(You've read Andy's take; here's mine.)
What might we expect from Michael Cuddyer during his 3-year, $31.5-million deal with Colorado? I approached it by looking at what others of the same age and similar value had done during the live-ball era.
For age 30-32 combined, Cuddyer produced 7.9 WAR. So I searched for outfielders from 1920-2008 who produced from 7.0 to 8.8 WAR combined for ages 30-32.
For age 33-35 combined, those 47 players had a median of 1.9 WAR. Uh-oh....
But perhaps Cuddyer's 2011 season, with a career-high value of 3.0 WAR at age 32, is a fairer starting point. So I looked at all outfielders from 1920-2008 who produced between 2.5 and 3.7 WAR at age 32. (I used 3.7 instead of 3.5 as the high end in order to get the median of that group up to Cuddyer's 3.0.)
For age 33-35 combined, those 54 players produced a median of 3.1 WAR. Hmmm....
But Cuddyer also has the ability to play 1B and 3B (never mind how well he plays them), which may indicate a better aging pattern. How do the comparisons look if we include corner infielders as well as OFs, with the same WAR range as before?
The 3-year-base group, for age 33-35 combined, had a median of 2.4 WAR.
11% produced at least 10 WAR; 17% produced at least 8 WAR; and 28% produced at least 6 WAR.
The 1-year-base group, for age 33-35 combined, had a median of 2.9 WAR.
10% produced at least 10 WAR; 23% produced at least 8 WAR; and 28% produced at least 6 WAR.
From a purely financial standpoint, it seems that Cuddyer needs to produce at least 6 WAR to make Colorado's investment worthwhile. These historical comparisons suggest there's less than a 1 in 3 chance of that happening.
Now here's the best case I can make for Cuddyer. Last year, he had 2 other markers that may suggest a better aging pattern: He had good baserunning data (11 steals in 12 tries, and a 56% rate of extra bases taken as a baserunner), and he started 17 games at 2B. So I'll run a comparison group of 32-year-olds with 10 to 20 steals and at least 2.5 WAR, with no position requirement. (That group's medians were 4.3 WAR at age 32 and 10.9 WAR for 30-32 combined, both well above Cuddyer's 3.0 and 7.9; but we're looking for a best-case study here.)
For age 33-35 combined, that group produced a median of 6.9 WAR. If Cuddyer does that over the next 3 years, he'll have just about earned his dough. But it's a slippery slope; the bottom 1/3 of that group produced less than 3 WAR for age 33-35 combined. No matter how you slice it, investing in the mid-30s seasons of a non-star is a risky proposition.
Finally ... If you're a Rockies fan and don't want any more agita about this deal, avoid the list of players statistically most similar to Cuddyer through age 32. Of the nine who are retired, six didn't even play 100 more games from age 33 to the end of their career, and the group averaged just 179 games. Only one of them produced any good seasons from age 33 onward.