On Monday, the Washington Wizards will play the Atlanta Hawks in a game that Russell Westbrook will break the record for most career triple doubles. That night, I think almost everyone in the NBA world will be talking about it. And I hate it.
Triple doubles are a combination of everything I dislike about how statistics are used in mainstream NBA media. The main reasons for this are that it's misleading, it relies on the rule of 10, and the stats it uses are not stand alone means of evaluation.
Triple doubles are really misleading. The crux of the argument as to why a triple double is impressive is because it is supposed to signify a player having an all-around game. My problem with this is that the stat categories of points, rebounds, and assists fail to capture even close to everything that happens on the floor. It dumbs down a sport which is incredibly complicated. I love analytics, but I also know that we are not even close to being able to properly quantify individual performance. With all the tracking data, play by play data, and machine learning techniques, we still can't figure out how to tell the actual impact an individual player had on a game. SO how could we possibly do it with some of the game’s oldest stats? Here is what triple doubles are really missing to me: defense, efficiency, decision-making, shot creation, communication, etc. You could sit down forever and make a list of all the impactful actions that happen from a play-to-play basis that does not get captured with the triple double.
Triple doubles rely on the rule of 10. The rule of 10 states that when we look at statistics, we tend to have a bias to think that numbers that end in 0 (or 5) are more statistically significant than other numbers. For example, if a player has 27 points and makes a three pointer, most audiences will be impressed that the player just reached 30 points. However, if he only makes a two pointer to give him 29 points, that will probably not warrant much of a reaction. As someone who looks at numbers all the time, I always try to keep this rule in the back of my mind, because I fall for it all the time as well. The triple double relies on people falling for this rule. The difference between 9 and 10 rebounds is not very significant from the standpoint of impacting a basketball game, but it could be the difference between 0 and 1 triple doubles. That's why a lot of these all or nothing stats that oversimplify the game never make sense to me. Lastly, who decided that the number after 9 should be the first one with double digits? The answer is Hindu-Arabic mathematicians from the 6th or 7th century. So just because these people came up with that particular counting convention 15,000 years ago, I'm supposed to think that getting a 10th rebound is someone a statistically significant feat?
Lastly, the counting stats that triple doubles rely on are all inherently flawed. I want to start with the most flawed one in my opinion, which is rebounds. Here's something that every front office in the NBA believes is 100% true: rebounds are not a good way of measuring rebounding skill. If you haven't really thought about it before, it may sound weird, but hear me out. Rebounds are a product of a series of actions that may or may not lead to the ball bouncing in your direction. First of all, a player has to miss a shot to get a rebound. This is why rebounds are so positively correlated with winning. Second, the ball has to bounce in our general direction for you to even have a chance at grabbing the rebound. A lot of that just relies on luck. Also, teams have schemes for how they rebound. Most NBA teams don't even bother going after offensive rebounds. They would rather just get back on defense and stop the fast break opportunity. That means that most defensive rebounds are uncontested, and whoever grabs them either happened to be in the right spot or the defensive rebounding scheme cleared the way for them to grab it. Next time you're watching a game, spend some time actively thinking about how the rebounds are being grabbed. You'll be surprised. Next, I'll talk about assists. Assists do not do a good job of measuring an individual players passing ability. They are really a team stat. You can't get an assist unless someone subsequently puts the ball in the basket. The best passers in the league will proactively create passing lanes for their teammates, but it's on their teammates to make the proper read or finish. Sometimes, reactive playmakers get a lot of assists just because they happen to have the ball in their hands a lot of the game. Sometimes, the best passers don't average a lot of assists because of their surrounding talent or particular offensive scheme. Assists are very much a team-oriented stat that for some reason has been accepted as an individual measure of playmaking. Finally, there's points. I've written a ton about how I think points per game is one of the most negatively influential stats we have, so I won't talk about it too much here. I will say that the major issue I have here is with efficiency and how the points are scored. There's a difference between creating shots and finishing plays. Also, it's not very impressive in most cases to put up a lot of points while using a lot of shooting possessions. Substantial inefficiency should not be rewarded.
