Games in the NBA have always been 48 minutes long. Other rules have come and gone — zone defenses, the 3-point shot, a ban on zone defenses, the shot clock, zone defenses again, the 82-game schedule — but game length has been sacred.
Those 48 minutes stretch longer than ever now, as the league and its television partners1 cooperate in accepting heaps of cash from advertisers willing to pay more and more for a slot during one of approximately 967 timeouts that interrupt any given nationally televised game. The games are too long. An NBA playoff game should not nudge up against Major League Baseball–level game times.
"Listen," says Jeff Van Gundy, the former head coach and current TV analyst. "There just shouldn't be three-hour NBA games." A three-hour NBA game goes against everything basketball is supposed to be — a fast-paced ballet of whirring athleticism, back-and-forth action, and fantastic players looping around the floor in coordinated geometric patterns.
Almost everyone associated with the league knows game lengths have gotten out of hand. Not everyone actually cares; multiple ownership sources suggested the only people concerned enough about game length to propose semi-radical solutions are those tasked with working the games in some capacity. But the league has telegraphed its concern. The NBA cracked down on pregame dance-and-handshake rituals last season, and two seasons ago it had in-game operators sound warning horns in order to hustle teams out of timeouts. The D-League over the last three seasons has discussed a bundle of proposals (and even implemented some) in hopes of containing games within a tidy two-hour window. A lot of these ideas centered on the unending stop-and-start of NBA/D-League crunch time, including a limit on substitutions and a FIBA-style rule that would ban live-ball timeouts. Every expansion of instant replay comes with the requisite platitudes about balancing accuracy with game flow.
Some proposals to streamline games are unrealistic, and others would take such tiny bites out of the problem as to be almost irrelevant unless enacted simultaneously with a dozen other jabs at the issue. Perhaps the most elegant solution is the simplest: shorten the games from 48 minutes to 40 minutes. Several sources around the league, both at the highest team levels and within the league office, say commissioner-in-waiting Adam Silver has signaled a desire to at least discuss moving to 40-minute games. About two years ago, Silver informally polled all 30 league general managers on the notion of cutting overtime from five to three minutes, per several GMs who remember the poll.
The 40-minute move has two immediate sources of appeal, beyond increased watchability:
1. It would align NBA and FIBA game times. This is a globalized sport, and it's only going to get more globalized. Next year marks the first FIBA Basketball World Cup, an event that could be massively lucrative for the league — provided the NBA and FIBA strike a deal to split revenues in some way.2 The NBA has considered adopting FIBA's goaltending rules, which allow players to slap at the ball once it is on the rim. FIBA scrapped the old trapezoidal lane and will face more pressure to adopt the NBA's longer 3-point shot.
"The 3-point line is too short in FIBA," Van Gundy says. "We already shoot too many 3s in the NBA. We should keep moving that thing out to get it to a point where instead of everybody shooting 3s, it becomes a much more difficult decision."3
"There should be one rule for all of basketball, college included," says Dirk Nowitzki. "The same minutes, the same shot clock, and the same 3s. It's ridiculous we play the sport three different ways."
2. Cutting game length should make NBA games more unpredictable. This is the money argument — the line of thinking that could win over entrenched ownership groups afraid of any step that would reduce TV ad time and jeopardize the growing cash bonanza it provides. Stat geeks are in something near universal agreement that the NBA is the most predictable of the four major U.S. sports, in terms of both the outcome of single games and of playoff series. This is not the same thing as saying the NBA lacks parity. (The lack of competitive balance in the NBA is a related problem, but it might be even harder to solve than game-to-game predictability, as strange as that sounds.) In simple terms, favorites do better in the NBA than they do in other sports, often winning games by gradually pulling away as their superior skill cements its impact over an increasing number of possessions. The NBA is unfriendly to underdogs. Fans seem to like sports that are friendlier to underdogs, and especially sports where every game carries a do-or-die significance; the NFL is by far the most popular sport in the U.S., and TV networks pay nearly as much to broadcast a three-week single-elimination college basketball tournament as they do for an entire season's worth of NBA games.
There is some debate over how unpredictable the league could ever be, but it's a statistical truism that reducing sample size — i.e., possessions — cuts the talent edge. "The NBA by far is the sport most dictated by skill," says Michael Mauboussin, an adjunct professor at Columbia Business School and author of one of the go-to analyses on the role of randomness in sports. "You reduce the sample size, and you let luck play more of a role."
