Any thorough evaluation of baseball performance requires some sort of abstract “replacement level.” The idea is simple: a player’s value is not measured by how much he produces, but by how much more he produces than the guy who would replace him in the lineup.
There are plenty of competing versions of “replacement” floating around: most notably, “replacement player” is the “RP” in WARP and VORP. Our stats section provides a different take on things, using a baseline of “bench” instead of replacement level.
The difference is a big one. A replacement player, in theory, is the guy who gets called up from Triple-A, or acquired off of waivers, to fill an open roster spot. A bench player, however, is the guy already on your team who steps into the starting lineup. Thus, a bench player is (or should be, anyway) better than a replacement level—we’ve seen millions of dollars spent this offseason on “bench” players such as Craig Counsell and Toby Hall.
While these abstractions are tremendously useful—they give us a way to compare production across positions, for one thing—they remain just that: abstractions. To give us a better sense of just what these numbers mean, I used Win Shares Above Bench (WSAB) to find five position players who epitomized bench-level production in 2006.
Brian Schneider has fought injuries, and to some extent, the injuries have won. He appeared to be poised for a breakout after a partial season in 2002, but he settled in eventually as a league-average producer at his position.
Last year, he took a massive step back, finishing at .256/.320/.329. So, while he amassed nine Win Shares for the Nationals, that’s exactly bench level for someone who played as much as he did. As you might expect, the baseline for catchers is low compared to that of other positions; despite that, several starters—Jason Varitek, Rod Barajas, and Yadier Molina—didn’t beat it.
My first reaction upon seeing that several established starting catchers were apparently bench-level players was that the baseline is set too high. On the other hand, look through the list of catchers: along with these four starters at 0 WSAB, there are a whole lot of backup catchers who are, in fact, performing at bench level.
Win Shares may underrate Molina’s defensive contributions, and it’s certainly true that Schneider and Varitek had off years. Both catchers accumulated at least 17 WSAB in the previous two seasons, and could well return to form in 2007.
The Second Baseman
Craig Biggio may be a Hall of Famer someday, but last year, he should have been on the bench. As he climbs toward 3,000 hits, his rate stats keep falling; in 2006, he dropped all the way to .246/.306/.422, a far cry from his career averages of .283/.367/.436. His 2006 numbers translated into exactly 0 WSAB.
As we’ve already seen, some bench-like seasons are the result of long slumps; Biggio’s was the next step in a predictable decline. While Houston’s dedication to Biggio is admirable, slotting him in as a starter until he reaches 3,000 hits is equivalent to giving the job to, say, Alex Cora. With worse defense.
Win Shares Above Bench may give Jack Wilson a short shrift because it discounts his defensive contributions; on the other hand, maybe Wilson just isn’t any better than your typical backup shortstop. Wilson is the best-compensated of the players on this list, and he got more playing time than anyone discussed here except for Biggio.
At least for the last two years, Wilson is a very good approximation of the bench-player shortstop: in addition to his 0 WSAB for 2006, he accumulated only 2 WSAB in 2005. Last year, he accomplished that with a .273/.316/.370; the year before, Win Shares gave him more credit for defense, but he hit even worse: .257/.299/.363.
A slightly more optimistic reading of Wilson’s glovework would make him better than a theoretical backup. However, he’d have to be Ozzie Smith‘s equal to be worth his nearly $7 million per year.
The Center Fielder
Some of these guys just had bad years; Aaron Rowand had an excuse. When he ran into a wall in May, his OPS was well above .800; by the end of the year, it had fallen to .745. Not only that, but his defense may have faltered as well; Win Shares certainly thinks so.
Over the entire season, Rowand hit .262/.321/.425, which with above-average defense, equals a bench-level center fielder. Like Schneider and Varitek, Rowand was coming off a pair of seasons that combined for nearly 20 WSAB, so he’ll probably return to form as a better-than-bench player in 2007.
The Corner Guy
It seems a little unfair to call out Jeremy Hermida as a bench-level performer; there are plenty of highly touted 23-year-olds who haven’t yet accomplished as much. It’s a useful reminder that even bench-level production has a lot of value to major league teams. The teams that manage their budgets the most effectively are often the ones that use internal (read: cheap) options as bench players rather than spending a couple million per bat on the free agent market.
Hermida hit .251/.332/.368 last year, managing a very weak slugging percentage but the best on-base percentage among those players I’ve discussed here. Like Rowand and Schneider, Hermida’s bench-like performance may be more representative of his injuries than his skills. PECOTA thinks that Hermida is still on track to become a star, forecasting him at .283/.378/.479 for 2007.
All Together Now
To provide some familiar benchmarks for bench-level performance, you’ve surely noticed that I picked players who aren’t often thought of as bench guys. Of these players, only Wilson has a bench-player talent level, and at least of couple of them (Hermida, especially) ought to perform much better in 2007.
Many of the guys who we more commonly think of as bench players also amassed 0 WSAB in 2006: among dozens of others, Joe Borchard and Sal Fasano did it. In 2007, many actual bench players will manage positive WSAB numbers, while plenty of starters will end up exactly at bench level. It’s that variance that keeps baseball interesting from April to October, no matter how aggressively we pick it apart from November to March.