Playoff Preview: Roster Construction in Oakland and Los Angeles

With the elimination of the Rangers, only two teams from the West divisions remain contenders for the World Series. Here at Midnight Baseball, we will always root for an AL West-NL West championship, even if we have to put aside a decades-long resentment to do so. Some years will offer us more than one rooting option, but any all-West World Series will ultimately fight East Coast Bias. So, here’s our postseason bracket, as quick and meaningless as most other predictions out there:

Oakland over Detroit, Tampa Bay over Boston; Los Angeles over Atlanta, St. Louis over Pittsburgh.

Oakland over Tampa Bay; Los Angeles over St. Louis.

Oakland over Los Angeles.

(If I’m being unbiased, I’ll take St. Louis over LA and Tampa Bay over Oakland. Tampa Bay wins.)

Homerism aside, an Oakland-LA match-up is intriguing for the easy contrast it offers. Of course I’m going to talk about the largest payroll in the game against the Moneyball pioneers. Both team-building strategies succeeded this year. With the Dodgers, it’s easy to see how. They paid and paid for good veteran players, with a few key assists from young homegrown players. (Yasiel Puig counts as both.) The A’s used their limited budget to stack the back end of their roster with depth. Here’s my attempt to visualize those strategies.

oakla16war

Both teams had 16 players with at least 100 plate appearances. Above I’ve ranked them from most to 16th most, going from top to bottom. The horizontal bars represent that player’s WAR on the season. The color of the bars indicates the team, yellow for Oakland and blue for Los Angeles.

You probably see Josh Donaldson’s stellar performance first of all. You’ll also notice the strong performances from eight of the Dodgers’ top nine players. Really, though, you can classify the ninth player, Hanley Ramirez, in the top eight. The only reason he was ninth on the Dodgers in plate appearances was injury; had he stayed healthy, he’d be a regular, somewhere in the top eight, and with an even more impressive WAR. It’s in the bottom half of the graph that the A’s found an edge. Only one player, Adam Rosales, who is no longer with the team, performed at or below replacement level, compared to four on the Dodgers. Below there’s a table that chunks players into groups of four: those whose number of PAs ranked 1-4 on their team, then players ranked 5-8, 9-12 and 13-16. In the table I’ve made the Hanley correction, switching Hanley with Skip Schumaker, the actual No. 8.

oaklawartable

Here the strength of the Oakland bench is a bit more obvious. The top eight on the Dodgers performed up to their contracts, fueling the team’s success. But you’ll see players 9-16 get about the same number of plate appearances on both teams. In other words, due to some combination of injury, rest and manager disposition, both teams’ benches have gotten the same amount of playing time. For the Dodgers, all that playing time has resulted in -0.3 WAR; for the A’s, 5.2. Holy cow!

It becomes even more jarring when we turn WAR into a rate stat. I put WAR/600 in bold because I was planning to give it a whole paragraph. Here goes: WAR/600 is simply what a given player’s WAR would be if he had gotten 600 plate appearances, which is about the average for an everyday player. Here I calculated the WAR/600 for the entire four-man units, to give you a rate-adjusted sense of the quality of the A’s bench. What the results say is that if you made the players at the end of the A’s bench into one everyday player, that player would accumulate either 2.0 or 1.4 WAR, depending on how far down the bench you go (chunk 9-12 and 13-16 on the chart, respectively). A WAR of 2.0 is considered an average major league starter, so an easy conclusion would be to say that the A’s have a bench full of average to slightly-below-average major league starters and that’s why they’re so good.

It isn’t that simple, of course, because selectivity is part of the reason those bench players have performed so well in their limited appearances. The A’s platoon frequently, giving their batters more playing time against opposite-handed pitchers, inflating their numbers. (Conversely, they don’t give their batters the opportunity to deflate their numbers against same-handed pitchers.) Platooning will be covered tomorrow. For now, we have concrete evidence that the A’s had a bright idea about roster construction and executed that idea well.

Advertisements

2 thoughts on “Playoff Preview: Roster Construction in Oakland and Los Angeles

  1. Pingback: About Platooning in Oakland, Briefly | Midnight Baseball

  2. Pingback: The NLCS Dodgers and Robert Andino, and Other Game 3 Notes | Midnight Baseball

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s