Is the NHL Really More Popular than the NBA?
by Jes
Over at The FanHouse, we were sent a link to something called "The Harris Poll", which asks a bunch of (American) people what their favourite sport is, and then ranks them, plays around with the data, and shows us just how popular the boring NFL brand of football really is.
What made our eyes pop out was the fact that 5% of respondents claimed hockey as their favourite sport, while only 3% claimed professional basketball, aka The NBA, as their favourite sport.
'Sup, bitches?
The NHL is more popular than the NBA?? WTF??
Of course, this news is far too good to be true, as any set of TV ratings will tell you.
As with any poll, this instantly screams "SMALL SAMPLE SIZE!". 2032 adults is hardly going to capture the true tastes of the general American public.
Given how regional NHL interest is, the figures could wildly fluctuate if you hit a good patch (Minnesota) or bad patch (West Virginia) when polling people.
It's always been a pet peeve of mine that the media, and the pollsters themselves, seem to forget a simple little concept called SAMPLING ERROR.
In statistics, sampling error is the error caused by observing a sample instead of the whole population.
An estimate of a quantity of interest, such as an average or percentage, will generally be subject to sample-to-sample variation. These variations in the possible sample values of a statistic can theoretically be expressed as sampling errors, although in practice the exact sampling error is typically unknown. Sampling error also refers more broadly to this phenomenon of random sampling variation.
For example, say a recent poll asked 2,000 Canadians which flavour of ice cream they like the most. The first poll had a +/- margin of error of 3%, with 38% choosing Vanilla and 23% choosing Chocolate.
Now, the second poll shows 35% choosing Vanilla, and 26% choosing chocolate.
If the media were covering the poll, the headline would be like "Support for Chocolate growing!" or "Vanilla Losing Popularity".
Nevermind the fact that the growth/drop was 3% for each, which was the exactly sampling error, the media seems to think that any such movement is indicative of the mindset of the population, rather than the expected variation from one poll to the next.
Just remember, kiddies, that any movement that is within, or very close to, a margin of error on any poll does not necessarily indicate a change in public opinion, as a whole.
Of course, the Harris Poll doesn't even bother publishing a NUMBER for sampling error, but at least they acknowledge there is a such thing and give their reasons for not putting a hard number out there.
Still, I did enjoy the poll, and would love to see a similar poll done with a far large sample size. Perhaps ESPN, with all of their resources, could conduct a poll through their website.
Anyway, back to the original topic at hand ... Big Mac took the reigns of this story for us at FanHouse, and made a good point about how the NHL really is gaining ground to the NBA.
The numbers for the NHL are pretty much flat throughout the survey. It's the NBA's numbers that have really dwindles, as the American audience seems to be turned off by the new generation of American 'ballers.
Over at The FanHouse, we were sent a link to something called "The Harris Poll", which asks a bunch of (American) people what their favourite sport is, and then ranks them, plays around with the data, and shows us just how popular the boring NFL brand of football really is.
What made our eyes pop out was the fact that 5% of respondents claimed hockey as their favourite sport, while only 3% claimed professional basketball, aka The NBA, as their favourite sport.
The NHL is more popular than the NBA?? WTF??
Of course, this news is far too good to be true, as any set of TV ratings will tell you.
As with any poll, this instantly screams "SMALL SAMPLE SIZE!". 2032 adults is hardly going to capture the true tastes of the general American public.
Given how regional NHL interest is, the figures could wildly fluctuate if you hit a good patch (Minnesota) or bad patch (West Virginia) when polling people.
It's always been a pet peeve of mine that the media, and the pollsters themselves, seem to forget a simple little concept called SAMPLING ERROR.
In statistics, sampling error is the error caused by observing a sample instead of the whole population.
An estimate of a quantity of interest, such as an average or percentage, will generally be subject to sample-to-sample variation. These variations in the possible sample values of a statistic can theoretically be expressed as sampling errors, although in practice the exact sampling error is typically unknown. Sampling error also refers more broadly to this phenomenon of random sampling variation.
For example, say a recent poll asked 2,000 Canadians which flavour of ice cream they like the most. The first poll had a +/- margin of error of 3%, with 38% choosing Vanilla and 23% choosing Chocolate.
Now, the second poll shows 35% choosing Vanilla, and 26% choosing chocolate.
If the media were covering the poll, the headline would be like "Support for Chocolate growing!" or "Vanilla Losing Popularity".
Nevermind the fact that the growth/drop was 3% for each, which was the exactly sampling error, the media seems to think that any such movement is indicative of the mindset of the population, rather than the expected variation from one poll to the next.
Just remember, kiddies, that any movement that is within, or very close to, a margin of error on any poll does not necessarily indicate a change in public opinion, as a whole.
Of course, the Harris Poll doesn't even bother publishing a NUMBER for sampling error, but at least they acknowledge there is a such thing and give their reasons for not putting a hard number out there.
Still, I did enjoy the poll, and would love to see a similar poll done with a far large sample size. Perhaps ESPN, with all of their resources, could conduct a poll through their website.
Anyway, back to the original topic at hand ... Big Mac took the reigns of this story for us at FanHouse, and made a good point about how the NHL really is gaining ground to the NBA.
This survey does dovetail well with some other data that we've seen about the relative popularity of the two sports. Back in November, Sports Business Daily published a study about the most popular sports sites on the Web, and that survey pegged the total traffic of NBA.com and NHL.com as pretty much even. Later, I called the league office in New York, and they told me that had the numbers from Canada been included, that NHL.com would have been far more popular.
Of course, once you add in those numbers from China, the NHL suffers a bit. The global village and all that.
If I was working for the NHL, I might take some time to send these numbers to a couple of sports pages around the country to make a simple case: You might be ignoring the NHL, but there's a market of people out there that you aren't necessarily serving. You'd think there might be some potential in that.
The numbers for the NHL are pretty much flat throughout the survey. It's the NBA's numbers that have really dwindles, as the American audience seems to be turned off by the new generation of American 'ballers.
posted by Unknown at
12:10
0 Comments:
Post a Comment
Subscribe to Post Comments [Atom]
<< Home