National TV Rondo Actually Exists

By Ryan Fortin

Throughout Rajon Rondo’s career, he has always shone under the spotlight. He is, of course, Boston’s best point guard and arguably the best player on the team. But as Grantland’s Bill Simmons has noted on occasion, he seems to save his best games for when he appears on national television, including many of his triple-doubles. Is it true that Rondo actually tries harder or performs better when he is in front of the entire country?

I decided to test this theory by compiling data over the past two years and running a t-test, using his non-nationally televised stats and his nationally televised stats to see if the two groups were significantly different. The results seem to back up the critics:

Stat Nationally
Televised
Average
Non-Nationally
Televised
Average
T-Test
P-Value
PPG 16.1 13 0.1029
RPG 6.8 5.3 0.1375
APG 10.2 11.6 0.8349
STL 2.3 1.8 0.1457
BPG .5 .2 0.0678
FGA 13.7 11.6 0.1183
TO 3.6 4.1 0.2699

All of his averages, including points, rebounds, steals, and field goal attempts are higher in nationally televised games. Though none of the p-values are below 0.05, and thus not “statistically significant,” the sample of games is small enough that we should still take notice — I think they can still be considered practically significant results. Interestingly, his assist total is lower, but the high p-value suggests that there’s not much to read into there.

It would appear Simmons is right — Rondo does play better when the most people are watching him. Whether it’s a positive thing that he can elevate his game in big contests or negative that he doesn’t play this way all the time is difficult to say, but it shows why Celtics fans often become frustrated with their talented point guard.

About the author

harvardsports

View all posts

12 Comments

  • I love it.

    Here’s my question: did you control for the types of games/opponents/home vs. road stuff? I’m guessing nationally televised games occur only for specific kinds of games, and perhaps those games lend themselves to more scoring?

    • My thoughts exactly, national TV games generally consist of two top teams. So I can’t say how pace affects top teams, but you would assume he would have worse stats against top teams but instead he has better stats. This means 1 of 2 things, one that he is trying harder to rack in stats or two that national TV games are less likely to be blowouts so he gets more minutes in nationally telivised games. More interested in per minute numbers controlling for opponent.

      • Zachary/Jay,

        I was going to ask the same question about minutes, and ultimately, possessions. It would seem that controlling for possessions, if possible, could shed some more light on the question!

        • My guess is that, given the already statistically insignificant p-values, adding more regressors would render the analysis useless. Maybe we can check back after another 5 seasons? 🙂

          • Maybe, but not necessarily! Sometimes the lack of a much-needed variable (possessions) can shed light on the value of the other variables (nationally televised?), making them even more significant… I mean, sometimes.

  • Actually, from my observation, rondo’s better stats vs the better defenses on national tv shows just how much rondo loafs vs the bad teams when the spotlight is off. It’s pretty much insulting to Celtic Pride and the greats who came before him.

    Dianna Ross probably showed up late and gave half-azed performances in smaller venues, it’s the Diva disorder.
    Just last week we saw justin bieber do just that in london, his excuse being that it was the first of 4 shows at the same venue and thus not very important.
    The Diva Disorder wouldn’t matter for the celtics, given his postseason prowess, but the season is long. Teammates don’t like inconsistent effort, especially from a ‘leader’, and they dont like having to work extra hard because rondo wiffed on the sportscenter steal/assist/offensive rebound.

    There are other aspects of diva behavior, like acting better than your non-hall-of-fame teammates, wanting to control everything on the court, and whatever else we can’t see in practice and lockerroom.

    For all the stat crunching, there was a significant lightening of spirits, increased confidence, and increased effort among rondos non-hof teammates after he left.

    This was rondo’s year to show he could be the cornerstone. He blew it.

    • There is no Diva Disorder. You’re clearly no shrink. How can playing better than what he does on average when facing a tough opposition be a disorder? Rondo was playing better than he ever did at the time of his injury this year. What you may want to look at is how much more efficient Pierce and the others have been since he went down. Pierce went through a really forgettable stretch of 6 games — the last 6 games Rondo played, which the celtics lost. Rondo had 2 efficient triple-doubles in them. Rondo didn’t blow nothing. He’ll be back and do just fine.

  • Ignoring what the quantitative analysis tells you (tests are inconclusive, at best) in order to push a predetermined “eyeball test” conclusion is simply bad form.

    And then there are also legitimate questions about the needs of additional controls, which is a whole other story.

  • I reckon you’ll find similar results for the likes of Chris Paul, Kevin Garnett, Paul Pierce, Dwyane Wade etc.. The minutes played is an important covariate.

Leave a Reply to Zachary Cancel reply

Your email address will not be published. Required fields are marked *