If the Phillies have a Terrell Owens equivalent, it's Billy Wagner. The diminutive closer has a mouth matched only by the speed of his fastball. What's worse, the Philadelphia media eats up Wagner's proclamations as if he's preaching gospel: earlier this year, the closer's split-personality statements first praising the team for its character, then forecasting doom for the team because of its character the next month, and then again praising its scrappiness days later wasted column inches by the dozen. It seems as if Wagner can't help himself.
Now, he's making headlines because of his contract situation. Wagner will be a free agent at the end of this season if the Phillies don't sign him to a contract extension. Although we have to be skeptical of whether he actually means it (see, e.g., his springtime claims that he would retire at the end of this season), he's publicly set a deadline: if there's no agreement by the end of August, Wagner has pledged to test the free agent waters this winter.
Press reports suggest that the two sides are not very close in contract negotiations: Wagner reportedly wants $24 million over the next three years with a full no-trade clause, while the Phils have offered two years at $16 million total, without the clause. Wagner has dismissed the Phils' offer as insulting, saying that he's having the best season of his career and that he won't accept a pay cut (although, curiously, the contract that he is reportedly seeking is for $8 million per year, the same rate the Phils are offering and lower than the $9 million he's making this season).
But we're not here today to quibble over Wagner's poor math skills. Instead, let's examine the question of whether and for how long the Phils should re-sign Wagner. (I'm not going to be addressing the "for how much" question here.)
We start with a look at historically comparable players, those who have had careers similar to Wagner. Because baseball has such a long history with so many players, it is often easy to find other similar players, the historical comparables, to see how a player?s future may play out. Of course, nothing?s set in stone, but past performance by comparable players may help predict future performance by the player you?re curious about.
One way to determine historical comparables is by using Bill James? similarity scores, described here by Baseball Reference. We can also break down similarity scores based on that player?s age, hopefully providing some insight into the future value of a contract as the player we?re investigating ages.
For Billy Wagner, we can look at his Baseball Reference page to see his historical comparables. Here?s the list of similar players at age 32 (Wagner?s age for his last complete season, the 2004 season). The number in parentheses indicates how similar these players are to Wagner, with 1000 being the highest possible. Thus, Mariano Rivera, at age 32, is the best comparable to Wagner at age 32; former Phil Al Holland at age 32 (the year he was traded from the Phils to the Pirates to the Angels) is the 10th best.
Mariano Rivera (935) Troy Percival (934) Trevor Hoffman (930) Jeff Montgomery (919) Randy Myers (919) John Wetteland (909) Tom Henke (908) John Franco (904) Joe Sambito (898) Al Holland (897)
The relief pitchers listed here as Wagner?s top comparables break down pretty nicely into two very different career trajectories that can be seen by looking at the top two comparables (separated in similarity score by just 1 point). Mariano Rivera followed his age 32 season in 2002 with continued excellence: a 1.66 ERA in 2003; 1.94 ERA in 2004; and so far an incredible 0.92 ERA in 2005. On the other hand, Troy Percival?s career path raises serious red flags: 3.47 ERA in 2003; 2.90 in 2004; and an awful 5.76 ERA in 2005. Percival has also battled hip and elbow injuries in his career, including a dreaded July Dr. Yocum visit resulting in him being shut down for the year.
If the Phils get their way, they?d sign Wagner for his age 34 and 35 seasons; if the Phils accede to Wagner, they?d sign him from age 34 through 36 with no possibility of trading him. So, it?s useful to look at what these comparable players have done at those ages, as the following charts show. (In the charts, ERA+ is the player?s ERA compared to the league ERA that year; R indicates the player was retired that year; N/A indicates the information is not available yet because the player is still active.)
Age 34 ERA ERA+ Mariano Rivera (935) 1.94 231 Troy Percival (934) 2.90 158 Trevor Hoffman (930) 2.73 140 Jeff Montgomery (919) 4.26 115 Randy Myers (919) 1.51 291 John Wetteland (909) R R Tom Henke (908) 2.26 181 John Franco (904) 2.44 172 Joe Sambito (898) 4.84 87 Al Holland (897) 14.21 31
At age 34, this group was still doing reasonably well: six of the pitchers were still dominant, with ERAs under 3 and ERA+s of 140 or over. Two were mediocre, only one had retired (Wetteland), and only one was awful (Holland).
Age 35 ERA ERA+ Mariano Rivera (935) 0.92 N/A Troy Percival (934) 5.76 N/A Trevor Hoffman (930) 2.00 197 Jeff Montgomery (919) 3.49 135 Randy Myers (919) 4.92 90 John Wetteland (909) R R Tom Henke (908) 2.91 143 John Franco (904) 1.83 214 Joe Sambito (898) 6.93 66 Al Holland (897) R R
By age 35, though, advanced age was setting in. Rivera (at age 35 this year) has so far continued his dominant ways, with perhaps his best season yet (which says a ton). Hoffman, Henke, and Franco all had strong seasons as well. However, the rest were either mediocre (Montgomery), had poor to atrocious seasons (Percival, Myers, and Sambito), or were retired (Wetteland and Holland).
36 ERA ERA+ Mariano Rivera (935) N/A N/A Troy Percival (934) N/A N/A Trevor Hoffman (930) 2.30 174 Jeff Montgomery (919) 4.98 99 Randy Myers (919) R R John Wetteland (909) R R Tom Henke (908) 3.79 127 John Franco (904) 2.55 158 Joe Sambito (898) R R Al Holland (897) R R
At age 36, we see lots of trouble. Only Hoffman had a truly dominant age 36 season. Franco and Henke remained very effective, and Montgomery was almost exactly league average. But the other four--half of the comparable players who have already reached age 36--had retired. In other words, the data shows what you?d expect: overall, the players declined with age and started to head for the golf course.
The age 36 comparables are not good news for the Phillies if they?re considering taking on Wagner for his requested 3 years, and the results at age 35 are not encouraging either. It appears that it would be a significant risk for the Phils to commit big money to Wagner for his age 35 and 36 seasons, especially given his injury history (although not as significant as Percival?s, he hasn?t had as clean a bill of health as Rivera either).
There is one missing piece of this puzzle, however, and it argues to the contrary. Take a look at this chart, which I conspicuously left out from the charts above:
Age 33 ERA ERA+ Mariano Rivera (935) 1.66 265 Troy Percival (934) 3.47 123 Trevor Hoffman (930) 3.43 117 Jeff Montgomery (919) 3.43 142 Randy Myers (919) 3.53 139 John Wetteland (909) 4.20 122 Tom Henke (908) 2.32 182 John Franco (904) 2.70 155 Joe Sambito (898) 12.66 27 Al Holland (897) 5.09 81
This year, 2005, is Wagner?s age 33 year. So far, he has had an amazing season, with a 1.86 ERA through mid-August. His season, if he keeps it up, would increase his similarity to Rivera, Henke, and Franco--the three comparables who had the most dominant age 33 seasons. Interestingly, those three also were three of the four on Wagner?s list of comparables who had excellent seasons for ages 34 through 36 (Hoffman being the other).
Thus, if we take a smaller sample of comparables and include Wagner?s performance this year, signing him to a three-year contract may not be such a bad idea after all. The ever-present risks of age and injury are always yellow flags, but if he can perform like Rivera, Henke, or Franco through age 36, he will make some team, possibly the Phillies, happy for the next three years.
Of course, this is all without saying how much any team should pay a guy who comes in just to record the last few outs of a game, but that?s another discussion for another day.