clock menu more-arrow no yes

Filed under:

The reports of the NHL's goal-scoring woes have been greatly exaggerated

New, comments

You've probably heard about the lack of goals being scored in the NHL this season. You've been misled.

Joe Camporeale-USA TODAY Sports

With the goals-per-game average going down in the NHL, there is what can almost be perceived as panic to increase goal-scoring in the league. I chose to take a deeper look into goal-scoring in the NHL as well as the other three leagues that can be considered as the top tier of the international hockey world, to see how the goals-per-game average had changed over the last few years.

Comparing Finland's Liiga, Sweden's SHL and the pan-European KHL over the last five years along with NHL numbers should help see what is going on in the world of hockey more in a general trend than what might just be an NHL trend.

NHL the last ten years

We have all been used to seeing the below goals-per-game graph now, telling us in certain terms that GPG has gone down from 6.20 to 5.34 as of Friday, November 27. The decline in GPG for the NHL looks very dramatic indeed with these numbers:

GPG NHL last 10

No one refutes this graph; there is a general decline of goals. But the problem is in how to interpret the data.

The mean goals per game in this period is 5.63 with a standard deviation of 0.24 (this would give us a maximum value of 5.87 and a minimum value of 5.39 to be within the confidence interval. Now let's see how many years are outside this confidence interval: 2005-06 (6.17), 2006-07 (5.89) and 2015-16 (5.34).

Being only a quarter of the way into the season, the NHL is just 0.05 GPG from being within the confidence this year. The 2005-06 season used as a baseline is much further away from being within the confidence interval of the period. That means that to use 2005-06 as a baseline is definitely not the right way of approaching this "goal-scoring drought."

Removing the three outliers gives us a low variation graph looking like this, showing that, for eight years, the NHL has been very stable indeed:

GPG no Outliers

There are three outliers during the ten-year span, and the last outlier isn't even a full season yet. The normally used 2005-06 campaign is just one season, not a plateau, and not a stable secure value. This begs the question: why was 2005-06 chosen rather arbitrarily as the baseline?

NHL within the big four

2011 2012 2013 2014 2015 2016 Mean Standard Deviation
NHL 5.59 5.47 5.44 5.49 5.46 5.34 5.47 0.08
KHL 5.57 5.28 5.46 5.02 5.07 4.84 5.21 0.28
SHL 5.25 4.89 4.79 5.28 5.11 5.94 5.21 0.41
Liiga 5.25 5.46 5.28 4.94 5.00 4.78 5.12 0.25


Within the group of the big four, the NHL has the highest goal expectancy per game for the six-year period included, and has been the top league for all seasons except the current one, which the SHL is claiming with almost six goals per game.

We can also see here that the NHL does not move much at all: it's quite stable around 5.5 goals per game, especially compared to the dips in goal production in the KHL and Liiga, with an obvious outlier this season for the SHL. Something that becomes even more apparent when plotting the leagues against each other:


GPG Graph

The dip in goal production for the NHL is marginal in comparison to the decline in the KHL and Liiga, who, after the 2012-13 season, follow each other closely. The NHL has a much lower standard deviation than any other league, at 0.08, which in this case is next to nothing, and points to the fact that the GPG is actually stable. This is especially true when compared to the massive 0.41 of the SHL, and the KHL and Liiga with 0.28 and 0.25, respectively.

There is a small decline in the goals-per game production in all leagues of the world, but the fact is that so far in the 2015-16 season the goal production in the world is trending upwards from 5.16 to 5.23 obviously driven by the situation in the SHL. But it is not an extreme situation by any means, and this is a process that we should not rush in to counter with a quick-fix band-aid but with intelligence and caution.

What happened in Sweden this season to make the goals per game go from 5.11 to a 5.94? The easy explanation is that the Swedish league went from 12 to 14 teams, and it seems the new teams were not as prepared for the top division; judging by the standings table, the three promoted teams and the team that avoided relegation last season occupy the bottom four positions.

The other thing that has happened is that the power play success has increased from an average of 17.7% to 20.7%. So early on, it remains to be seen if this will be sustained over the season, and if it continues next season as well, but for now the SHL has become the highest-scoring league of the big four, this after having been the lowest-scoring league just two-and-a-half years ago. There have been no drastic rule changes to facilitate this sudden spike in goals, either.

Why the panic?

Why is the NHL panicking now, with almost three-quarters of the season remaining? Is it because of the empty seats in the arenas, such as Carolina (something Raw Charge touched upon in this article by John Fontana) that makes it bad for business? Is it because of the possible expansion with Quebec and Las Vegas hovering around the corner (and trying to bait Seattle to put in a bid as well)? Do goals really sell a product? How come the North American sports media is so focused on this perceived issue, creating a storm in a water glass? How come they don't present the stats in a proper way, instead using extreme outliers to sell a version that isn't really true?

Also, shouldn't the trend in global ice hockey be considered? If we talk about the economy, don't we look at the global economy parallel to the economy in our own country? Compared to global ice hockey, the NHL's goal production is stable. You could argue that it's a general decline in the hockey world, but every year the NHL is over the average goals per game compared to the others. Globally, the average is 5.2 goals per game, and has been for the last five years.

Compared to international hockey, the NHL's goal production is stable.

Those in the public eye who have been pushing for change have something to answer for when choosing to use one season (2005-06) and the maximum value for goals production as a baseline for something as important as this, especially as they are considering changing equipment and potentially the rules of the game.

In many ways, the NHL gets the best players in the world, be it defensemen, forwards, or goalkeepers. It is also often easier to shut down an opponent rather than totally outplay them. If coaches take the approach to limit the opportunity for the opposition, then that is the reason for a decline, not the padding on the goalkeepers nor the size of the goals behind them.

Do more goals increase attendance?

The big question is revenue though, and has the large increase in goal production been matched by a rise in attendance in SHL?

Checking home attendance for the teams that have been in the top tier for the last two seasons, the sum of the average home attendance for those teams has actually gone down about 1000 total spectators so far (mostly from lower fan support for Djurgården and the struggling Modo). Last season's numbers included playoffs and relegation rounds in the average, so it will probably even out in the end, or reverse course to a slight increase, but probably not by much.

This would contradict the perception that a "better product" (i.e., more goals) automatically means greater attendance and, in the long run, more revenue. This is obviously a small sample size, and should be treated as such, but clearly more goals does not look to have an immediate effect on attendance in a positive way.

The global economy has gone down, the average global citizen has less money in the pocket. Given the current price of tickets at an average NHL game, maybe it is time to consider that if there is more people there, it is likely that concession revenue will increase, it could therefore be worth making the tickets cheaper to draw in more revenue elsewhere. It would also look better on television with a big crowd in the arena.

Maybe the NHL should look into something other than just a quick goal fix — and the massive alteration to the game those changes would effect — instead looking in the mirror to see what's really wrong with its business model and discovering a better way to fix attendance. If the consumers don't want to pay for the product, maybe it's time to look into the cost you're asking them to pay.

The important question here is why the media just took up the NHL's position. Journalists are supposed to research independently and find facts, not run a PR service for the entities they cover. Hard questions need to be asked, not about the size of the nets or goalkeeper equipment, but about the application of mathematics by those in the media who haven't done their homework (since seventh grade math it seems) when it comes to the use of the statistics they have been so quick to report upon.