Looking Back At The 2004 Free Agent Class

(I realize this is very similar to what Jon wrote at Dodger Thoughts today, but, really, I started this yesterday.)

Since I'm back on a Paul DePodesta kick, I wanted to take a look back at his controversial signings in the 2004 off season. I'll do this by looking at the 56 free agents who signed two year or longer contracts in that off season.

First, we'll sort these players by WARP1 (Unadjusted Wins Above Replacement Player), just to show what players have been most productive since 2004. All salary figures are given in millions.

On the plus side, the Dodgers signed three of the seven most productive players from the 2004 off season. I'm pretty surprised that J.D. Drew is ranked as high as he is, considering he missed half a season. Another shocker was Derek Lowe not just being the best pitcher for the money, he's been the best pitcher over the last two seasons, period out of this free agent class. Not bad for a guy who's signing was almost universally panned. The other big surprise is that Beltre ranked 12th in overall performance. Granted, he should be higher since he's made the second most money out of the 2004 free agent class, but he's not the complete disaster that he was perceived to be.

On the down side, the Dodgers also signed two of the least productive players of this class, Odalis Perez, and Ricky Ledee. Ledee's failure isn't a big thing, he struggled through injuries and wasn't very highly paid anyway. Even though he had a higher dollar per win ratio than Gagne, I doubt anyone would say his contract was worse. That leaves Odalis Perez as the big disaster from this free agent period for the Dodgers. While I was overjoyed to hear that the Dodgers resigned Perez, his performance took a nose dive after 2004. If you have to take a lesson from this, it's that if you want to criticize Paul DePodesta, blame him for signing Odalis Perez, not J.D. Drew.

Now, we'll sort the players by wins per dollar. Before I do, keep a couple things in mind. First, wins per dollar is not a linear function. That is, a ten win season is far more than twice as valuable than a five win season. This is because a ten win player can't just be replaced by two five win players, unless you pair the ten win player with a replacement player. Since the amount of wins you can put on a team is limited by the number of players that can be on the field, it would take some kind of exponential function to describe what a good deal really is. Maybe I'll answer that question someday, but for now, I'll stick with the linear function.

To make this process a little fairer, I've cut out any player that was worth less than six wins who haven't made more than 10 million dollars. This keeps me from reaching the conclusion that Henry Blanco was one of the best signings of the off-season, while leaving the truly bad contracts intact. Another thing to keep in mind is that a lot of these players contracts are either back loaded, Carlos Delgado and Richie Sexson being the prime examples, or front loaded, like Adrian Beltre's. These can make the deals look better or worse than they really are. With these adjustments in mind, we get the following chart:

Of these remaining players, the average cost is 1.97 million per win. But, as I said before, due to the exponential nature of wins, this doesn't work out properly. No one would argue that Carlos Beltran signed a bad contract because he averaged 2.3 million per win. The ten wins he collected are worth it at any cost. Conversely, it makes disastrous contracts like the one handed to Magglio Ordonez even worse, since finding a guy that can collect six wins in two years shouldn't cost all that much.

I think the lesson here is that if you're looking for free agents to rebuild your team, stop. If you consider 10 wins to be MVP type numbers, then only 10 of the 56 free agents signed in 2004 pulled this off, in the last two years combined. If you need a warm body to fill a position and you don't have anyone capable in the farm system, the free agent market is great, since five or six win players can be found relatively cheaply. If you're looking for a savior, then you are just asking to get burned.

As a bonus, by looking at these figures, these were the three best long term contracts of the 2004 off season, in my opinion:

3. Craig Counsell - Signed to a "filling out the bench" type contract, Counsell proved to be more worthwhile than Adrian Beltre for a fifteenth of the cost. Counsell served as an above average middle infielder for 1.5 million dollars a year.

2. Jeff Kent - The third most valuable player out of the players signed in the 2004 off season, Kent only trailed Carlos Beltran, who was much more highly paid, and Richie Sexson, whose contract is heavily back loaded, in wins. All this for the reasonable cost of 17 million dollars.

1. Jermaine Dye - Jermaine Dye got a smaller contract that Juan Encarnacion, think about that. In return, Dye hit 75 home runs over two seasons and put together an MVP type year this year, hitting .315/.385/.622 while playing his best defense since he was in Kansas City.

And the worst:

3. Odalis Perez - While Odalis struggled with injury in 2005, he was just plain bad in 2006. While his DIPS ERA with Kansas City, (4.33, which would be sub four when converted to the NL) suggests he could come back and be productive, Odalis just isn't getting results. His massively back loaded contract could make this look even worse if he doesn't have that bounce back year.

2. Carl Pavano - While I try not to punish someone for injuries (like the ones that afflicted Troy Percival) Pavano's avoiding pitching at all costs has to be noted. But, at least he wasn't hurting the team, unlike...

1. Russ Ortiz - This is the one signing the sabermetricians can look back on and say, yep saw that one coming all the way. Ortiz's terrible walk rates, mediocre strikeout totals, and high wins made him bait for the few G.M's who don't like "all that Moneyball stuff". That he collapsed so dramatically just makes it icing on the cake.