Freelance Friday is a project that lets us share our platform with the multitude of talented writers and basketball analysts who aren’t part of our regular staff of contributors. As part of that series we’re proud to present this guest post from Kevin Hetrick, one of the minds behind GotBuckets, a basketball website which focuses on statistical analysis using Adjusted Plus Minus metrics. Kevin is here to walk you through a few of the metrics his site uses and break down some of this summer’s transactions looking for winners and losers. You can also find Kevin on twitter, @Hetrick46.
What a thrilling day?!?! A chance to bring our niche at GotBuckets to Nylon Calculus! While many popular, catch-all, advanced stats (PER, Win Shares, etc) are derived from the box score, GotBuckets deals in play-by-play regression derived stats, namely Regularized Adjusted Plus-Minus (RAPM), from @talkingpractice, and 2-year Adjusted Plus-Minus (2-year APM) from @laughingcavs. From those, wins equivalents are developed; we affectionately call these SWAgR (Summed Wins de-Aggregated, Regularized) and SWAg. Simply stated, RAPM and 2-year APM use NBA play-by-play data of who was on the court (independent variables) and how their teams point differential faired while they played (dependent variables), churning this information through a massive regression, and spitting out a number presenting the per possession value that each player contributes to their team’s bottom line; a bit more detail is available here.
Gotbuckets also features a similar regression, Four Factors Adjusted Plus Minus (FFAPM, from @laughingcavs), but instead of point differential, the dependent variables are effective field goal percentage, turnover rate, rebounding rate, and free throw rate; the output offers an approximation of how each player impacts their squad’s offense and defense in those areas.
RAPM and 2-year APM are an improvement on predicting future results compared to PER and Win Shares. With that as a general guide, this article delves into likely “Winners” and “Losers” from the first month of the NBA’s off-season. The article exclusively uses RAPM, along with @talkingpractice’s simplified Aging Curve —change in RAPM from season to season = (27 – Age) * 0.06. Using SWAgR as the estimation of wins that a player provides, the value of a win is based on an $80 million luxury tax; i.e. $80 million divided by 60 wins, means $1.3 million of salary or less per win is a great value. $80 million divided by 40 wins…$2 million in salary per win is an overpay. Now onto the early “Winners” and “Losers”.
Winner – Dallas Mavericks
Let’s start by acknowledging that the biggest reason the Mavericks are a Winner of Summer 2014 is that their star, the best player in franchise history, agreed to a deal that is worth far less than he warrants. While a shame that franchise icons need to take below market deals to help their team compete, for 3 years and $25 million, Dirk Nowitzki did exactly that. According to SWAgR, last season Dirk was worth over 16 wins, or 0.209 wins per 36 minutes. Assuming Dirk finally begins succumbing to normal aging, he still manages 0.194 and 0.178 SWAgR per 36 for the next two seasons; if he can roll out 4800 minutes, that amounts to nearly 25 wins for $17 million. Production at that level would be an amazing value for the Mavericks; hopefully they treat Dirk really well in retirement.
This may not be a shocker from a team that used to employ APM originator Wayne Winston as a consultant, but RAPM & SWAgR like many of Dallas’ other moves this offseason, too. Devin Harris (3 years, $9 million), Greg Smith (1 year, $1 million), Al Farouq Aminu (2 years, $2.5 million), and even Raymond Felton (2 years, $9 million) look poised to chalk up SWAgR at a cost around $1 million per win over the course of their contracts. Felton, for better and worse, is a guy with an on/off switch. RAPM traditionally likes him though, rating him better than Jose Calderon for the past six years. Felton is younger, has $13 million less in guaranteed money, and if he acclimates well to his new, presumably happier surroundings and plays with ambition, his two years could work out very nicely for the Mavericks.
Chandler Parsons is an overpay at around $2 million per forecasted win, projecting to net approximately 7 SWAgR per season over the next two years for his $14 and $15 million. Tyson Chandler also looks likely for overcompensation, but he is better than Samuel Dalembert, it is one year, and the Mavericks are under the luxury tax; no blood, no foul, right?
While accounting for aging, and estimating that Nowitzki, Parsons, Tyson Chandler, Felton, Devin Harris, Farouq-Aminu, Greg Smith and Richard Jefferson play roughly 60% of the Mavericks minutes next year, the octet projects to create over 36 wins, for $48 million. The crew of Dirk, Parsons, Felton and Harris seems reasonably likely to procure 31 wins for $26 million in 2015 – 2016; both of those projections fall very tidily into a range of $1.2 – $1.3 million per victory. Despite never landing the free agent superstar they seek, the Mavericks look pointed towards another 50-ish win season, while keeping the space to be free agent players in summer of 2015 or 2016. This franchise continues to target the right consolation players (Vince Carter, Monta Ellis, etc) to stay competitive and appears to have done it again this summer. It also helps that Dirk is way too nice.
Loser – New York Knicks
Next — the Mavericks’ trade partners this summer. A perennially underachieving unit in the nation’s biggest market; RAPM expects them to continue on that path.
Here’s the thing about Carmelo Anthony. First, his defensive RAPM has never been in the top half of NBA players. That doesn’t help his cause. Second, his offensive RAPM has never hit the top ten in any given season, peaking with the 11th and 12th best offensive RAPM in the NBA. While certainly a great & entertaining scorer, RAPM does not consider him among the truly elite talents at generating buckets for his team. Add those two ideas together, and he never surpassed 11 SWAgR in season…that’s right, according to the method used here, at his best, his on-court performance didn’t merit $20+ million per season. With a new contract taking him towards his mid-30’s for age, and mid-$20-millions for salary, it becomes even more unlikely for Melo to take the reins of a true contender.