The NBA is doing very well. Its popularity curve is trending the right way, and its audience is young. Change carries risk, and there is a sense among folks in all branches of the league — players, coaches, GMs, owners, NBA officials — that there is no need to upset the current balance of things. Adopting the FIBA system of four 10-minute quarters would reduce game length, naturally cutting the raw amount of advertising time networks would have available. It might cost the league two in-game TV timeouts.4 The league and its network partners could try to salvage as much ad time as possible, perhaps by lengthening halftime or some other sort of gimmickry, but a decrease in commercial minutes would seem inevitable.
The league's TV deal expires after the 2015-16 season. The next one will almost certainly be worth hundreds of millions more annually, and if league popularity keeps mushrooming during the course of that deal, caution may overwhelm innovation. There is also concern that teams would have to reduce ticket prices if a ticket brings less in-arena entertainment time. Scoring totals will drop, and fans have long had a visceral, negative reaction to team point totals that start with 7s and 8s.
But caution can also lead to damaging stagnancy. Slashing 17 percent of game length might engineer the kind of uncertainty that draws more eyeballs and holds them longer. It could create an even better in-arena experience for which people would pay more. Every possession would take on a slightly greater importance, and thus (hopefully) an even greater intensity of play. The pace of games would seem brisker, and watching one — either in person or at home — would be less of a commitment. Players wouldn't have to log as many minutes, a change that could have a dozen positive side benefits.
Ever seen a super-close FIBA game? Holy cow, are those things exciting! And they're over in 90 minutes!
But there are factors beyond money that could torpedo the dream of NBA games that actually feel fast and leave time for Regular Human Life things:
• The possibility that the change is actually anti-competitive. This was the most common objection I heard to the 40-minute dream: What if coaches keep minutes levels for stars at or near current levels? In other words: What if LeBron James, Dwyane Wade, and Chris Bosh can play literally every second of the games when they matter? Several teams that have stayed close with Miami in the postseason have done so at least in part by pouncing on Miami when James is off the floor. What if he's never off the floor? "Before I say [the change] would make games less demanding, I just wonder if starters would still play the same minutes," says Steve Nash. "I can see the positives — more meaningful possessions, more intensity. But I could see a lot of coaches playing their stars 37 to 40 minutes anyway."
Starters hoarding a larger percentage of minutes could have implications for both cap management and roster size. Why pay bench guys if they are barely going to play in high-leverage games? General managers like the challenge of unearthing bench players and constructing deep rosters, and NBA life would be just a little less entertaining if the bench characters we all love logged fewer minutes or even moved to other leagues. Any reduction in roster size would have to go through the players union via the collective bargaining process,5 and the union historically has worked hard to protect the interests of midlevel and fringe players.
Ditto for any move to lift or eliminate limits on maximum player salaries, a change several sources around the league suggested would have to come along with the 40-minute game. If stars are going to be even more important, the theory goes, the league better do its damndest to make it hard for two or three of them to join up on one team. And that becomes much harder if suitors could offer LeBron $35 million — much more than half the salary cap.
The current CBA mandates that players, as a collective, receive about half of all league revenues; maintaining that system and slashing a roster spot would result only in fewer players splitting the same giant pool of money. Playing eight fewer minutes also reduces the players' workload. Earning more + working less = happy people, at least in general.
But doing all this and scrapping the limit on max salaries could bring a salary distribution curve that would hurt a huge majority of the league's players. That such a curve would represent a more accurate version of what each player is actually "worth," in basketball terms, would probably not mollify mid-rotation players looking at large pay cuts.
But let's just imagine the league adopts a 40-minute game while keeping caps on max salaries. Even then, most league executives and stats experts I spoke with are confident the reduction in minutes and possessions would outweigh the potential anti-competitive effects of stars hogging more minutes — that games would still be less predictable, a healthy goal. The degree to which randomness would increase is unclear, and there are lots of people around the league who view even a small bit of randomness as a bad thing. They like that the NBA is the sport in which the objectively best team wins most often.
All valid arguments. I will say this: I wonder if the league would find a "new normal" of minutes distribution after a short transition period in which coaches would indeed play stars the customary 36 or 38 or 40 minutes. Coaches in international leagues have long used substitution patterns and minutes totals for stars that would baffle NBA fans.
