James Click has written a very interesting column (subscription only, sadly) for Baseball Prospectus trying to nail down a better idea of "replacement player." The problem is this: for stats such as VORP (Value Over Replacement Player), each player is measured against a theoretical "replacement player"--that is, the type of guy freely available, AAA free agents and the like who move around a lot and can be had in mid-season if the need arises.
However, when you're making decisions for a specific team and replacing specific players, it doesn't make sense to use theoretical numbers. For instance, if we decide to replace the Branyan/Cirillo/Helms/Hall mishmash with Billy Hall full-time in 2006 and Billy puts up a VORP of 25 runs, does that mean the Brewers are 25 runs better? Heck, no. It depends on how much the 2005 mishmash was worth.
The point of all this is, depending on where a team's strengths lie, the value of upgrading at certain positions differs greatly. To some degree, this is obvious: it'd be pretty silly for the Crew to sign Rondell White, whereas a very good backup catcher would have a significant impact.
Click came up with the "replacement level" for batters, the rotation, and the bullpen by measuring how well the team's replacements (bench players, #6+ starters, and 7th+ bullpen guys) performed last year. Surprise: the Brewers had the highest replacement level for batters of any team in MLB, thanks in part to that killer third base platoon. In other words, the Crew had in-house replacements who were, on average, superior to those of any other team in baseball.
In the rotation, Milwaukee's replacement level was 8th in all of baseball--this time mostly due to Rick Helling's excellent 7 starts, which overcame Obermueller's and Glover's not-so-excellent contributions.
The bullpen is another story. The Brewers ranked 25th in replacement level there--replacement relievers for the Crew pitched below league replacement level. Theoretically, that means that if Melvin had picked the best guys off the waiver wire beyond the 6 regulars (presumably Turnbow, Santana, Wise, Bottalico, DLR, and Obie), the bullpen would've been better. I'd like to see exactly how he came up with that, because many of the other relievers on the team (Lehr, Kane Davis, Capellan, Adams) were well above league-average) but I suppose the relief contributions of Santos, Obie, and Glover were particularly bad.
Long story short, that means that--among free agent signings and trades--good bullpen upgrades have the potential to be the most valuable additions to the 2006 Brewers. It appears that Doug Melvin knows this--he's said he's on the market for bullpen help since the beginning of the offseason, and is still looking into players like Joe Borowski and Shigetoshi Hasegawa, both of whom became available at yesterday's non-tender deadline. If the Brewers are going to spend more money, it appears--according to Click, anyway--that the bullpen is where it can be best spent.