2020. Sorting the wheat from the chaff is difficult to begin with and it will be especially so this coming offseason when the team will need to make updated assessments about players like Jorge Polanco, Byron Buxton, Mitch Garver, and Max Kepler.

Whether the Twins’ fastball woes were small sample noise that would have corrected with 102 more contests, we’ll never know. But that’s beside the point. Regardless of if it is two games, 60 games, or 162 games, it remains critical for major league hitters to handle fastballs to have sustained success. In 2020, Minnesota’s struggles against fastballs were signal enough for a postseason opponent to actively target and successfully exploit. Houston’s strategy wasn’t complex or novel. Nor was it based on elite stuff. It’s one thing to struggle versus Justin Verlander and Gerrit Cole fastballs. It’s another to be dominated by the fastballs the Astros offered last week. Greinke ranked in the third percentile in fastball velocity, Javier was in the 39^th, and Valdez and Urquidy in the 52^nd. Velocity isn’t everything, but it’s significant.

It was simple strategy the Twins couldn’t answer and it resulted in another disappointing end to an otherwise positive season. That alone should be enough to merit the Twins management and coaching staff giving the issue of fastball production significant offseason attention.

---

John is a contributor to Twinkie Town with an emphasis on analytics. He is a lifelong Twins fan and former college pitcher.