If you follow me on Twitter and were listening to me today, you’ll already know the story to this and could probably skip ahead a couple paragraphs. If you don’t, then here’s why I’m about to apologize.
Early today, an article from Joe Nelson of KFAN surfaced where he was arguing that PER, or player efficiency rating, should not be trusted as much as it should. His reasoning was that you could create a team with the leaders of PER at their respective positions, and on name and star power alone would not be as good as another team that Nelson cherry-picked. His team was made up of Steve Nash, Monta Ellis, Kevin Durant, Blake Griffin, and Greg Monroe, while his “Stats Team” consisted of Jeremy Lin, Sundiata Gaines, Nic Batum, Ryan Anderson, and Nikola Pekovic.
Nelson (at least I believe this is what he did) added up the average points per game for the starting five players on each team, and reasoned that his team could beat the “Stats Team” 103-72. He then added that the reader might argue that Sundiata Gaines plays less than 20 minutes a night, and with more time could be a better scorer. After all, Gaines has a PPG of 5.7 this season, but a points-per-36 minutes of 14.9. However, Nelson argued that there’s a reason why Gaines doesn’t get more playing time, and that it’s because he’s not as good as Kevin Durant (I feel a more accurate comparison would have been with Ellis, as they actually play the same position).
Regardless, I set to work on tearing Nelson’s article apart, Fire Joe Morgan-style. I copied and pasted his text, bolded everything, and started filling in my own words in between each of Nelson’s paragraphs.
The first thing I started with was the ragtag group of players Nelson chose for the “Stats Team.” Why Sundiata Gaines, seriously? I went over to Basketball Reference (because where else was I going to find PER?) and instantly found the Top 20 players in PER this season.* I found it notable that Gaines was nowhere to be found on this list. Then I thought that Nelson disregarded playing time and just found Gaines’ PER and threw him into the leaderboard, just for fun.
* It’s there, I promise. Second column from the left, seventh row down.
Well, that didn’t work either, as Gaines’ PER of 18.7 didn’t even crack the Top 20. That’s when I remembered something. Nelson was picking players that were tops at their respective positions. Alright, on to the top point guards!
Unfortunately, I couldn’t sort the Top 20 in PER by position on Basketball Reference, so I had to do everything manually. Chris Paul, Derrick Rose, Russell Westbrook, Steve Nash, Tony Parker… nope, Gaines would not have fit on this list.
At this point, I thought of something odd. Why the hell did Nelson completely ignore LeBron James, the leader in PER this season? If you were going to compile a Stats Team from leaders in PER, he must be on it, and yet he wasn’t. Instead, Nelson chose Nic Batum, some guy I’ve never even heard of (I watch the Wolves all the time, but I’ll admit I don’t know my basketball players as well as my baseball players) as his small forward.
I decided to continue with my argument, and my next step was to create my own Stats Team, from the actual leaders at each position in PER. Using that Top 20 list, I came up with Chris Paul, Kobe Bryant, LeBron James, Kevin Love, and Dwight Howard. Now THIS was a Stats Team you could be proud of, and it would easily rival Nelson’s preferred team.
Next, I noticed that Nelson had said that each team he had created was “loaded with top-five PER talent.” Seriously? Remember, Sundiata Gaines was nowhere to be found. After taking some time to decipher what Nelson meant by “top-five PER,” I put together the top five players at each position. This came to be difficult on Basketball Reference, as it would not sort the players by position. I resorted to using my memory, the player’s pages, and Wikipedia for help. I even consulted a box score for the Portland Trailblazers to figure out if LaMarcus Aldridge was a power forward or a center.
After what felt like half of my afternoon, I finally started looking at possible other places on the Internet that would have PER. The NBA website had an “efficiency” statistic, but the numbers didn’t match up with Basketball Reference, so I felt that these were not the same statistic. Eventually, I found – what else? – John Hollinger’s pages on the ESPN website. Hollinger, if you read Nelson’s article, was also made fun of as he is the creator of PER. Here on the ESPN site, I could finally sort players by their position and see how they rated in PER.
I ended up making this list, and remembering Nelson’s “loaded with top-five PER talent” quote, started comparing players on all three teams: Nelson’s best team, his Stats Team, and my Stats Team. Obviously my Stats Team was superior to his, as I took the best player at each position (bolded). Nelson’s inferior Stats Team was underlined, and his hand-picked team was italicized, as shown below. Interestingly, I found out that Gaines was listed as a shooting guard instead of a point guard like when he was with the Timberwolves.
Note: I excluded Andrea Bargnani from the Center list. According to ESPN, he would have ranked fourth, but I took him off because he had only played 13 games this year, significantly the fewest among all 25 players listed. Besides, I thought he was more of a power forward than a center anyway.
I still had something bothering me in the back of my mind. Why Sundiata Gaines? Why? Obviously I knew about Jeremy Lin. Nikola Pekovic was an odd choice, but I went with it. Same with Batum and Anderson. But Gaines?
Struggling to find the answer, I reread Nelson’s article. Finally, after having 75% of my Fire Joe Morgan-style post finished, I found a word, a single word that answered why Sundiata Gaines was on Nelson’s Stats Team (emphasis in the following quote is mine).
If you look at the top five players at each position and draft a stats geeks team and an NBA fans team, this is what it could look like.
Well, shit. Go back to my top five list at each position above. Remember, Nelson’s Stats Team was the underlined group above, and indeed all 5 of them were in the top five in PER at their respective position.
I must admit, it was at this point that I had a deflating feeling. I was primed to just absolutely tear this article apart, and that feeling had been disarmed by a single word, of all things. On Twitter, I compared it to comedian Demetri Martin and how the two words “sort of” can change the meaning of an entire sentence. “You’re going to be just fine…. sort of.” “I love you… sort of.” You get what I mean.
Joe Nelson, if you happened to take my invitation and read this, you may not have realized this on Twitter, but I was trash talking you all day for this article. But now, I see that I was wrong, and I’m sorry. I still disagree with the gist of your post – after all, I bet this discovery means that a smart team should be trying to acquire Sundiata Gaines for relative peanuts – but not to the extent that I was feeling earlier today.
I know bloggers like myself get a bad reputation for what we sometimes do, and because of that I feel that I should hold myself accountable when I’m in the wrong. I hope that from reading this, others can do the same. At the same time, I feel that I should use this as a learning experience. If I disagree with something, I should get the facts first, like I did with this PER post. By diligently sorting through the statistics and attempting to get the same frame of mind as Nelson, I was able to figure out his intentions with his article, and it turns out that it wasn’t as off-base as I first thought.
Nelson, I was going to unveil a catchphrase of sorts at the end of what this post was originally going to be, but it ends up that I’m going to have to use it on myself. Playing with the name of this blog, today, I was the one that was off the mark.