“You could study that”--as I mentioned in my intro to the first “Hey Bill” post, that was like a mantra with Bill James. It’s how he approached the kinds of basic questions baseball fans were always asking, many of which had answers that seemed so self-evident, no one had ever bothered to see if they were true or not…”Who should be the MVP this year?” (In the ‘79 Abstract, he wrote a long thing comparing Jim Rice’s case to Ron Guidry’s, based on their relative contributions as a hitter vs. a pitcher--which was, above all else, a refutation of the idea that a pitcher simply couldn’t be the MVP pitching once every four or five days.) “Are the _________ for real this year?” (In the ‘85 Abstract, prompted by the 84 Tigers’ 35-5 start, he tried to figure out at what point a fast start became meaningful--very early on, it turned out.) “That guy is so over/underrated”--he once tried to systematically figure out what causes a player to be over- or underrated. (I remember him concluding that Steve Garvey was like a perfect storm of overrated-ness.) That was the foundation of what eventually became sabermetrics; ask a question, figure out a method for addressing it, start collecting data.
With that in mind, I thought I’d test out something I posted on Facebook in connection to Davey Lopes’ death last week. Lopes was a solid player, just the kind of player that James identified as often underrated (working from memory here):
1) his batting average was ordinary (.263), but he took a lot of walks (74/162 games), so his on-base percentage (.349) was better than many other guys who’d hit .280 or .290, the Bill Buckner type of hitter;
2) his contributed defensively as much as he did offensively (harder to quantify, but he did have a lifetime dWAR in the positive range, one GG, and had an errorless streak of 28 games once);
3) he played on a team filled with stars;
4) he had (what was then) good medium-range power for a middle infielder; 10-15 HR a year, with a high of 28, and a slugging percentage of almost .400;
6) he was Black.
The most amazing thing to me that turned up in his obituaries--something I knew but had forgotten--was the 8-1/2 consecutive seasons that the Dodgers’ infield of Lopes (2B), Garvey (1B), Bill Russell (SS), and Ron Cey (3B) remained intact. Their first game together was June 23, 1973, and they stayed together as a unit until 1982, when Lopes left for Oakland. That doesn’t mean that they all played every single game, but every year from 1974 through 1981, they are listed as the Dodgers’ primary infield on Baseball Reference’s team page. I’m not sure how they determine that, but my guess is that each of them played more games at their given position than anyone else. Example: Russell was injured for much of the ‘75 season, but he still played more games (83) at short than Rick Auerbach (81). (Incredible: there was a season where the Dodgers split shortstop duties between a Bill Russell and an Auerbach.) All told, I’m pretty sure they were all playing 140+ games in almost all of those years.
What I wrote on Facebook: “That just couldn't happen today. As an experiment, think of the most successful teams this century, and I'd be surprised if you could find an infield that stayed intact for more than three seasons. Even the Dodgers from 2020-2025 did not go two seasons without at least one infield change.”
Just an intuitive, off-the-cuff remark. “You could study that,” I thought, so I have.
Immediate disclaimer: me studying something (you’ll find a handful of other such things elsewhere on this blog: Triple Crowns, the ‘92 Jays’ pitching staff, Dave Stieb’s bizarre 1985 season) is not Bill James studying something. I’m working with Baseball Reference and an ordinary desktop computer; I don’t pay for their Stathead feature, where, if you know how to set up a query, you can hit a button and generate a list of left-handed hitters who hit between 35-40 doubles while playing more than one position; and even if I were a paid subscriber, I doubt I’d know how to set up a query for what I wanted in this case. So I have to really narrow things down, and I have to search manually, skipping from screen to screen. I’m in a better place than if it were 1975 and I was using the MacMillan Encyclopedia, a Texas Instruments calculator, and a pad of graph paper, true; let’s say I’m about halfway between here and there.
I thought of two relatively manageable ways I could look at the question of how likely it is that a team could put together that kind of streak today. The first would be to take all teams this decade and track their yearly turnover in the infield. I eliminated the 2020 season because of COVID, leaving four seasonal changeovers for each team: 2021-22, 2022-23, 2023-24, 2024-25. That gave me 120 data points to look at. Compared to the whole history of baseball, that’s a small sample size. But: in trying to figure out the likelihood of doing today what Lopes’s Dodgers teams did in the ‘70s, only the most recent data seems relevant anyway. Patterns of player movement in 2026 are obviously not the same as those patterns were in 1936 or 1956, before free agency. They’re probably not even the same as in 1996, when Albert Belle signed a five-year contract for $55 million with the White Sox; Shohei makes almost that much per year now. Anyway, I think the recency of data here is more important than the quantity. As Cyndi Lauper and the Brains said, money changes everything.
The A.L. data:
The N.L. data:
Pretty straightforward what the numbers mean…that 3 under 2022 for the Yankees means that the 2022 team had three different players manning the infield when compared to the 2021 team. Actually, that was the first team I looked at, and I immediately encountered a situation that would turn up now and again: a player who stayed in the infield but switched positions from season to season, like Gleyber Torres did when he went from SS in 2021 to 2B in 2022. I counted that as a change. Lopes, Garvey, Russell, and Cey didn’t move around; they played where they played and stayed there.
All the 0s (no changes) and 4s (someone new at every position) are highlighted in yellow. There’s a 2 under the 2024 Padres highlighted in blue: the 2024 team had the same four guys in the infield as the 2023 team, but Xander Bogaerts and Ha-Seong Kim flip-flopped at second and short.
The really hard part of doing this, and where I’m just a softball-league/armchair version of James, is drawing conclusions from whatever data you assemble. (And also knowing when not to draw conclusions--realizing that you have to reframe the question to arrive at something meaningful, or when you have to discard it all together.) I’ll get to that part in a follow-up post.
No comments:
Post a Comment