« The Liberaltarian Class Action | Main | Saving the Senate, the Court, and the Country »

Thursday, April 06, 2017

The solution to late-game fouling?

No one likes late-game intentional fouling in basketball, because it drags out games and produces boring basketball of constant stoppages and endless free throws. On the other hand, there is no way around the strategy, as it reflects the only way that a trailing defensive team can save time and get the ball back.

But it appears Nick Elam, a middle-school principle principal and MENSA member from Dayton, has a solution: In the final three minutes of the NBA game (final four in college), turn off the game clock and play until either team reaches a target score, set at +7 from the leading team's score when the clock is turned off. So if Team A leads 99-91 when the clock goes off, the teams play to 106. Elam has been sending his proposal around to basketball types, some of whom purportedly find it interesting, but too radical to implement just yet. But it is going to be used in the early rounds of The Basketball Tournament, a $2-million 64-team tournment featuring teams of former college players. (Elam is interviewed on the tournament podcast).

The proposal does eliminate any incentive to take fouls at the end of the game, because a trailing team can simply play good defense without having to worry about preserving time on the clock. The only fouls we might see are to stop a three-pointer, although that strategy is so time-sensitive (it only works under :04 or so) that it might dissolve on its own. Eliminating the game clock somewhat changes the nature of the game somewhat, which is played in a rhythm of time, but not as much as soccer shoot-outs or college football overtime. And the shot clock remains, so there still is a time element to keep possessions and the game moving.

The proposal may not succeed in shortening games and might lengthen them--not because the clock is stopping constantly, but because teams are not scoring. This will be especially true in close playoff games, where the defense ratchets up in the final minutes. For example, at the 3:00 mark of Game 7 of the 2016 NBA Finals, the score was 89-89, meaning the game would have been played to 96. The final score was 93-89, and one of those points came on a made free throw off an intentional foul with :10 left. The defense was that good and the players were that tired (this included LeBron James's block of a fast-break layup).

On the other hand, perhaps offenses would be freer to look for the best shot at anytime, no longer worried about any time considerations. Teams now get as many possessions as it takes to score the requisite points, so they need not save or waste time. Back to Game 7: After Cleveland's Kyrie Irving hit a go-ahead 3 with :53 left, Golden State used almost the entire shot clock to get Steph Curry isolated on a weak defender, who forced Curry to miss a three-pointer. But Golden State does not need a three in that situation; it can get a better two-point shot, knowing that, if it plays good defense, it will have a greater number of possessions and opportunities to score.

Posted by Howard Wasserman on April 6, 2017 at 09:37 AM in Howard Wasserman, Sports | Permalink


Why have a clock at all? Games have rules and boundaries in order to reward the most skilled team for a period of time within those rules and boundaries. There is an inherent arbitrariness to rules because that is what humans like to do--impose arbitrary rules to make things more exciting and challenging. Not every rule can be justified on efficiency or other grounds. Some just make the game fun because quirks add wrinkles to the achievement of success. Every rule could be subject to a searching "why do we put up with this analysis"...

Which brings me to the next point: too often it is the "we" that craves change rather than the game itself that needs to be changed. Changing the end-of-game rules to alleviate burdens on the trailing team (the difficulties of coming back) due to the superior play of the winning team, within the rules for the previous 37 (or 45 if we are talking about the NBA) minutes fails to reward the advantages accrued by the winning team due to superior play. Altering the end of game scenario without considering fully how much weight we should afford to the winning team's success over the previous 93% of the game most definitely (and arbitrarily--this is inspired by our shortening attention spans and inability to appreciate sustained success in athletics--see e.g., calls for shortening baseball games, amongst other similar proposed reforms) diminishes the situation of the winning team. The proposal's argument that the results of fouling (foul shots, inbounds plays, resetting possessions, etc) are not basketball is remarkably short-sighted. Surely those aspects of the game are basketball--but I suppose that depends on how one defines the game...

