Photo from blmiers2 via Flickr
There isn’t a single phenomenon more fascinating to the typical box score scavenger than the triple double. Supposedly the ultimate achievement for all-around dominance in a single game, we crown the nobel achievers and bemoan those who taint its special glow by besmirching the game or by being Ricky Davis.
Of course, this is stupid. Tom Haberstroh’s fantastic work regarding the experience that was last season’s Miami Heat served as a constant reminder to the risks presented by arbitrary endpoints, and it doesn’t get much more arbitrary than the triple double. Who are we to decide that the conversion from single figures to double figures is what makes a statistical achievement impressive? What if we were all Simpson characters with only 4 fingers on each hand? Would we then have the triple-octo? Would anybody in their right mind prefer a 12-13-11 night from your prime Jason Kidd over a 30-15-5 masterpiece from your prime Kevin Garnett?
And yet, we are drawn to the triple double like internet moths to a more-digits-in-box-scores flame. The likes of Oscar Robertson, Magic Johnson, Jason Kidd and LeBron James would be more than worthy of our adulation regardless of where they stand in respect to this somewhat pointless achievement, but their legacies have been enhanced by how often they ascend to the top of our trivia books. And on the other side, we can always remember that ridiculous night when Chuck Hayes, completely unprovoked, went and got himself that trip-dub.
This shortened season has been completely insane in every way, so it shouldn’t surprise anybody that the triple double world has gone beserk as well. After more than a quarter of the season, we’ve had just two triple doubles league-wide: Rajon Rondo against the Wizards and Kyle Lowry against the Wolves. After discussing this a bit on twitter with HP’s own Andrew Lynch  I was left both baffled and curious: is the arbitrary nature of the triple double creating an arbitrarily small amount of them, or did something actually happen that created this numerical singularity? Could this be a post-lockout swag, such as the dip in offensive efficiency, or am I wasting my time?
Determined, I ran to basketballreference.com’s all-powerful play index. For each of the past three seasons, I checked:
- How many triple doubles we’ve had.
- How many players missed triple doubles by 1 rebound or 1 assist.
- How many players missed triple doubles by 2 rebounds or 2 assists.
- How many players missed triple doubles by 1 rebound and 1 assist.
I decided that missing triple doubles by a maximum of 2 combined rebounds and assists was a fair cutting point, but it’s very easy to see that this is yet another arbitrary endpoint – which, in a very similar manner to last season’s Heat, was created by sheer laziness. That’s life, I guess.
After taking out the games that fit my criteria, I adjusted the numbers to the shortened season by dividing the results by the number of games played (1230 for regular seasons, 271 so far this year), to create what is effective a “triple doubles per 100 games†stat for the entire league. Also, for the sake of this exercise, I chose to ignore the ever-rare points-rebounds-blocks triple double, both due to its scarcity, and because I refuse to include JaVale McGee in an otherwise pure chart. Sorry, Vale.
The results!
It’s very easy to see that 2007-2008 had dramatically more triple doubles than other seasons, and the reason why is very clear: Jason Kidd was still in New Jersey. In the 4 months of Jason Kidd wallowing in mediocrity before coming back to Dallas, the future Hall of Famer posted a ridiculous 12 triple doubles; he’s only posted 8 in nearly 4 whole seasons since. With Kidd’s usage rate dropping dramatically once he got actual teammates next to him, this should be no surprise.
Kidd aside, we can very clearly see that individual columns are susceptible to major variations on a per-year basis – clearly, a result of the small sample size that is eligible games. For example, last year we had 36 triple doubles, and another 22 games of 10+ points, 10+ rebounds, but only 8 assists. Conversely, we had only 6 games of 10+, 10+ and 9, after exactly 11 such games in each of the previous 3 seasons. Indeed, looking at this chart, it is very easy to dismiss it as completely random.
However, the column you should be looking at is the one at the far right. When totalling all of the games that fit into one of the categories we presented, we see so far this year, we’ve had 7.38 games classified as either a triple double, or close to one – a number that is nearly identical to previous years. Again, the only outlier is 2007-2008 – which drops right back into the 2008-2012 range once you take out all Nets-Jason-Kidd games.
Sometimes, things are just that simple. While the triple double is, in general, an impressive feat, it could also just happen randomly because JaVale McGee was trying too hard, or because Kyle Lowry is too good not to do so. The lockout has done many things that would seemingly help/hurt triple doubles – minutes, assists and points are down, while rebounds are up – but sometimes, things are just random.