• There are less dramatic tweaks the league should try first. This sentiment stems in part out of respect for the game's tradition, and especially for the game's historical statistics. Per-game totals for every statistic would drop, and we'd have to recalibrate all of our statistical benchmarks for greatness. "The history of the game demands that before you do something this radical," Van Gundy says, "you try other means."
The historical numbers argument is the one I don't really buy. Career counting stats never seem to have resonated with NBA fans the way they do in baseball. Quick: How many career rebounds does Tim Duncan have? How many 3s has Jason Terry — no. 4 all time! — knocked down? What's a Hall of Fame–level total of blocked shots for a center, or steals for a guard?
There's definitely an understanding that 20,000 career points is a vague marker of historical greatness, and you can deduce from there that 10,000 rebounds or assists work the same way. But do you have any clue how many guys have, say, 5,000 career assists or rebounds? Or 15,000 points?
Per-game stats factor into the day-to-day conversation about the league in a deeper way; we still talk about "20-10 guys," and the scoring title always gets some late-season press.
But the league again and again has adopted rules that changed the per-game statistical landscape, and fans and media have adapted just fine in rejiggering our scales. The 3-point shot is probably the single most important innovation since the shot clock, and the league's move to legalize zone-style defenses has rendered the 20-plus-point scorer almost extinct.
Sports generates its expected share of uninformed shouting media commentators and doltish fans who ruin Twitter. But the huge majority of people are well-meaning, and we're all smart enough to adapt along with the league. If I can change, and you can change, everybody can change!
The smaller tweaks to address game length would include all the stuff referenced above, as well as other ideas the D-League and team executives have tossed around: shorter halftimes;6 making sure a game scheduled to tip at 7:30 actually tips in the general vicinity of 7:30; reducing timeouts in crunch time; eliminating one timeout and replacing it with a free advance of the ball from the baseline to midcourt; and allowing teams with the lead late in games to decline free throws after intentional fouls and take the ball out of bounds again instead.
Perhaps there are also smaller ways to inject some outcome unpredictability. One suggestion that came up a few times: eliminate fouling out. Coaches probably overreact to foul trouble, both early and late, and underdogs who sit their best players for extended stretches due to real or imagined risks obviously hamper their chances of pulling off an upset.7 The proposed fix would charge teams a technical for a player's sixth foul, and for each one after that, but the player would never foul out. (A dream scenario for Jan Vesely!)
• Tackle 82 games first. There is broad agreement that the NBA season involves too many games. There is a large overlap between those concerned about season length and game length, but addressing two issues so core to the league's structure is a lot to ask; changing either by itself would require in-depth negotiations with major stakeholders, including TV networks and the players union. Cutting the number of games wouldn't inject as much unpredictability as a minutes reduction, but it would create a scarcity effect that might eventually drive up ratings, ticket prices, and the advertising revenue that determines what the league can wring from networks. If there are fewer games, each one means more.
Cutting the number of games is also the surest way to reduce the wear and tear on all players without disrupting current patterns of minutes distribution. It's not necessarily an either-or proposition, especially in the long run, but in the short run, there are lots of folks in the league who would prefer tackling 82 games before tackling 48 minutes.
• The direction of change. There is an undercurrent of thought that the NBA, as the world's best league, should be the global standard-bearer of basketball — and that most rule changes should flow in the direction of other leagues adapting to the NBA brand. "FIBA should go to 48 minutes," Nowitzki says. "Forty is too short for me." Adds Van Gundy: "There should be some compromise, but FIBA should align with the NBA."
And a few powerful league sources told me they liked how the 48-minute game length differentiated the NBA from both FIBA play and college basketball.
Look, any major change is going to be difficult. Inertia is powerful, and doing nothing is always the easiest option — and it might be the best one for the NBA. But too many games are just too long, and the momentum of instant replay is moving the NBA even further in that direction — to the point that a few small changes may not be enough to find the happy medium for game length. The 40-minute game is the easiest catch-all solution, and the league should consider it seriously. All indications suggest that the 40-minute game will be one of many such ideas to get a detailed hearing under Silver, and the sport will be better off if the league fosters thoughtful discussion of lots of different proposals. Sports and consumer habits, like anything else, are always evolving.