Changing rules such as these under the guise of interest is misleading and fails to appreciate how those rules interact with the rest of the rules in the game--which the winning team have navigated well enough to achieve an advantage worthy of a reward that is not subject to the whims of excitement that comes with fickle fan base. What the proposal is really saying is this: we value live possessions more than other aspects of the game. That is *not* controversial to the average fan, and accurate. But, it begs the question: should those *other* aspects of the game remain at all?

Posted by: Brian | Apr 10, 2017 9:49:49 AM

Upon further reflection, this gives up too much of what makes the end of basketball games potentially great. One Valparaiso or Duke-Kentucky ending makes up for thousands of less than compelling foul-fests, and this proposal eliminates that possibility.

Posted by: jph12 | Apr 9, 2017 5:27:41 PM

If the game clock isn't a meaningful part of the game, why not just get rid of the game clock all together and play first to 80 (if you want to shorten games, 100 if you want to make them longer) like a pick up game? You can have breaks at 20, 40, and 60 points.

Posted by: jph12 | Apr 7, 2017 4:48:03 PM

Hash, it is a sliding scale surely, with fouling making sense in fewer situations as the shot clock gets shorter. Even making it 20 instead of 24 seconds might mean that with 25 seconds on the game clock and trailing by 2 points it is better to let the opposition shoot than foul them and send them to the line.

Howard: Was it not 400 games where intentional fouling was tried and it nearly never worked? That sounds like a respectable sample to me.

Asher: It is actually interesting to consider whether it will make outcomes more or less random. I am thinking that theoretically it might make it less random, since the trailing team gets a chance to win but only if they play better in the final part than the leading team. We are actually testing the skills of the teams for a longer period you could argue.

Posted by: Jr | Apr 7, 2017 6:25:27 AM

Howard -- I still don't understand why you don't think shortening the shot-clock wouldn't solve, or at least mitigate, the problem. Take an extreme case: if the short clock was reduced to five seconds for the final five minutes, there'd be no reason for the team behind to foul; they'd be better off playing defense for five seconds and potentially forcing a miss than fouling after one second just to gain four seconds but likely lose two points. In other words, the fact that they'd get many more possessions reduces the incentive to foul, because the lost points become more harmful than the lost time. Now, obviously the reduction in the shot clock has to be material: reducing it to 20 seconds would still have the same problem, because the four-second speed-up wouldn't generate enough extra possessions to be worth it. And I've got no clue where the right tipping point in between 20 seconds and 5 seconds is. But, mathematically, I'm pretty sure there is one.

Posted by: Hash | Apr 7, 2017 3:24:09 AM

Why speed up the end of the game? Personally, my favorite games are the nailbiters where it comes down to the last few minutes that just seem to go on and on. That's what makes a game exciting to me, much more so than a humdrum blowout.

Posted by: Facepalm | Apr 7, 2017 2:17:44 AM

It isn't about making comebacks easier--it's about making them possible. It would be next to impossible for a team to come back when the game clock moves below the shot clock and probably even in the final 1-2 minutes.

How does this structurally advantage the trailing team? It seems to me it is easier and quicker to score 7 points than to score 16 points.

Posted by: Howard Wasserman | Apr 7, 2017 1:03:24 AM

Anothe source of major resistance you'd get is that you could no longer do spread betting. Vegas would never let it happen!

Posted by: Anon | Apr 6, 2017 10:35:19 PM

I thought you said that "[n]o one likes late-game intentional fouling in basketball;" I assumed you were at least including yourself.

I also see no reason why we should make comebacks easier. A first to leading team's score + 7 game, where time is unlimited, structurally advantages the trailing team and would seem to make the outcomes of basketball games considerably more random and thus less meaningful (if you think, as I do, that the results of athletic contests are meaningful insofar as they're revelatory of which competitors are superior).

Posted by: Asher Steinberg | Apr 6, 2017 8:30:25 PM

There are three goals: 1) Speed up the end of the game; 2) Make the final minutes more fun to watch; and 3) Still enable the trailing team to come from behind. The only way that can happen is if the trailing team has some way to stop the clock and get the ball back. Under current rules, fouling is the only way to do that.

