Welcome to Project Goliath, the ultimate statistical test to take just about every stat you can dream up, traditional and advanced, and use them to see which stat (or combination of stats) produces the most reliable lists of the greatest NBA players of all time.
Because let’s face it. We know all-time greats when we see them. A stat that leaves Michael Jordan off the list is obviously going to be either a garbage stat or else slanted too far away from shooting guards to be worth comparisons across positions.
Likewise, a stat that leaves out Hakeem Olajuwon or Tim Duncan or Karl Malone is probably too guard-heavy a stat to work on its own, but might be useful when combined with a stat that favors power forwards and centers.
So we’re going to try everything.
Keep in mind, we’re going to run into some problems here, namely that certain stats are naturally going to leave out guys who played before 1973; the introduction of offensive rebounds, steals, blocked shots, and turnovers as official NBA stats (and individual turnovers starting in 1977) are going to create more opportunities in counting stats for guys who played more recently.
Similarly, you can probably already see a problem with this very first introduction to the method (and that’s what this is this week); what about adjusting for pace?
Don’t worry. We’ll get to advanced stats later. Let’s begin at the beginning.
Ranking the GOAT by Points Scored
Let’s begin at the beginning. The object of the game of basketball is to put the ball in the net, and the guy who scores the most points is usually the best player, right?
So, the top 12 players by points scored, NBA and ABA combined, in history:
Hmm. That’s…good, but there’s no point guard, I refuse to believe that Issel and Hayes are two of the 12 best players ever to don a uniform, but OK, not a bad stat.
Plus, if you want to rank the Mailman and Kobe ahead of LeBron and MJ in a GOAT debate, you’re doing it wrong.
No, we’re going to have to dig deeper. Points are a nice start, but…
How About Per Game?
How about it?
OK, any stat that makes Michael Jordan the GOAT can’t be all bad. But Elgin Baylor the third-best player who ever lived? Kevin Durant a top-6 guy all time?
At least we got a couple of point guards on this list. You could actually mold this 12-man unit into a team that plays actual basketball.
Kareem is 15th in points per game, for what that’s worth. Hang around the NBA long enough and you’re going to score lots of points seems to be the moral of the story.
No, we’re going to have to get creative here.
Points Plus Rebounds Plus Assists Equals GOAT?
Let’s jump right past rebounds and assists as individual stats for what should be obvious reasons, shall we?
We’ll take the total of points, rebounds, and assists, rank them 1 through 12, and…
Kareem, Wilt, Mailman, LeBron, Kobe, Dirk, Kevin Garnett, Hayes, Tim Duncan, Moses Malone, Shaq, and MJ.
Nope. Not buying it. Michael Jordan 12th and LeBron 4th? No. Clearly counting stats are not going to do the job here.
OK, Once Again, How About Per Game?
OK, how about it? Here’s your top 12:
Wilt, Pettit, Baylor, Big O, LeBron, MJ, Bird, West, Kareem, Mailman, Russell Westbrook, and Durant.
Nope. Wrong again.
PART 1 CONCLUSIONS
So what have we learned, besides “counting stats are a lousy measure, especially when comparing across eras?”
Nothing. We’re gonna need some better stats. And I’m going to need to brush up on my Excel skills, because this is one hell of a data dump I’m doing from Basketball Reference, whose Play Index is making just about all of this possible.
These will run on Wednesdays and Sundays until such time as we’ve indeed fine-tuned this measure to find the most useful stats in NBA history for truly sorting the all-time greats from the stat-padding glory hounds.
So stay tuned for Part 2 on April 21 and thanks for reading!