clock menu more-arrow no yes

Filed under:

Astros 2018 Team Defense in Review

New, 39 comments
MLB: Los Angeles Angels at Houston Astros Troy Taormina-USA TODAY Sports

Baseball has always been a game of numbers. Ted Williams batting .400, Hank Aaron’s 755 home runs (and Barry Bonds’ 762), Joe DiMaggio’s 56-game hit streak, Nolan Ryan’s 5,714 strikeouts. The list of numerical and statistical accomplishments is endless (in a quite satisfying way). In recent years, data collection methodology and consumption has intensified (e.g., Trackman, Statcast, FanGraphs, Baseball Savant) and driven what can be best described as a revolution within the sport. The intricate analyses being performed have shifted the way fans watch and experience the game, how players approach the game, and how teams construct their rosters. Launch angles, exit velocities, spin rates, shifts, “Openers,” have all become common lexicon for baseball fans.

An area that has lagged, however, is the way defense and fielding are measured. Fielding percentage was relied upon as the gold standard for years, but the data revolution has changed that (straight-up replaced it, really). There are more defensive metrics and information available than ever, but still not much certainty about what truly classifies a defender as great.

Which is quite surprising because the Astros have shown how important defense can be to win a championship. Sure, the 2017 team had an all-time offense, but some of the biggest plays during that year’s postseason run came in the field for the Astros.

Litany of videos

That litany of videos was gloriously presented to show the impact defense can have on an outcome (and to indulge the Astros first World Championship again, of course), and most of them occurred in separate games. Defense is a startlingly important part of the game, and yet it still seems undervalued, underappreciated.

With all that said, I want to look at how the Astros defense fared in 2018 using current fielding metrics.

Astros 2018 Team Fielding Stats

Fielding % DefEff wOBACON xwOBACON xwOBACON - wOBACON 1-10% (Remote) 10-40% (Unlikely) 40-60% (About Even) 60-90% (Likely) 90-100% (Almost Certain)
Fielding % DefEff wOBACON xwOBACON xwOBACON - wOBACON 1-10% (Remote) 10-40% (Unlikely) 40-60% (About Even) 60-90% (Likely) 90-100% (Almost Certain)
0.989 0.717 0.352 0.343 -0.009 6.2% (162) 6th 34.7% (121) 1st 54.3% (105) 4th 81.9% (199) 4th 98.1% (2311) 8th

The Astros had the highest fielding percentage in baseball last season, but we know that is a flawed stat. A better metric to evaluate team fielding is Defensive Efficiency (DefEff), which represents the percentage of balls in play converted into outs. A year ago, the Astros ranked second in the AL and fourth in baseball in DefEff.

wOBACON and xwOBACON represent the actual and expected weighted on-base average, respectively, when contact is made. As eloquently explained to me by Willie McGee’s Twin, “subtracting wOBACON from xwOBACON will result in the difference between the value of what was expected to happen given the launch angle and exit velocity of balls hit in play (i.e., quality of contact) and what actually did happen. In other words, if a defense can turn more balls that have a high expected value into outs, then its actual wOBACON will be less than its xwOBACON. xWOBACON is what the pitchers give up, wOBACON is what the defense does with those balls after the ball is struck.”

The Astros had the lowest xwOBACON in the Majors last year, mostly because they had historic starting pitching. However, the Astros had a higher wOBACON than xwOBACON, which indicates their defense was worse than could be expected based on the quality of contact surrendered.

I also investigated the Inside Edge Fielding statistics from FanGraphs. A good summary on how to interpret these stats is provided here and here. Simply put, the Inside Edge Fielding categories represent how often a player at that position has made a very similar play. Houston was adept at making all types of plays, ranking in the Top 10 across all Inside Edge Fielding categories, and was the best at making “Unlikely” plays. (Note: The numbers in parentheses refer to the number of opportunities in a given category and league-wide rankings are shown to the far right).

One team metric I didn’t include in the table was Outs Above Average (OAA), which is the cumulative effect of all individual Catch Probability plays a fielder has been credited or debited with, making it a range-based metric of fielding skill that accounts for the number of plays made and the difficulty of them. According to Statcast, the Astros created 20 OAA, third-best in MLB.

Of course, the Astros shift their defensive positioning more than any other team in the league (in history, probably). The Astros shifted—which is defined as three or more infielders on the same side of second base—on 37.3% of plate appearances in 2018. That’s, like, more often than every three hitters! The next closest team in terms of shifting was Tampa Bay at 30.0%, and league-wide teams shifted for 17.4% of plate appearances. The Astros like to shift against lefthanded hitters in particular, doing so on 59.7% of plate appearances! The New York Yankees were second in that regard at 52.6%, and no other team shifted against lefties more than 44% of the time (league-average was 29.6%). Clearly, the shift is a big part of what the Astros do defensively. I wonder if their tendency to shift against lefties somewhat mitigates the need for a strong lefthanded presence in the bullpen?

Anyway, I’ll wait to dive into the defense by position for another article, but offseason additions—namely, Aledmys Diaz, Robinson Chirinos and Michael Brantley— as well as an expected increase in playing time for Kyle Tucker will have ramifications on the Astros defense in 2019. That said, although the team defense in 2018 may not have been elite, it was certainly solid. Considering the absence of and limitations to a number of starters for extended periods (e.g., Correa, Jose Altuve, Jake Marisnick), I’d say the Astros had a pretty good defensive unit in 2018 and are set up for similar success next season.

One final (and likely obvious) note. There are also still abilities unaccounted for by the mainstream fielding metrics, like the amount of respect an opposing team has for an outfielder’s arm. Such things can prevent great defensive plays from showing up on the highlight reel yet still have a heavy impact on the game. It’s a good demonstration that even though we have so many tools available to help make baseball decisions—in-game, during off days, throughout training in the offseason, in the front office—there’s no correct formula to quantify defense. Yet.