This proposal achieves ## 2 and 3 by taking the clock out of the mix (there is nothing to stop by fouling) and requiring the leading team to score more points, which incentivizes it to shoot. So now the trailing team only has to play good defense and rebound against a team that is going to try to score. And the game has a normal flow and is more fun to watch.

It is not clear whether it will achieve # 1, for the reasons I described in the original post. The only way to find out is to implement the change and see.

Posted by: Howard Wasserman | Apr 6, 2017 5:49:28 PM

I don't think this is a fix. If the goal is to speed up a game, then the rule should just be that every foul committed within the last 3 minutes is a technical on the coach.

Posted by: YesterdayIKilledAMammoth | Apr 6, 2017 5:25:47 PM

I don't either. My introduction to basketball was NC State's run to the 1983 NCAA title, where it won at least two or three tourney games, including the final, thanks to late-game fouling (back in the days when college only had 1-and-1s).

Posted by: Howard Wasserman | Apr 6, 2017 4:53:21 PM

I don't dislike late-game fouling. Late-game fouling has helped create a lot of riveting late-game finishes, it can be fun when good free-throw shooters miss or when a poor one is forced to shoot free-throws, and I think proposals of this kind, like the elimination of the pitched intentional walk, overlook the tradition and memory that inheres in any aspect of a sport that's been around for a while, whether or not it would be the most enjoyable way in which to design the sport as an original matter.

Posted by: Asher Steinberg | Apr 6, 2017 4:50:01 PM

Hash: The incentive is still to foul as quickly as possible. That the shot clock is :14 rather than :24 does not change my need to foul after two seconds.

Jr: Note that Elam acknowledges looking at a very small sample size--about 400 televised games since 2014. It does work sometimes, if only through the 3-for-2 possibilities.

Posted by: Howard Wasserman | Apr 6, 2017 3:40:43 PM

jph12: It should dilute the incentive to foul for sure. The likely effect of fouling is to make you go further behind, and unlike the end of the game today, where it does not matter how much you lose by you, when the untimed portion begins it is easier to comeback the less you are behind.

Elam claims to have research showing that intentional fouling very rarely works to close the gap so in your example the trailing team are probably better off playing normally, hopefully closing the gap somewhat before the game clock expires and then outscoring the leading team in the untimed period to win.

Posted by: Jr | Apr 6, 2017 2:49:24 PM

Couldn't this lead to the trailing team treating the three-minute mark as if it were the end of the game and committing fouls so that the score is as close as possible when the clock goes away? In other words, if I'm down 10 with five minutes to play, why wouldn't I spend the next two minutes fouling in an effort to close the gap? You would have to be a little more careful to manage the foul situation, but I'm not sure it eliminates the incentive for late game fouls as much as shifts it.

Posted by: jph12 | Apr 6, 2017 1:55:13 PM

Why not just shorten the shot clock in the final x minutes of the game?

Posted by: Hash | Apr 6, 2017 1:17:06 PM

ADD: Your solution is what the NBA adopted for Hack-a-Shaq, at least in the last two minutes of the quarter, with the additional wrinkle that the offense can choose any player to shoot the free throws.

Posted by: Howard Wasserman | Apr 6, 2017 11:48:33 AM

Oops. I always make that mistake.

Curious: Late-game fouling is different from Hack-a-Shaq. We need a way for trailing teams to be able to extend the game and get the ball back when the leading team with the ball has no incentive to shoot. It's also not intentional in the same way as Hack-a-Shaq, because the foul is always on the person with the ball and usually in the guise of trying for a steal.

Posted by: Howard Wasserman | Apr 6, 2017 11:47:25 AM

You have a typo: "middle school principle"

Posted by: gdanning | Apr 6, 2017 11:07:21 AM

Genuine question: Why is this fix superior to having an intentional foul result in two free throws and retained possession?

Posted by: Curious | Apr 6, 2017 10:35:21 AM

The comments to this entry are closed.