Now let's talk about the man who inspired this article, the one and only Russell Westbrook. If you've read a lot of my previous articles, you know I am not very high on Westbrook. The reason for that is because I don't think he can raise the ceiling of an already good team to push them into contention. I value winning championships so much more than any other outcome, and I think the path to doing so with Russell Westbrook as a centerpiece on your roster is extremely difficult. However, if you want him to be the secondary player on your undertalented team chasing the eighth seed, then I think he's great. Westbrook is still an excellent floor raiser, and that's exactly what he's been able to do this season. There are plenty of really good players out there that fit this mold (DeMar DeRozan, D'Angelo Russell), and those guys can be really valuable in the right situation. I don't think you can win a championship with Russell Westbrook because the scoring is just too inefficient, the shooting still isn't there, and the defense is too poor. Westbrook's shot selection this season has been atrocious. There are way too many mid-range pull ups and three-point jumpers, and not enough attacking the rim. I think some of this can be blamed on the poor spacing around him, but a lot of times Westbrook has mismatches that he can get to the rim against but then opts for the jumper. The thing is that Westbrook is just a poor jump shooter, despite his inclination to take a lot of them. Without Beal on the floor, Washington's offense devolves into Westbrook isolation's, which have not been good this year. The advantage creation off the dribble is just not the same as it was a few years ago. Teams are also able to help across the floor with whoever is guarding Westbrook. We saw that last season with every team that played the Rockets. Centers would guard Russell Westbrook, giving the opposing team an easy out against the Rockets revolutionary roster construction. He ultimately put a cap on the offense. Defensively, Westbrook just isn't impactful as a point of attack defender. He really struggles to get around screens, the closeouts are lazy and slow, and the help defense and instincts are non-existent. Some of that can be because he does a lot offensively, but a lot of the lapses he tends to have are inexcusable. Now, let's talk about some positives. Westbrook is an unbelievably good playmaker. That is by far his best skill. He has gotten really good at reading the helper's help and has gotten to an elite level at fitting in drop down passes and skip passes to the weakside corner. He has developed as a passer with either hand and throws the ball on a string. He has also been impressive with improvisation during sets when he sees an opening of defensive coverage he knows is advantageous. When Westbrook gets downhill, he is still hard to stop. He is incredibly strong for a guard, and once he gets you on his hip it's curtains for the defense. He can finish well at the rim, still draws a lot of fouls, and can make plays like I mentioned earlier. He's also great as a grab and go guy, instantly getting the ball up the court and constantly pushing off opponent misses. This is where the downhill advantage creation really shines. Westbrook is also strong enough to switch some matchups defensively and do a good job. He has a ton of lower body strength, which allows him to push post players away from the rim and force them into longer shots over him. I actually think he's pretty underutilized as a switch guy. Westbrook is also a great communicator and is phenomenal as an energizer. By all accounts, he's a fantastic teammate who is a great leader in the locker room and a world-class human being. That stuff matters.
The thing about Westbrook is that I think he is one of the most interesting evaluations we have in the NBA today. I just wrote about 700 words about Westbrook without bringing up a single number (and if you know me, that's really uncommon). I did that on purpose because I think two smart people can talk about the intricacies and idiosyncrasies of Westbrook's game for hours without bringing up any numbers. Westbrook's game is so unique and so intriguing, but every conversation about him starts with triple doubles. That's really frustrating for me. I just hate how people decide to talk about Westbrook. Everything starts and ends with triple doubles. What are we doing here? Why are we letting conversations about one of the most fascinating players ever get boiled down to a stupid stat that bears no actual weight on impact of winning? I want to leave you with a couple of things to think about:
1. Russell Westbrook averages 2.8 contested rebounds per game. That's not very impressive. He's not one of the greatest rebounders of all time.
2. Russell Westbrook's True Shooting is 50.6%, about 7 points below league average. He is in the 29th percentile in isolation, 28th percentile in scoring on post ups, 32nd percentile on long midrange shots, 7th percentile on 3's, and has a 39.4% eFG% on pull up shots. By no means is this guy scoring efficiently.
3. Russell Westbrook has an assist rate of 47.1%, about 5% higher than the next player. He averages 21.4 potential assists per game, which leads the league by about 2.5.
4. What to watch for next time you watch Westbrook play: How does he get around screens? What is the degree of difficulty on the reads he is making? Is he a good decision maker as a shooter? Was that a contested rebound? Watch how much his fingers and mouth move (communication). Watch how he reads the backline of the defense. Watch him actively increase the pace of the game. Watch if he has a high motor or is just fast. Watch the decision making with mismatches. Watch how he gets guarded when he is off the ball. Watch the decision making and playmaking when getting a defender on his hip. Watch how he defends when he is the bottom weakside defender. And finally, watch his interactions with his teammates (you'll see some fun stuff).
Comentários