Goal scoring in the NHL is, as you may have heard, a lot harder than it used to be. In the early 1980s there were roughly 8 goals scored per game. That number fell to about 7 by the early 90s, then down to just 5.3 in the late 90s, as the clutch and grab era slowed the game to a standstill. The obstruction crackdown at the start of the 2005-06 season brought goal scoring temporarily back up to 6 goals a game, but as powerplays have declined, the number has rested relatively consistently around 5.3 GPG in recent years.

Most hockey fans seem to think that number is too low. Lots of people inside the game seem to think that too, and there are a bunch of suggestions floating around on how to increase scoring. In recent years the NHL has often tinkered around the edges with things like faceoffs or smaller goalie pads, but they haven't been willing to make more substantial changes.

The tide seems to be turning in favour of bigger changes, however. Toronto Maple Leafs head coach Mike Babcock told reporters last week that he favoured increasing the size of NHL nets. "The net's too small for the size of the goalies," Babcock said.

I'm not necessarily opposed to increasing the size of the nets, but there's a simple change that the NHL could make the would increase scoring without altering the game on any fundamental level. How? By making teams have the long change in all three periods, instead of just the 2nd period as is the case now.

What Is the Effect of the Long Change?

There are two elements to the effect the long change has on scoring. The first is on the frequency with which goals are scored. I pulled some numbers from the three seasons since the last NHL lockout. Here are the per-60 minutes goal rates in each of the periods in various manpower situations (all stats used in the remainder of this post were pulled from War On Ice or calculated by me using WOI numbers).

Period 5v5 G/60 PP G/60 SH G/60
1 4.12 12.85 1.40
2 5.01 13.58 1.55
3 4.35 12.74 1.61

A few things are evident here. Scoring goes up in all situations in the 2nd period relative to the first. The 3rd period sees a boost over the 1st period as well, though much smaller. And the biggest boost, by far, is to even strength scoring.

The fact that 3rd period scoring is higher than 1st period scoring presents a potential complication to my efforts, but I'll get to that below.

The other element to the scoring boost in the 2nd period is that there are more penalties called. Here are the relative ratios of ice time in 5v5, PP, and SH situations in each period:

Period 5v5 PP SH
1 0.83 0.08 0.08
2 0.80 0.10 0.10
3 0.84 0.08 0.08

There are more penalties called in the 2nd period, which boosts special teams scoring. This is likely for the same reason scoring rates go up: players have a harder time changing, making them more tired or prone to mistakes (or bad line changes).

How Much Would This Actually Boost Scoring?

There are a number of different ways to calculate this, so I'm going to provide a few of them to give you a possible range of results. All of these results assume the 2nd period totals will remain unchanged relative to now, since that period already has the long change.

On the low end, if I calculate the difference between the 1st/2nd and 2nd/3rd period based on their own baselines and assume no change in special teams time, the boost to scoring would be 0.48 goals per game.

In the middle, if I ignore the boost to scoring in the 3rd period, which is likely caused by score effects, and simply treat the difference between the 1st and 2nd periods as the "true" effect of the long change, while still not changing special teams time, the boost to scoring would be 0.54 goals per game.

At the high end, if I take scenario 2 but also change the ice time ratios in the 1st and 3rd to match the boost to special teams TOI seen in the 2nd period, the boost to scoring would be 0.60 goals per game.


There are other possible effects that I'm not able to account for. As one example, perhaps penalties would increase more dramatically in the 3rd period, as players become even more tired. You might argue that coaches will find a way to neutralise the effect, though I'm pretty skeptical of that, since they haven't been able to do it in the 2nd period yet and there isn't a lot you can do about the physical dimensions of the rink.

But taking into account the information I can reasonably know right now, I get a possible range of 0.48 to 0.6 extra goals per game. That might not sound like very much, but I think it's actually reasonably substantial.

0.6 extra goals per game would bring the total to somewhere around or a bit above 6 GPG league-wide. That number is in line with NHL scoring from the late 50s to early 70s, as well as right around where the NHL was in the first post-2005-lockout season, which most people seemed to be OK with. It would be 738 extra goals scored over the course of a full season, about 25 more per team.

Maybe for some people that wouldn't be enough. Some fans might want to get closer to the 7 GPG of the late 70s/early 90s, or even the 8 GPG of the early 80s. But before the league makes major changes with unknowable impacts, like increasing the size of the net, I think it's worth pursuing this very simple idea that has a relatively quantifiable likely outcome, especially since it won't change anything about how the game is played.


There are two main criticisms of this idea that I'm aware of, so I'll try to address both of those before closing this post out.

The first criticism is that the NHLPA would never sign on to it. More long changes would result in players being more tired, and could lead to an increase in injuries. I don't know of anywhere that tracks injuries by period, so there's no way for me to test that claim, but I have to admit that an increase in injuries would be a legitimate concern from the PA.

That said, any major change to the rules would go through the Competition Committee first, which has player representatives on it, so you're going to have to get the players on board with anything more substantial than minor tweaks to faceoff rules.

The other criticism I've heard is that fans who buy tickets to see games live aren't going to like the idea that, depending on what side of the rink they're sitting at, they'll only ever be at the end of the rink that one team shoots at.

My primary response to this would be that the increase in scoring should make the game more fun, and that it's a reasonable trade-off to make.

One thing you could to alleviate this concern would be to designate the far ends of each rink for ticket sales as "home" and "away", to take advantage of the fact that fans of the home team are more likely to want to see the end their own team is shooting at anyway.

You could also have players swap benches in the 2nd period so that you could have them switch which net they're shooting at, but still have the long change. I know some NHL arenas aren't necessarily designed to accommodate that at the moment, but it's a relatively simple change to make.

In the end, though, I think it's a pretty easy change to make that would likely have a pretty big impact on scoring, so it's worth pursuing and figuring out the details along the way.