"It isn't the high price of stars that is expensive, it's the high price of mediocrity," former Chicago White Sox owner Bill Veeck once famously observed. I've borrowed this quote, which I remember learning from the Baseball Prospectus annual, in the past. More and more all the time, however, I'm not certain Veeck's quote really sums up the issue. It's not even the high cost of mediocrity that kills teams--it's the duration of it.
This jumped out at me over the course of compiling my position-by-position rankings of free agents. Sprinkled throughout these lists are players who have long since ceased to have any value to their team and have hung around solely because of their contract. Now, this is not any sort of criticism of the guaranteed contract in general or player motivation. Often, like Antonio Daniels or the more extreme case of Cuttino Mobley, injury is a culprit. Sometimes, the player just aged. And occasionally, as in the case of Trenton Hassell, the contract was simply a mistake from day one.
Instead, I see the takeaway as how teams can get themselves in trouble with the lengths of the contracts they offer as much as the yearly salary. To try to gain some insight into this phenomenon, I went through and looked up all the contracts of five years or longer that were signed in free agency (no extensions) and concluded at the end of last season.
I found 23 such contracts, with the biggest (Ray Allen for $80 million over five years) and smallest (Damien Wilkins for $14.5 million over five years) both handed out by the Seattle SuperSonics during the summer of 2005. The majority of the deals (13) were for five years, with several six-year contracts and a couple of seven-year ones signed during the 2003 offseason by Brad Miller and Kenny Thomas rounding out the group.
Even without going to the numbers, it's obvious that teams came to regret many of the long-term offers they handed out. Of the 23 players, just four spent the duration of their contract with the team that signed them, and even one of those comes with an asterisk (Zydrunas Ilgauskas, who was bought out of his contract and returned to Cleveland after being traded to Washington). While not all of the decisions were made because of performance, nearly half the 23 players (11) were either waived or bought out before their contracts ended. And five players (Daniels, Mark Blount, Jerome James, Darius Miles and Cuttino Mobley) did not see a single minute of action in 2009-10.
The summary statistics reinforce this conclusion. I looked at both average minutes played by year and average WARP totals to provide two different perspectives. Note that I only looked at the first five years, so for Miller and Thomas, for example, this ignores their performance the last two seasons so the pool of players remains consistent.
Stat Yr1 Yr2 Yr3 Yr4 Yr5
Minutes 2183 1859 1985 1508 1170
WARP 3.7 1.9 2.6 1.5 1.4
The first three years, players averaged about starters' minutes (26-30 minutes a night) and posted WARP totals around league average. Things went downhill the next two seasons, when free agents' minutes plummeted and their WARP totals dropped. Exacerbating the problem is the fact that most NBA contracts escalate year over year, so these players were making progressively more money for less production.
One other way to break down the numbers is by the annual salary, separating the star players like Allen from the role players. Are long-term contracts more of an issue for one group or the other? I came up with four groups--star players (at least $10 million on average over the life of the contract), those above the mid-level exception ($7 million-$10 million), the mid-level exception ($5.7 million to $6.3 million) and everything below. Here's how they compare in the same breakdown.
Type Yr1 Yr2 Yr3 Yr4 Yr5
Stars 2340 2067 2555 2414 1871
AboveMLE 1638 1842 1189 898 1061
MLE 1615 1486 1479 1050 846
BelowMLE 2366 2438 2297 1626 1252
Type Yr1 Yr2 Yr3 Yr4 Yr5
Stars 7.4 5.0 6.7 6.9 3.3
AboveMLE 4.5 1.4 2.2 0.2 1.4
MLE 1.1 0.6 0.8 -0.6 0.3
BelowMLE 2.6 1.3 1.4 0.5 0.9
The sample sizes are small here (no more than seven players in any group), so drawing any strong conclusions would be dangerous. Still, what we see generally confirms conventional wisdom. The star players remained good values through year four before finally dropping in year five. Everyone else dropped considerably after year three.
Two sidenotes. First, players in this group signed to mid-level contracts were uniformly awful. This ties in nicely to my breakdown two summers ago of how poorly mid-level deals have failed, which I speculated might be partly due to the length of the contracts handed out to them. Surprisingly, players on smaller contracts actually faired quite well, with Rafer Alston, Raja Bell and Kyle Korver all proving bargains among this group.
Looking over all the data, it appears the sweet spot for free-agent contracts is about three years. Past this, the risk escalates quickly. Predicting performance next season is hard enough, so trying to figure out what a player will be like five years down the line (or even whether they'll be healthy then) is nigh impossible. For stars, that risk is an acceptable part of acquiring an elite talent. Among the NBA's middle class, it's difficult to justify going much more than three years. A four-year deal can be justified because the expiring contract becomes valuable as a trade asset the last year, but five-year deals to marginal talents simply aren't worth it.
Looking at the contracts handed out so far this summer, several fall into this category. Phoenix's Channing Frye, Milwaukee's Drew Gooden, Denver's Al Harrington (only fully guaranteed for the first three years), Toronto's Amir Johnson, Portland's Wesley Matthews (an offer sheet the Jazz reportedly will not match), Miami's Mike Miller and New Jersey's Travis Outlaw all got five-year deals right around the mid-level exception. Meanwhile, even if Brendan Haywood's sixth year only becomes guaranteed based on conditional performance, his contract is both too long and too lucrative.
By next summer, six-year contracts may be a thing of the past. The last CBA limited teams to six years when re-signing their own free agents or five years for other teams' players, down from seven and six, respectively. The league will surely look to cut down contract lengths again, and they'll probably use data like this. But there's already a simpler answer for intelligent teams: Don't make long offers and let other teams deal with the painful consequences.
Kevin Pelton is an author of Basketball Prospectus.
You can contact Kevin by clicking here or click here to see Kevin's other articles.