In September, an old man’s fancy turns to the Greatest American Sport – greatest, that is, if you’re addicted to statistics. Mavens of Major League Baseball are poring over the numbers and chicken entrails to predict the playoffs.
How to size up hitters? The popular yardstick is the batting average – the share of times at bat that the player gets a hit, discounting an error or a fielder’s choice (when the other team lets the batter go to first while it throws out another player). The batting average is normally about one in four. As I write, the hottest hand is attached to Yordan Alvarez, designated hitter for Houston, who is merrily clicking along at .312. The Astros aren’t complaining: They’re leading the American League West, with 75 games won and 72 lost.
The batting average is easy to understand, but it misses the point. To win the game, you must cross home base more often than the rival. First base is just the first step.
What other yardsticks can we use? Well, the number of home runs is spectacular – Kyle Schwarber, a designated hitter for Philadelphia, has hit 44 so far, although the Phillies could use a few more: With a record of 82-65, they’re only second in the National League East, 4 games behind the Atlanta Braves. Maybe that’s because this statistic counts only some of the runs that the batter has set into motion. Even his single can be batted in later. So we should tally all his hits – singles, doubles, triples, and home runs, giving more weight to the hit that is closer to home, since it is more likely to become a run.
The slugging percentage fills the bill. It rewards a point for a single, two points for a double, three points for a triple, and four points for a home run. Thus weighted, the sum of hits can be compared to the number of at bats. In short, the slugging percentage is the hitter’s number of bases per at-bat. Alvarez of the Astros is slugging away this season at .593, or nearly three-fifths of a base per at-bat.
Slugger’s choice
Already you see the problem. In reality, the impact on the team’s score of a homer rather than a triple is greater than the impact of a double rather than a single. And yet the slugging percentage awards just one more point in either case. It also ignores walks and pitches that hit the batter, although these put him on base just as surely as a single.
Let’s add to the brew the on-base percentage. This is how often the player gets to base in a typical at-bat. Hits, walks, beanies, whatever works. Houston’s Alvarez sizzles at .428; 43% of his at-bats wind up on a base. As always, we will exclude base trips due to an error or a fielder’s choice. The idea is to gauge the batter’s skill, not his luck. (So why do we count beanies? Search me.) However, the on-base percentage strangely weights all base trips equally. A homer is worth no more than a single.
Back to the drawing board. Since the slugging percentage and the on-base percentage reflect different aspects of the batter’s performance, let’s add them together. That’s the On-Base Plus Slugging statistic. King of the stat is Babe Ruth, with a career value of 1.164. Breathing down his neck are Ted Williams (1.116) of the Boston Red Sox (the Yanks’ historical nemesis, though not this year, yet) and Damn Yankee Lou Gehrig (1.079).
The OPS may be the favorite batting statistic of aficionados of inside baseball. Unfortunately, just adding two statistics together makes no sense here. Why should the on-base percentage get the same weight as the slugging percentage? And what does the sum really mean?
OMG, another statistic
Maybe we should take another whack at the question. Our meta is a statistic that shows how much a batter contributes to the team’s runs. It considers two possibilities: He batted in runs himself; or he will be batted in later in the inning. Let’s look at the formula, then demystify it:
Chance of Runs = [Player’s runs batted in per Player at-bat] + [Team runs batted in per Team at-bat] * [Player’s slugging per Player's at-bat + Player's stolen bases per Player's at-bat]
Chance of Runs is the probability that the player generates a run. On the right-hand side of the equation, the first term is the probability that the player bats in a run. The second term is the probability that the player himself is batted in. This second term is the number of bases that the player reaches in a typical at bat, times the probability that someone else on the team will bat him in, in a typical at-bat.
Unlike my previous posts, this one subtracts the player's RBIs and at-bats to calculate the team's, to avoid double-counting. It also adds stolen bases. These changes are not trivial. On a team like the Nats, where most batters are as weak as kittens, an outstanding hitter like Abrams accounts for a respectable share of the team RBIs. Failing to exclude it boosts the calculated team RBIs per at-bat and consequently the player's Chance of Runs. Stolen Bases also matter for an Artful Dodger (apologies to LA) like Pete Crow-Armstrong of the Chicago Cubs. The center fielder has stolen 35 bases, and the night is still young. However, the formula excludes walks and beanies; it focuses on what the player achieves by his own bootstraps.
The table below lists the top 25 players in Major League Baseball in the current season up to September 7, using MLB’s official statistics. The standouts are Jordan Walker, right fielder of the St. Louis Cardinals, and Alvarez of the Astros. They exceeded a chance of one in four of sparking a run in an at-bat. All 25 players topped a chance of one in seven of an eventual run in a typical at-bat.
Of the 25 players, 9 are first basemen. This may reflect the fact that first base is one of the easier infield positions, leaving the player free time to hone his batting chops. On the other hand, shortstop may be the most demanding position on the diamond, and only two shortstops made the list of the top 25 hitters. CJ, of the bottom-feeding Nats, really is a magician.
The table also suggests that home runs are not the alpha and omega of baseball. Schwarber, king of homers, ranks only 16th on the Chance statistic.
As statistics go, mine is undoubtedly a swing and a miss. For instance, it assumes that the increase in the chance of a run is the same for a homer rather than a triple as for a double rather than a single. More tinkering in the garage is on the agenda. My only excuse for Chance is that it's easy to interpret. And it keeps me off the streets at night.
– Leon Taylor, Seymour, Indiana tayloralmaty@gmail.com
Rank Player Team Chance Position
1 Jordan Walker STL 0.259 RF
2 Yordan Alvarez HOU 0.251 DH
3 James Wood WASH 0.242 RF
4 Junior Caminero TB 0.233 3B
5 Max Muncy LAD 0.232 3B
6 CJ Abrams WASH 0.229 SS
7 Brandon Lowe PIT 0.228 2B
8 Manny Machado SD 0.211 3B
9 Miguel Vargas CWS 0.207 3B
10 Sal Stewart CINCY 0.197 1B
11 Luis Garcia NYY 0.192 1B
12 Willson Contreras BOS 0.183 1B
13 Pete Crow-Armstrong CHC 0.183 CF
14 Pete Alonso BALT 0.177 1B
15 Ben Rice NYY 0.172 DH
16 Kyle Schwarber PHI 0.170 DH
17 Rafael Devers SF 0.168 1B
18 Hunter Goodman COL 0.167 C
19 Dilson Dingler DET 0.164 C
20 Shohei Ohtani LAD 0.162 DH
21 Kazuma Okamoto TOR 0.161 3B
22 Colson Montgomery CWS 0.161 SS
23 Bryce Harper PHI 0.160 1B
24 Muneta Murakami CWS 0.160 1B
25 Matt Olson ATL 0.144 1B
No comments:
Post a Comment