The Knicks’ other big acquisition was Jose Calderon, 33-years old on opening night, with three years and $22 million left on his contract. RAPM considered him the 10th worst defensive player in the NBA last season, and the 30th worst the season before that. If he plays 2200 minutes per season the next two years and experiences typical declines with age, he estimates to produce only 3.5 total SWAgR for nearly $15 million. Perhaps the triangle reinvigorates him, but he seems very unlikely to generate one win per $1.5 million of salary, and the Calderon, Hardaway Jr, Carmelo, Bargnani, Stoudemire defensive unit could be a dumpster fire for history.
Winner – Toronto Raptors
By re-signing Kyle Lowry (4 years, $48 million), Greivis Vasquez and Patrick Patterson, all the Raptors did was keep together the core of a team that finished last season on a 42 win, 22 loss run, returning seven of their top eight players for minutes, with 28-year old Kyle Lowry being the oldest of the crew. The lost player? I don’t think John Salmons will be terribly missed. The success of this off-season starts with the Lowry deal, as SWAgR projects him to be worth nearly 22 wins over the next two years. Between he, Vasquez and Patterson, with an estimated 5700 minutes per season, the trio appears poised to generate approximately 21 wins for $24 million of salary in both 2014 – 2015 and 2015 – 2016. Those deals definitely leave needed flexibility to re-sign Amir Johnson, extend Jonas Valanciunas and Terrence Ross, and possibly add another talented free agent piece. Nice work, Masai Ujiri.
Loser – Los Angeles Lakers
It’s not all bad for the Lakers. Ed Davis at $1 million per season is a nice deal; with his forecasted RAPM, he should provide 0.8 SWAgR in only 600 minutes of play (for $1.3 million salary / win). If Swaggy P (4 years, $21 million) plays a similar number of minutes as he did last season, he should produce 6.5 wins over the next two years; $1.6 million per SWAgR can build a 50-win team. Jeremy Lin is a solid point guard with only one year remaining on his contract, and he brought the Rockets first round draft pick with him.
Ultimately, none of these guys are game changers though, as the Lakers struck out on the big names. Carlos Boozer and Jordan Hill ranked in the bottom ten percent of NBA players for RAPM last year. Actually, Hill sinks to that level in three of the last four years, generating negative SWAgR over that entire timeframe. And the Lakers are paying him $9 million next season? Is that a misprint? Kobe’s career does not look destined to end similarly to Tim Duncan or Dirk Nowitzki’s, playing on competitive, fun teams.
Winner – Orlando Magic
Let’s ignore Ben Gordon’s expiring contract and instead focus on the two guys that will still be around in 2015 – 2016, when Orlando likely hopes to start being fun and competitive. Assuming the then 24-year old Evan Fournier and 32-year Channing Frye are each good for 2000 minutes, the young Frenchman appears capable of producing nearly six wins, as per SWAgR. All for a paltry rookie contract price of $2.28 million. Frye projects as offering seven wins, for the relatively low salary of $8 million. Almost thirteen wins? Only $10 million in salary? That sounds like some building blocks for a young, fun team in two seasons.
Loser – Detroit Pistons
There are some teams that obviously employ regression based metrics…and some that clearly do not. Detroit falls in the latter group. Or maybe Stan Van Gundy is looking at his charts upside down. Or he just doesn’t care about this stuff. Caron Butler (1 year, $4.5 million), D.J. Augustin (2 years, $6 million), and Aaron Gray (2 years, minimum) all finished in the bottom 15 percent of NBA players according to RAPM, each around replacement level or worse. Augustin rated as the 15th worst defender in the NBA last season and 6th worst in 2012 – 2013; Butler turns 35 this season.
Jodie Meeks is a nice enough player, finishing in the top half of NBA players for RAPM each of the last four seasons and producing a remarkably consistent 3 – 3.5 SWAgR each of those years, but turning 27 this summer, it seems unlikely that he improves greatly on that. At three years and $19 million, he is likely to offer one win per $1.5 – $2 million of salary.
Winner – Phoenix Suns
Phoenix is definitely a team that employs regression based metrics. There were 482 guys that played in the NBA last year; Isaiah Thomas (4 years, $27 million) was 55th for per possession RAPM and 54th for cumulative SWAgR. P.J. Tucker (4 years, $22 million) and Anthony Tolliver (2 years, $6 million) both finished in the top 25th percentile according to RAPM and SWAgR. Assuming a modest 5000 minutes between the trio, they project to provide nearly 14 wins each in 2014 – 2015 and 2015 – 2016, all for an excellent $14 million annual salary tab. I guess this isn’t a surprise, but Phoenix knows what they’re doing.
Loser – Sacramento Kings
Isaiah Thomas produced 8.5 SWAgR in 2500 minutes last year. Darren Collison (3 years, $16 million) churned out 2.5 SWAgR in 2300 minutes. Thomas is 18 months younger. And way more exciting. Even a dummy like me can do the math regarding what the smart move for the Kings was.
Winner – Cleveland Cavaliers
This one seems too easy, right? Lebron has lead the league in SWAgR each of the last six seasons. Even if the Cavs limit Lebron’s regular season minutes to 2800 per season (career low for him), and he starts slowing a bit on account of age, SWAgR projects him as worth 20 wins per season in 2014 – 2015 and 2015 – 2016 compared to a replacement player, right around $1 million in salary per win. In 2500 minutes, Earl Clark, Alonzo Gee, Anthony Bennett and Sergey Karasev combined to be worth -2 wins compared to replacement last year. Replace those minutes with James, and suddenly the dysfunctional, disastrous 2013 – 2014 Cleveland Cavaliers are a 53 game winner. The guy is a big deal.