When Did “The Nerds” Start Ruining Each Sport?
Let's look at the development of certain trends that are allegedly making sports unwatchable.
Amidst all of the online discourse over NBA ratings dropping, one of the strongest undercurrents of blame — granted, in a sea of other proposed reasons — centers around the (over) abundance of 3-pointers in the game today. In 2024-25 so far, an all-time high 42.3 percent of field goal attempts are taken from downtown, on track to break 2021-22’s record of 40 percent. The defending champion Boston Celtics are currently taking 56.1 percent of their shots from deep, which would obliterate the 2018-19 Houston Rockets’ record of 51.9 percent.
As recently as the mid-2000s, no team in the league jacked up 3s even half as often as the Celtics do now. But this is how basketball — winning basketball, even — is played now. And like many strategic trends that have drawn complaints in recent years, it’s easily traced back to those darned nerds infiltrating the game.
Here’s a timeline of leaguewide 3-point attempt rate since 1999-2000, annotated to show key moments on the path to today’s style of play:
Many — though not all — of those important developments trace through the Houston Rockets front office tenure of Daryl Morey, the earliest and most famous purveyor of analytics among NBA decision-makers, and his partnership (starting in 2012-13) with James Harden, one of the most prolific 3-point shooters in NBA history. Since the heatmap of efficiency on an NBA court confirms the soundness of taking 3-pointers over pretty much every shot beyond the immediate vicinity of the rim, it’s not difficult to draw a line of causality from Morey’s influence — and thus, the widespread adoption of numbers in the game — to the current NBA gameplay meta, with its many discontents.
Basketball isn’t the only sport whose aesthetic changes have been blamed on the proliferation of analytics types within the game. Baseball’s excess amount of walks, strikeouts and home runs — the “Three True Outcomes” (3TO) that obviate the need for fielders because the ball never lands in play — has also been laid at the feet of nerds, since they simultaneously pushed for more strikeouts by pitchers while disregarding the negative signal of strikeouts by batters.1
MLB’s trend toward fewer balls in play has been more gradual than we see with 3-pointers in basketball, however, with detours along the way. There was a brief increase in 3TO% during the steroid era of the 1990s, for instance, which predated the widespread adoption of analytics in MLB — though the upward trend from 2005 into the 2020s could more fairly be linked to data-driven strategies:
Baseball may also provide a blueprint of sorts for nerds to save the game, by changing some of the incentives that led to fewer balls in play (and fewer stolen bases) through data-driven rule changes such as the pitch clock and larger bases. (Though it’s somewhat amusing that 3TO% actually went up the first year of those new rules, it dipped some in 2024.)
But the most interesting comparison of all comes in football, a sport where isolated player performance is vastly more difficult to measure through data, and thus the resident nerds tend to be more tape-watcher types than pure number-crunchers.2 There are still plenty of lessons to be gleaned from the math of football analytics, of course, many of which have hammered on the idea of less conservative play-calling — passing more in traditional running situations, and going for it both on fourth down and as 2-point conversions after touchdowns.
We can see the clear uptick in both of those trends across the NFL, with the latter beginning to rise in 2015 and the former seeing an increasing trend starting in 2018:
But while those trends clearly show signs of analytical influence, football hasn’t drawn anywhere near as much criticism for its aesthetic changes as either baseball or basketball have.
Why not? It could be because these play types have a fundamentally more limited scope within the overall game. The average team goes for it on fourth down about 1.3 times per game and attempts a 2-point conversion once every 3.9 games — decisions that tend to come at moments of outsized importance, but don’t dominate the entire game on a play-to-play level.
Meanwhile, other NFL trends potentially driven by data, such as increased passing on 1st-and-10, have actually gone backwards in recent years:
With football, the lack of aesthetic complaining could also be that the analytically-recommended play style — “throw downfield, never punt, go for 2 a lot” — is just fundamentally more fun, especially to a generation of folks who grew up playing Madden exactly that way on PlayStation. The ability for nerds to penetrate the NFL game is inherently more limited by a complicated sport, generally more understood through tape and not numbers, and what penetration there is tends to lead to a more exciting product anyway.
But at the same time, those who say it’s boring to see NBA teams chuck a bunch of 3s nowadays — and instead pine for the golden age of ‘90s post-ups and mid-range J’s — should actually go back and watch the spacing from that era. Is that really what we want to go back to?
(It’s also not true that every team today plays the same style, which is a complaint you often hear. While every team takes at least 33 percent of their shots from downtown, which would have been the all-time record prior to the 2009-10 Orlando Magic, that still leaves 22 percentage points of variation between the most and least 3-happy teams in the league. Plus, some teams do cool things like the Grizzlies, who never run a pick-and-roll or dribble handoff.)
I’ll leave you with one final thing that stood out to me in Adam Silver’s comments about the TV ratings:
“It’s not unique to the NBA, where analytics start to be too controlling and create situations where players are doing seemingly unnatural things because they’re being directed to do something that is a more efficient shot,” [Silver] said. “And part of what we’re focusing on, too, is that what makes these players so incredible is the joy they bring to playing the game and the freestyle notion of the game too.”
It sounds like he’s saying it should be on the players — and the players alone — to determine the most effective way to play the game for themselves. But was that ever the case?
Obviously, no. Coaching has always been an aspect of sports, going back to the very beginning. While plenty of players have chafed under their coaches’ instructions throughout history, the role of someone who sees the big picture, directing players to do the efficient thing, is far from a recent development brought about by the nerds. Modern analysts may have added different insights, but the goal is and has always been finding better ways to score points for your team and prevent the opponent from doing the same.
It’s easy to romanticize a past where sports were “freer,” untethered by optimization and the influence of external strategists. But the truth is, sports have always been shaped by those who innovate. (The advent of the 3-point line itself was an invention of the 1960s, with the fledgling ABA trying a new gimmick to help compete against the NBA.) If the game is indeed suffering, in the ratings or otherwise, because of the recent innovations brought by data and analytics, it’s up to Silver and the league to change the incentives of the sport to encourage something else.
Filed under: General, Basketball, Football, Baseball, Statgeekery, Miscellany
At first glance, this may seem paradoxical — but it has to do with the fact that strikeouts are little worse than any other kind of out from the perspective of changing a team’s run expectancy as a batter, while they are hugely predictive of future effectiveness from the pitcher’s perspective.
Basically, wannabe coaches rather than wannabe GMs.
The fact that by the end of the season the ratings had largely recovered I think points to the NBA being an inventory sport and one which has figured out that it doesn't make sense to put out the good stuff when the NFL is still playing.