In a few short hours, I will board an airplane bound for Boston and the MIT Sloan Sports Analytics Conference. For the second year in a row, I will attend several panels discussing the ever-growing role of sports analytics across the world. The conference is a breeding ground for ideas and new ways to think about the games we love. For me, this was one of the highlights of my (admittedly brief) professional career.
And I could not be more eager to attend again. But this time around, the Sloan Conference is sure to be a different experience for me. I have a whole new batch of curiosities about the relationship between statistics and sports, considering my new perspective -- both literally and figuratively -- on the game of basketball.
You see, this time last year, I was crunching numbers, watching the NBA, and writing about the game from my cold apartment isolated in the sleepy town of Wethersfield, Conn. This season I'm covering the Miami Heat on the ground. As a result, I see the game in a whole new light.
Given the team’s recent struggles closing out games against winning teams, my trip to Sloan is a timely one. Because for someone who subscribes to the tenets of statistical analysis and empirical study, the Heat have tested what I thought to be true: a team's record in close games doesn't mean much.
The Heat have lost some nail-biters this season, especially as of late. They fell to the Knicks by five points on Sunday. Two days earlier, they lost to the Bulls by four points. Two Sundays ago, they lost to the Celtics by three points. This season, The Heat have been defeated by the Mavericks by two points, the Jazz by two points, the Hornets by three points, the Celtics (again) by five points, the Grizzlies by two points, etc. The list goes on and on.
All in all, the Heat are 2-11 in games decided within five points against winning teams.
Now, let me pause for a moment and explain why that 2-11 stat is a bit troubling -- not necessarily for the Heat, but for my analytical senses. This is a classic case of arbitrary endpoints. In other words, why cut it at five points? Because it’s a half of the round number of 10? Probably. But what if the Heat pull away at the end of a tight game to win by six points? Is that not worth our attention? Do we just discard that piece of evidence? Furthermore, are the Heat’s routs against winning teams (i.e. Lakers, Magic, and Knicks) not as meaningful?
Certainly, these are important questions that we must ask ourselves when we see a graphic on a broadcast or read a note in the latest morning column. What does the stat really tell us? Arbitrary endpoints have the power to mislead. Sometimes it’s not what is being sliced, but how it’s being sliced that matters.
And then there’s that pesky little thing called randomness. In close games, a tipped ball or a breath of wind can shift the outcome of a game. On the basketball court, no matter how clutch-tastic a player is or how stacked a team may be, sometimes a basketball game comes down to something as mundane as which team recovered the loose ball with 30 seconds left. These are random events that have the ability to make a winning team a losing one, all merits aside. The sober reality is that luck can determine the outcome of a game, no matter how disturbing that may be.
For this reason, a team’s record in close games isn’t all that predictive. Over the long term, luck tends to even out and the underlying talent rises to the surfaces. In the context of a mere 13 games, sometimes it’s better to be lucky than good. Statistical analysts have found that record in close games means little, if anything, come playoff time. Furthermore, the magic of close-game performance does not translate year-to-year, nor does it matter how many veterans are on the team. No matter how we slice it, records in close games really don’t tell us much. It appears we’re making a big deal out of nothing.
But then I come full circle, putting aside my statistical reservations, and gaze at that record.
How can that tell us nothing? And if it continues, at what point will it begin telling us something?
These questions bounced around my head as I waited for players and coaches to take the stage after the Heat lost to the Knicks on Sunday night. My mind was in the midst a tug of war. How could I reconcile what my statistical principles tell me and what I see on the floor? The Heat didn’t execute down the stretch and that seems symptomatic of something deeper, more permanent than a statistical fluke.
Usually, I lean toward my statistical background, which heeds the power of the butterfly effect of randomness. If Mike Miller’s last-second shot against the Celtics on Feb. 19 falls, things would be different. If Luol Deng misses that late three-pointer on Thursday like he has done 68 percent of his career, things would be different. If LeBron James hits that free throw with 12.5 seconds left back on Feb. 13 in Boston, things would be different. Heck, if Mario Chalmers doesn’t flub a wide-open fastbreak layup on Sunday night, things would be different.
The “ifs” are mounting. Each time the Heat lose a close game, the rhetoric from Heat coach Erik Spoelstra and the players suggests that they are victims of some hard luck. After Sunday’s loss to the Knicks, Spoelstra looked defeated as he spoke to the media. But despite his downtrodden body language, he expressed confidence that his team will get over the hump.
“We will have our breakthrough,” Spoelstra said. “And as painful as this is right now, there will be a time that we break through and we’re able to execute and win a game like this against a quality opponent down the stretch.”
Spoelstra believes that the Heat are bound to turn it around. “We’re close,” he likes to say. He said those exact words again at practice on Wednesday.
I can understand the unrelenting faith he has in his squad. Generally speaking, if the process is sound, soon enough the results will begin to reflect it. Spoelstra is still confident in the "process over results" approach, despite the losses. But that may be a hard pill to swallow for fans.
"No one wants to hear that," Spoelstra said after Wednesday's practice.
Spoelstra isn't satisfied with poor outcomes, but realizes that getting an open shot is the goal, even if the shot doesn't fall. But if he questions his team's execution, it’s an indication that the process side is hurting and, more importantly, that the struggles are not a product of bad luck. The team has watched extensive film from Sunday's loss to the Knicks and practiced their late-game strategies for much of Wednesday's practice. So they're working on it, even if they think they've experienced some hard luck.
The tests will continue for the Heat. Beginning on Thursday night against the Magic, the Heat will have 10 straight games against winning teams. At the end of that brutal stretch, we’ll learn a thing or two about this team. Is the 2-11 record a statistical mirage or is it something real? Honestly, I don’t know. But between the Heat’s next slate of games and this weekend’s Sloan conference, lessons will be learned.