SG John Salmons was outstanding in his short post All-Star break stint with the Milwaukee Bucks. He posted an MWS48 of +1.57, and won an estimated 3.6 games for the team while losing only 1.0 games. But if he opts out of his agreement and becomes a free agent, how much should the team invest in him?
I would be careful about investing very much. There is every sign that Salmons Milwaukee stint was an aberration. His high MWS48 was basically caused by his outstanding net effective scoring (basically his pts/scoring poss minus his opponents). His net effective scoring with the Bucks was a +3.02 per 48 minutes.
But his career average and career median net scoring are both well below that at +0.08. His 3 year average and median are also both well below his Bucks mark at +0.73 and +0.59. Indeed, for the whole of last season, both with the Bucks and Bulls, his net effective scoring was +1.65, just a bit more than half of what it was with the Bucks alone. Indeed, in no other stint with any other team, whether half or full season, has he ever posted anything better than +0.75.
To underscore how deviant the scoring performance was, if you remove the Bucks stint, then his career net effective scoring average is -0.41 and his net effective career scoring median is -0.44. His 3 year net effective scoring average would be +0.10, and his 3 year net effective scoring median would be +0.45, both far, far lower than his aforementioned Bucks mark.
So if you sign Salmons to a deal based upon the expectation that he will be the late 2010 Salmons for the course of the agreement, you would be banking on him, at the age of 31, replicating over time a performance that looks to be beyond even his single full season capabilities.
I would make an offer that anticipated a performance next season that is much closer to his 3 year marks, and declining from there because of age. For next season I would think the Bucks could expect him to produce just above average wins. I would project his expected MWS48 for next season at around +0.23. After that I think he will slowly slink down to bench level. Not bad, considering the Bucks have nothing else at the position, but still the Bucks shouldn’t offer him the moon. That’s bad value for money and a small market team, especially one with the Bucks woeful shopping record, can’t have that.
Here are Salmons recent and career efficiency numbers from which I made that MWS48 estimate:
Salmons Career Weighted MWS 48: (-0.33)
Salmons 3 year Weighted MWS48: (+0.05)
Salmons 3 year unWeighted MWS48: (+0.09)
Salmons 3 year median MWS48: (+0.53)
Salmons 3 year “category median” MWS48: (-0.03)
Salmons 3 year “Ceiling” MWS48 (using full seasons): (+1.26)
Salmons 3 year “Floor” MWS48 (using full seasons): (-0.79)
Salmons 3 year “Midground” MWS48 (using full seasons): (+0.23)
So as you can see, in his short stint with the Bucks Salmons exceeded even his projected “Ceiling” production (if you don’t understand these terms please scroll to the very next post, I don’t have time to reexplain, basically the “ceiling” is if he produced his best numbers in all 3 categories in one full season, with the categories being: Net Scoring, Net Possessions, and Net Helpful Stats).
Thus a regression to the mean has to be anticipated, and would be shocking if it did not occur.
So based on his age, his resume, and other factors, if I were the Bucks I would make him a decent offer, but certainly nothing extravagant. The team needs another overvalued contract albatross hanging on their collar like they need a hole in their heads.
By the way, if they do sign Salmons, here is what I would project:
Based on last three seasons, I would have him playing 2799 minutes, which if you figure the NBA average of 241 minutes per game (*you have to consider overtime), that is roughly 11.6 game responsibilities.
At a projected +0.23 Marginal Win Score, that equates to: