The Problem of Punishment
Intro to Hockey
To ensure that everyone can understand my point, I’ll provide some background on ice hockey. If you understand this hockey, please skip this section and call the Arizona Coyotes. They need some people who understand hockey. I probably don’t need hockey to explain my point, but I wanted to write about a certain rule even before I thought to connect it to other issues.
Let’s start with an NHL rink, which looks like this:
There’s a lot going on here, and I won’t explain it all. The most important action occurs in that light blue paint near the goal. After that, the most interesting activity occurs at the blue lines. These lines divide the rink into three zones: the attacking zone (where a team attempts to score), the defensive zone (where I team attempt to halt the opponent’s attack), and the middle “neutral zone.”
Every goal-scoring sport contains a rule to prevent an offensive player from “camping” near the goal. In basketball, offensive players may not station themselves in the painted area for more than three seconds. In soccer, players can’t pass to a teammate who’s body extends past the opposing defense. Hockey deters camping by preventing a player from entering the attacking the zone before the puck enters this zone. When an attacking player crosses the blue line early, the linesman will call the play “offside,” and play will reset from the red dots in the neutral zone.
This rule creates a few interesting dynamics. First, offensive players must decide how to legally enter the attacking zone. An attacker can try to skate into the zone with the puck, but this gives defenders the chance to impede the play. For a safer play, the attacker may “dump and chase,” meaning he throws the puck behind the goal and skates after it. The more exciting play, in my opinion, occur when a team attempts the first option. The offside rule also empowers the defense. Defenders can thwart the opposing attack by moving the puck into the neutral zone. This creates some of the fiercest battles in the sport, where the players from each team battle to move the puck to their preferred side of the blue line. If a defender successfully moves the puck to the neutral zone, all members of the offensive team must skate into the neutral zone to avoid an offside call.
The defensive has a few options in how to move the puck out of the zone. Ideally, the defense takes clean possession of the puck and begins their own attack. That’s hard, so often the defense will just try to knock the puck into the neutral zone. If they knock the puck all the way to the other end of the ice, the linesman will call them for “icing” and return the turn puck to the offensive zone.1 In other words, there’s no advantage smacking the puck down the ice, so teams try hit it lightly into the neutral zone. The last option for removing the puck from play involves hitting it over the boards and into the stands. This maneuver grew in popularity through the early aughts, and an annoyed National Hockey League (NHL) responded by banning it. Since the 2005 lockout, any defender who hits the puck into the stands from their defensive zone will be called for a “puck-over-glass” minor penalty. A minor penalty provides the opponent a two-minute “power play”, during which the opponent has about 20% chance of scoring a goal. Hence, it’s never advantageous to hit the puck over the glass, and players never do so on purpose.
Enough Background, Time For the Point
While the NHL bans both offside and puck-over-glass plays, these rules impact the game in different ways. With respect to offside, offenses rarely proceed with caution. If a team remains onside while entering the zone, they can proceed with their possession. If a player enters the zone offside, play stops, and the linesmen initiate a faceoff just outside the attacking zone. If the offense wins the faceoff, they can re-try a zone entry. If they lose it, the opponent remains backed into their defensive end. Given the high-reward, low-cost nature of the offside rule, teams often get called for several offsides in a game.
Meanwhile, the risk-reward tradeoff points the other direction for the puck-over-glass penalty. Knocking the puck out of play accomplishes nothing, since, rather than impeding the opponent’s attack, it initiates a more dangerous powerplay. One might therefore expect that puck-over-glass penalties rarely occur. On the contrary, it remains a moderately common penalty in the NHL. No one tries to throw the puck over the glass, but players inadvertently do so while attempting to pass or clear the puck. Thus, the NHL reaches an awkward equilibrium: no one would commit the penalty on purpose, so the rule only punishes those who violate in on accident. Though the NHL never intended to ban sloppy passes and zone clearings, it effectively does so through the puck-over-glass rule. The rule therefore discourages some passing or clearing attempts that would occur in its absence
Offside and puck-over-glass represent two distinct types of bans. Players put little effort into avoiding offside2 despite its ban. Meanwhile, the puck-over-glass rule discourages legal plays while punishing well-intentioned play. Imagine if NHL instituted a minor penalty for committing offside. In this case, teams would always dump and chase or enter the zone slowly. This would maintain a ban on offside, of course, but it would also ban the aggressive zone entries even if the league insisted that it wanted to maintain those plays. In any instance, the NHL doesn’t get to choose what it bans. The incentives make that choice.
In each case, have we can separate the “spirit” of the rule from its actual enforcement. In the offside case, the league attempts to prevent camping3 via a blue line that sits 64 feet from the goal. I don’t think attackers gain an unfair advantage if they’re one inch in front of the line when the puck enters the zone. We could eliminate the arbitrary blue line, but that would require the official to enforce some vague anti-camping rule. The same hold for the puck-over-glass penalty, which seeks to eliminate an annoying defensive tactic. In both cases, it helps to have some standard, even a somewhat arbitrary one, that the refs can enforce in a uniform manner. It would seem awkward to let the discretion of individual refs to determine the outcome of games. Yet, that’s how all nearly all the other rules in hockey work.
We allows refs a degree of discretion with regards to hooking, tripping, holding, interference, and all the other penalties. Any NHL fan knows that refs don’t enforce these rules uniformly game to game, or even within the same game. The refs also “let ‘em play” more in the post-season, which reduces the advantage of highly skilled rosters. In recent years, we’ve therefore seen less talented teams make the Stanley Cup finals, like under-.500 Canadiens and the 500-1 Golden Knights.
Why do we permit discretion for most penalties, but not for offside and puck-over-glass? Maybe it’s just a matter of necessity. In an ideal world, we would set an a specific standard for each rule, and allow the officials to call each game the same way. We don’t have to write a nebulous anti-camping rule; we can just draw big blue line on the ice. For slashing, on the other hand, we need can’t write a rule specific enough to remove all human judgement. I think there’s an additional element that allows us to accept this arbitrariness. Earlier, I argued that the puck-over-glass penalty effectively discourages otherwise banal passes and zone clearings. One could see this as a flaw, though the NHL added the rule as part of an initialize to increase scoring. More traditional penalties don’t suffer the same issue. We don’t mind discouraging actions that are just short of tripping, slashing or boarding since there’s little value added by a player almost boarding an opponent.
One final thought experiment: what if the NHL applied a weak punishment to penalties, similar to offside? I think basketball exemplifies such a system. Given the high scoring nature of basketball, an individual foul has much less impact than a penalty call does in hockey. Hence, we see dozens of fouls in every basketball game, and, in many instances, basketball teams foul their opponents on purpose. If hockey adopted a basketball-esque system, we’d see tons of penalties, resulting in a slower and more dangerous game.
I’ve discussed a few different thoughts here, so let me summarize
If violating a rule results in a weak punishment, we encourage some people to break the rule
If violating a rule results in a strong punishment, we discourage (and effectively ban) some behavior that’s adjacent to the punished behavior
Given the strong punishment for violating a rule, we assume that many of the violators will have broken the rule on accident
We need arbitrary human judgement to enforce some rules
This creates four different rule regimes:
Offside regime - clear rules, with a weak punishment for violators
Puck-over-glass regime - clear rules, with a strong punishment for violators
Slashing regime - judgement call, with a strong punishment for violators
Basketball foul regime - judgement call, with a weak punishment for violators
The Non-Hockey Point
In recent years, I’ve seen an uptick in normative debates around free speech. Free speech proponents often consider free speech it own end, without pointing to pragmatic or utilitarian concerns. I agree with this position, but I wonder if any dissenters find it convincing. I also don’t know how broadly to apply the mere principle, since I expect every social group and online community to engage in some moderation. While I rarely see people wholly oppose the principle of free speech, some want to carve out exceptions for hateful, bigoted, or otherwise low-value speech. Unfortunately, I don’t think hate speech rules will have effects that their advocates desire. Like the NHL, hate speech exception advocates (HSEA) don’t get to choose what they ban.
Let’s imagine we want to write a hate speech ban. First, we have to decide between a clear line (offside/puck over glass) or a judgement call (slashing/basketball foul). In the case of hate speech, I can’t imagine drawing a clear principle.
In some spaces, an offside regime works. Hobbyist communities, like the hockey subreddit, should ban any discussion on either side of the these controversial issues. Here, they can draw a clear line (e.g. any discussion of politics or identity issues) and delete all violations. Having one’s post deleted is also not a big deal, so it’s a safe policy. In larger institutions like universities, though, it's difficult to imagine a comparable regime.
I also worry that some people think they’re instituting a puck-over-glass regime, when they’re really instituting a slashing regime. I’m not usually a “where do we draw the line” guy, but can’t let M.C. Escher draw the line.
Board Games Again
I also want to return to the topic of colonialism in board games. I promise, this will be the last time. Even the most ardent free speech defenders agree that no one has the right to discuss board game theming in four consecutive articles.
As social awareness surrounding colonial themes increased, the industry developed an offside regime. Recent hits have set their game on desert island, like Lost Ruins of Arnak, on Mars, like, well, On Mars. Another hit, Spirit Island, tasks players with fighting off a colonial invasion. You won’t find a rulebook in a recent game that ends with “and the player with the most conquered subjects win,” though some minor social insensitivities probably remained. However, I feel as if we’re moving towards a slashing regime. The aforementioned Lost Ruins of Arnak, for example, received criticism for erasing colonialism by setting itself on an uninhabited island. I thought erasing colonialism was the thing you’re supposed to do!
In a slashing regime designers will want to avoid any behavior that could violate a broad set of a cultural norms. To do so, they might avoid setting works outside “safe” settings, leading to more games about Vikings, Romans, and medieval townsfolks. For people who enjoy diversity, myself included, represents a devolution of the hobby. I want to see more games that take me to new cultures and time periods. I love Castles of Burgundy, for instance, but I find the theme pretty drab. I would have preferred Large Architectural Structures of Literally Anywhere Other than Ancient Rome, Renaissance Europe, or Modern Europe and North America. If designers fear a large backlash for breaking an ill-defined ruleset, however, we’ll never see, LASoLAOtARREMEaNA, despite its catchy title. We’ve seen this occur in other media. YouTube creator Sarah Z, for example, noted that some authors de-diversified their books (i.e., made all the characters straight white people) to avoid progressive backlash.
In the same vein, I saw LinkedIn post warning users against committing “cultural appropriation” on Cinco de Mayo. The poster warned against wearing a Sombrero, considering it offensive to Mexicans (news to every Mexican I’ve met). It also asked users to refrain from Maracas as those come from Africa. Luckily, I don’t consider this a slashing regime. Few people care about, or even understand, the notion of cultural appropriation. Still, I fear a burgeoning slashing regime and the harm it would do the relationships between different ethnic groups. I don’t think of Mexican culture foreign in the first place, given it’s prevalent role in the American southwest. When I celebrated Jewish holidays in my youth, I never considered Jewish culture separate from “American culture,” and I never placed my Jewish activities in a dominant/minority culture framework. If this idea gains power, people will avoid ideas and individuals from other cultures, out of fear of violating some unwritten rule.
Coda: Torture
I don’t think this fit into the rest of the article, but I thought I’d discuss torture. During the War on Terror, the United States waterboarded4 alleged terrorists. The law unambiguously banned torture, but government officials avoided this ban by arguing that waterboarding wasn’t torture. I find this claim ridiculous, but it’s probably not even a debate that should occur in the first place. Ideally, the punishment for torture should be so strong that it discourages torture-esque practices. In other words, we should lean towards accidently banning non-torture rather than accidently allowing torture. Here, the “banning too much” element of a slashing regime can work in our favor.
Anyway, I’m tired of social criticism; next article will be an introduction to linear regression.
Icing is a bit more complicated than that, but I’ll leave the details out for now.
I’m simplifying a bit. Players put a lot of practice into timing their zone entries just right. I only mean to say that offensive players don’t slow to a halt to avoid an offside call.
Offside does a bit more than camping, as it also allows for the defense to clear the zone. Regardless, I think the point holds.
I’m using past tense optimistically.
Klaus - excellent analogy. May I borrow it for a hockey use? I think it would inform my current discussions on hockey forums concerning head hits in the NHL.
In fact, I want to just borrow it to have on hand because it's an elegant way to explain risk mitigation and unintended consequence.
I loved this for so many reasons. 1) hockey 2) your explanation of types of regimes 3) the seeming randomness of topics that in fact make total sense to me.
Looking at this through a sports rule lens is pretty brilliant because it forces us to think through consequences devoid of the emotional and intellectual baggage around the topic you’re really getting at: free speech.
So you’re from the southwest? Me too! Grew up in LA, my grandparents lived in Albuquerque, had cousins in Arizona. I feel the same as you about Mexican culture. It’s not *my* culture but it’s also not some foreign unknowable thing. Just like I didn’t grow up Jewish but I had a good friend who was Jewish so her holidays and rituals weren’t an alien occurrence. I didn’t realize you were Jewish. Cool!
Also how do you feel about Auston Matthews?