web analytics
<

NBA HD: This summer’s $740K premium

With Joe Johnson receiving a max contract and Darko Milicic taking in $20 million from David Kahn, it seems as though teams are recklessly showering free agents with money this summer. It’s a sellers market; teams are flush with cash and promising the world to their fans.

But what do we mean when we say that a team overpaid for a free agent?  Whether you know it or not, our minds gather bundles of basketball information (How good is this player?), transfers that to a dollar amount (What is that product worth?), and compares it to the price tag (Was it a good deal?).  The wondrous mind is able to perform this function in a matter of seconds.  But let’s try to slow it down and put it on paper.

One approach is to quantify player value on the court and then observe how much that product costs on the market.  The market prices stabilize only after several deals have been made and they change from year to year as player as new money enters the market (say, a Prokhorov arrives or salary cap threshold rises) and/or the product line changes.   The product line has never been stronger and the suit pockets have never been deeper.

There are several player metrics out there that attempt to quantify player value on a scale of wins: John Hollinger’s Estimated Wins Added (EWA), Dave Berri’s Wins Produced (WP) and Justin Kubatko’s Win Shares (WS).

The player metric I’ll use for these purposes is Kevin Pelton’s Wins Above Replacement (WARP) which applies the same framework in Baseball Prospectus’ WARP to the NBA.  To account for player value, I will use the player’s WARP for the 2010 season.

So far, the going rate this summer for one WARP is $2.23 million.  This means that in this climate, a 4 WARP player would generally command about a $9 million per annum contract.  Of course, this isn’t ironclad and as shown by Chris Duhon and Steve Blake, who both received four-year contracts after contributing sub-replacement level performance last year, this model will bend going forward.

Remember the Drew Gooden contract that people drew all sorts of insta-snark?  That measures out to be the best bang for the buck deal of the summer up to this point, along with Boozer’s deal.  The years may be long on Gooden but the Bucks got the veteran big man at a steep discount most likely because of his questionable motor and perception that a oft-traded player equals a flawed player.   If he continues to produce on the court, Gooden could be a steal at this climate.

Surprise, Surprise: Darko Milicic was one of the worst deals so far this off season.  The Timberwolves overpaid about $13 million ($2.23 x 0.8 x 4) on the fringe contributor and the signing did little to change David Kahn’s rep as a showrunner.

One shortcoming of this model, as you can see with the cases of Blake and Duhon, is that a straight $/WARP calculation can produce some wonky results off of a poor season.  I looked at adjusting the WARP input to reflect an average of the past two seasons but the going rate remained nearly unchanged ($2.23 per win vs. $2.1 per win).  With that adjustment, Duhon and Rudy Gay became the summer’s worst deals.

Another assumption that this model makes is that production is constant.  Joe Johnson’s contract doesn’t look nearly as bad as it would if we considered his career arc and likely depreciation.  It’s the length that’s egregious; a two-year, $20 million is a much better deal than six-years , $120 million.

So is $2.23m/win an inflated price?  Compared to last year’s free agency, yes.  In fact, teams are paying about a $740,000 premium per win this offseason compared to last summer.  Using the same system for last year’s free agency, teams paid $1.49 million for each WARP unit in 2009.

But there’s still plenty of time for the Grand Opening excitement to calm and the price will likely slide a bit.  The other capped max contracts have yet to be handed out (Wade, LeBron, and Bosh) and their contracts will actually drive the going rate downward since they’re not paid on the free market.  The near $1 million premium may drop down to $500K or $250K by the end of summer.

But right now, players are seeing green.

Post comment as twitter logo facebook logo
Sort: Newest | Oldest

your formula breaks down for negative numbers - Duhon produced better than Blake by WARP and signed for less money - he is a better value - but the way your formula works, a player who was even better than Duhon - say with a WARP of -.001 at the same 4million er year salary would cost -$400million per win and a player who was truly attrocious - say with a WARP of -10 at the same salary would only cost -$.4million per win

there has to be a way to accurately show that Duhon's contract (both less expensive and for a better player) was less bad than Blake's

It was all a SCAM!!! Lebron was lying through his teeth!!!

The whole deal was decided many months ago between Wade, Bosh and Lebron. His story about only making a decision after waking up yesterday is a total lie and it showed on his face!!! Sooo obvious.

When Bosh went with the Heat after posting twitters which amounted to taunts, that should have tipped everybody that the fix was in.

The fact that all three players are getting less than max money is another tip off, because that doesn't happen overnight. Negotiations were aimed to achieve this arrangement and contracts had many pages of detail. It was all done well ahead of yesterday. Believe it!!!

NBA fans were just strung along in an attempt to make it look like it was a difficult decision for Lebron. It was a total, cynical SCAM!!!!

For example, adding individual defensive impact impact would drive Gooden's win impact down and his $ per win up.

Attempt to adjust WARP by non-boxscore defense- using Adjusted +/-, Synergy counterpart data, visual judgments or a combo of all and the $ per win will of course swing up and down. I assume teams or at least some teams try to incorporate this element into their valuations and decision-making.

Of course, teams aren't necessarily paying free agents for last year's production. They're paying on anticipated production for this year, which is at least slightly different for many of the players on this list.

Your analysis of the Milicic contract is braindead. WARP is a counting stat - play more minutes, accumulate more WARP. (Look at his player card on BP. In the 08-09 season, Darko played more minutes at nearly the same "win%" and produced correspondingly more WARP.) The wolves signed Darko to be their starting center. He'll play at least 2000 minutes next year, twice as many as in 09 and thus we would expect him to contribute around twice as many WARP - 3.4. At $1.5M/WARP (forget the premium) that means he should be paid $5.1M, and - gasp - the average annual salary on his deal is $5M! Oh noes! David Kahn is such an idiot!

Trackbacks

  1. [...] heavily on the mid-level exception to bring in talent. Today Tom Haberstroh happened to look at the current value of a win on the free agent market, using WARP and the contracts signed thus far. His findings: the current cost of a win was $2.23 [...]

  2. [...] is a subject that has been tackled before. Tom Haberstroh- now at ESPN- took on the issue at this very site. While Haberstroh used Basketball Prospectus’ WARP system, Forbes’ list employs David Berri’s [...]