The Value of Alex Gordon Not Using His Arm by Jeff Sullivan June 23, 2014 There aren’t many plays quite like the challenging of an outfielder’s arm. If you think about it, hitters don’t really have a choice, when they’re in the box. Runners don’t really have a choice, when a ball’s hit to an infielder. But when the ball goes to an outfielder, runners can opt in to an arm test, wherein they attempt to beat the ball to a bag. It’s a challenge of arm against legs, and when the arm emerges victorious, it can make for some memorable moments. Just this past weekend, Marcell Ozuna went crazy in consecutive innings. Less recently but more memorably, the Angels made the mistake of challenging Yoenis Cespedes. Outfielders with the best arms tend to be outfielders who rack up the most kills. Alex Gordon‘s always had a great arm. Alex Gordon’s always piled up the kills. Between 2011 – 2013, Gordon led all outfielders in UZR’s arm rating. He led all outfielders in DRS’ arm rating. He led all outfielders in assists, with 54. The next-best was Jeff Francoeur’s 40. Gordon was drafted as a third baseman but he’s become an all-around star in left field. This season, Gordon has just five outfield assists, almost halfway through. The last three years, he’s finished with 20, 17, and 17. This season, Gordon’s also on pace for career-best arm ratings. Alex Gordon is showing the value of having a gun you seldom use. Rewind to the very beginning of April. The White Sox were in Kansas City, and Alexei Ramirez pulled a grounder down the line into left field’s foul territory. Gordon scooped the ball up. Said the White Sox broadcast: Harrelson: There’s a rule that is just a conceded double, but not with Gordon out there. Stone: There’s one guy in this outfield you don’t want to run on. Ramirez hit a possible double, and aggressively rounded first to see if he could go for a double. He looked up, saw Alex Gordon, and slammed on the brakes. Ramirez still wound up with a hit, but he wound up with half the total bases he probably expected to get. And this is where Gordon’s numbers are really outstanding. In the past, he’s racked up a lot of value by throwing runners out. So far, he’s racked up a lot of value by having his arm serve as a deterrent. Some assists, certainly, have still been there. And Gordon has yet to be charged with a throwing error. But the bulk of his arm value is coming from his not having to even use his arm. Here’s some data you might not have ever looked at before. Sure, there are assists, but there are also holds, where a runner doesn’t advance an extra base. That’s the other way for an arm to be valuable, and Gordon so far has been extraordinary in this regard. Let’s go over a simple rundown, shall we? On Gordon’s 2014 performance in left field: Single with a runner on first. So far, Gordon has held the runner 96% of the time. The league average is 80%. Single with a runner on second. So far, Gordon has held the runner 77% of the time. The league average is 36%. Double with a runner on first. So far, Gordon has held the runner 58% of the time. The league average is 54%. Fly out, runner on third, less than two out. So far, Gordon has held the runner 43% of the time. The league average is 24%. (The sample here is also really small.) (I feel stupid using percentages.) Fly out, runner on second, less than two out. So far, Gordon has held the runner 100% of the time. The league average is 89%. This completely ignores assists. This is just about runners challenging or not challenging, and so far, Gordon has held the runner 82% of the time, overall. The league average is 63%, so if you’re content with simple math, you could equate that to about 16 saved bases. A different way of looking at this: against Gordon, there’s been a 14% advance rate. The league average is 35%. Every saved base has a run value, and some of the saved bases are home plate. A saved base in this regard is less impactful and less dramatic than an assist, in that it doesn’t create an out, but value is value and Gordon’s on track for a career high. There are obvious issues with this data. Not every opportunity is created alike, and maybe Gordon has been defending against unusually slow runners. Maybe the balls in play he’s been fielding have made it extra unlikely for a runner to try to move up. With this sort of data, you want bigger sample sizes to try to get greater evenness of opportunities, and it’s worth noting that, just last season, Gordon had an overall 67% hold rate. His numbers held pretty steady before jumping up in 2014, so it would be strange if runners only just now suddenly lost the will to be aggressive. The short of this being, there are error bars, as there are whenever you break something down to the core components. It stands to reason runners don’t like challenging Gordon; it stands to reason we don’t know his current true-talent level, in terms of holding runners without advance. But this is why Gordon’s numbers are where they are. His assists are present, but down. He has yet to make a throwing error. And runners have more or less stayed put, instead of putting Gordon’s arm to the test. Maybe down the stretch, they’ll run a little more often, as the sample size balances out. But, let’s go back to that April White Sox/Royals game. The batter after Alexei Ramirez doubled. One way or another, Alex Gordon’s arm is going to accumulate its value. How it does that is up to the runners.