Tornicade
2018 RBF League Champion
- Joined
- May 11, 2011
- Messages
- 17,542
- Reaction score
- 10,441
- Points
- 123
That's uncalled for. As someone who looks to use statistics in arguments, you should not diminish your argument with ad hominems.
With respect, you are misusing the Wins Produced statistic. You are comparing two players with very different styles, on very different teams. You are comparing an offensively-gifted player in LaMarcus Aldridge who is a top scoring option on his team (@ 20PTS/38M) vs Tristan Thompson who is still trying to find an offensive game of his own.
However, the statistical flaw is that you are ignoring the input variables into the Wins Produced function. The components themselves demonstrate that the model is not applicable to your comparison. LMA's Per 48 stats compared to Tristan's aren't even close. Tristan gets you 2 additional rebounds at the cost of several turnovers, an assist, and 8 PTS. He has a higher FG%, but to suggest that would translate into 20+ PTS/48 misses the mark, I think.
Secondly, and perhaps more importantly, you're ignoring the team component to the metric, i.e., the strength (and more precisely, the needs) of the team as it relates to the player being measured. Tristan's Wins Produced will be higher on a #3 lottery seeded team than it would be on a mid-tier team. This can be evidenced anecodatally by looking at players who change teams with vastly different surrounding talent pools, as an example, LeBron James: his best year with respect to Wins Produced, was 08-09; however, after leaving Cleveland for Miami and playing with Wade and Bosh, even though his game has reached unprecedented levels, his Wins Produced has still not reached his 08-09 season. Why? Because he's playing with Wade and Bosh, who bring his share of wins down.
Concisely, if you swapped Tristan Thompson and LaMarcus Aldridge, Tristan's wins would decrease and LaMarcus' would increase. That is somewhat readily apparent, prima facie, by simply looking at the players actually play basketball, their individual (less abstract) statistical output, and their team performance.
Simply put, when you make a claim that Tristan is "better than LaMarcus Aldridge already..." you really need to dig a little deeper than quoting what amounts to a misused statistic. However, I would be willing to entertain an argument that Tristan will be better than LMA, because that is entirely possible. If that's what you meant, then I understand if you mispoke or were misunderstood.
Oh, and FWIW, I have a degree in Mathematics.
I agree. statistics and rating systems are tools for analysis not Analysis for tools