If you read a lot of hockey business stuff – and Lord knows I do – you know that the revenue and profit numbers published by Forbes get treated by the media as being about a half step below Gospel. There is a problem with this: there is absolutely no reason to think that their numbers are any good.
I’m vaguely aware that PR people talk about the need to get some sort of a message out when you’re a person who wants to control information, because an informational void will always be filled by something. I’ve never really believed that, or at least, never really believed that people that people won’t carefully evaluate information seeking to fill a void. Obviously, my cynical exterior conceals an utterly naive interior. Because man do people take these Forbes numbers seriously. And there have been obvious problems with them in the past.
The problem with refuting the Forbes numbers is that I have no better model to offer, simply the argument that an honest mess is better than a clean lie. I can’t point to sources and say “All of their numbers are wrong.” All I can do is look at some instances in which there has been information published and make an argument that it contradicts the Forbes numbers. I would suggest a test: if the Forbes numbers don’t pass muster in those circumstances where we can compare them to published information, we shouldn’t take them seriously where we can’t simply because there’s no better information available. An honest mess.
One of few saving graces of the 2004-05 lockout was the NHLPA’s inadvertent release of information that permitted the careful reader to assemble an ordered list of teams in terms of their regular season revenues. In their December 9, 2004 proposal, the PA proposed a revenue sharing mechanism based on regular season revenues. They went on to explain how this would affect the various NHL teams. (Side note: the PA proposal is hilarious because it repeatedly takes shots at the NHL for refusing to allow them to publish the revenue numbers and then inadvertently gives this away.)
The effect of this was to disclose where most teams ranked in the NHL in terms of regular season revenue in 2003-04 – the team with the highest revenue contributed the most to revenue sharing, the second highest contributed second most, etc. The NHLPA plan saw eleven teams contribute to revenue sharing, five teams (including Edmonton) which neither contributed nor received any money and fourteen teams that received money. It’s noteworthy that the Forbes numbers include playoff revenue while the NHLPA proposal was based on regular season revenue. To eliminate that as a factor, I’m just going to look at the teams that didn’t make the playoffs in 2003-04.
The second column is the Forbes revenue estimate for the year; the third is how that team would be affected by the NHLPA’s revenue sharing proposal, in terms of revenue lost or gained. Edmonton, neither paying nor receiving, sat somewhere between 12th and 16th in regular season revenues in 2003-04.
You can see that Forbes has to be significantly out of whack on a number of teams here. We know that Minnesota generated more regular season revenue than Los Angeles – Forbes says the Wild generated $9MM less. We know that Atlanta generated at least $8.8MM less than Edmonton; Forbes puts them $4MM ahead. That’s a massive swing – we’re talking about numbers that are off by at least 25%. We know that Phoenix generated at least $8.9MM less than Edmonton. Forbes puts them $2MM ahead. There are significant problems there.
(A brief digression: Twitter’s @woodguy, a knowledgeable guy when it comes to the Oilers, is adamant that the Oilers were prepared to walk away from the team if things didn’t go their way in the 2005 CBA. I am aware that there have been statements made by EIG guys to that effect. I don’t believe them.
If the environment that existed when they bought the team still existed, this was plausible. By 2003-04, not so much. According to Cal Nichols, the Oilers had revenue of CDN$85MM in 2003-04. I am assuming that he is not overestimating. With a 70 cent dollar, that’s US$60MM in revenue. My gut tells me that that’s a little low, given where they ended up in the NHLPA proposal but who knows. They reportedly turned a profit in 2003-04. They were spending something like US$34MM on players. They’d just seen teams with similar payrolls go to the Stanley Cup Finals. Also, owning a hockey team is a lot of fun, I’m told: see, for example, how much the EIG fought against the sale of the team when Katz started sniffing around. I have considerable difficulty believing that, by 2003-04, the EIG would have walked away from the team without the CBA they got.)
What about 2003-04′s Stanley Cup finalists? According to Forbes, Calgary generated $70MM in revenue while the Lightning generated $88MM. The NHLPA’s revenue sharing proposal would have seen Calgary receive $4.7MM while the Lightning received $1MM, so a difference of $3.7MM between the teams. Is it plausible that a playoff run that ends in a Cup win in Tampa Bay is worth US$14.3MM more than a playoff run that ends in a Cup loss in Calgary? It seems exceedingly unlikely to me.
What about the top end? Well, according to the Forbes, the Toronto Maple Leafs, who went two rounds in the playoffs that year, were second in the NHL in revenue, finishing $1MM behind the New York Rangers, who did not make the playoffs. The NHLPA proposal talks about one team (it doesn’t specify which team) that has revenues that dwarf even the second place team. You can infer from the proposal that this team had regular season revenues that were $19MM more than the team with the second highest revenues. You can further infer from the ordered list that that team was the Toronto Maple Leafs – they were the team that would have paid the most in revenue sharing and, therefore, the highest grossing team in the league.
Again, that’s a huge miss – according to Forbes, the Maple Leafs brought in $1MM in revenue less than the New York Rangers – counting playoff games – of which the Rangers played zero and the Leafs seven. According to the NHLPA’s proposal, the Rangers were more than $19MM back in regular season revenue alone. Again, they have to be out by more than 25% on someone here.
Want something more recent? I stumbled across a paper a few years ago written by a Professor Buser, who was given access to the audited books of the Columbus Blue Jackets – I wrote about it at the time. I’ve summarized his info, as well as that of Forbes, in the table at left. The takeaway is this: four of the six years, Forbes was 9.5% out or more and in every year, they overestimated Columbus’ revenue.
Want more? In 2006-07, the Oilers paid $2.7MM into revenue sharing. Cal Nichols was telling journalists in Nashville that Nashville could work: “It worked for us. The market size is very similar, and we went from being almost at the bottom of the league in revenue to seventh-best.”
According to Forbes, the Oilers were 18th in the NHL in revenue in 2006-07. Someone is wrong: either the guy with access to the Oilers books or the guys with a model. I know where my money is. It’s worth nothing as well that the Canadian dollar averaged about 88 cents for the 2006-07 season, applying the NHL’s method of calculation. The Oilers were seventh in revenue with an 88 cent dollar.
I suspect that Forbes has some sort of systemic bug in their model that causes them to underestimate Canadian team revenues and overestimate American team revenues. The problem seems to affect all of the Canadian teams, based on the various examples I’ve found. Their numbers roughly add up to the totals generated by the NHL, so it’s an allocation issue. The bio of Michael Ozanian, the guy at Forbes who writes this stuff for their site, reads as follows:
I like to take lots and lots of numbers and turn them into proprietary concepts and multi-platform content, such as the valuations of sports teams and ranking actors and movie studios on bang for the buck (ROI).
This is the problem with proprietary models – the assumptions can’t be checked and evaluated further. The brilliance of proprietary models that relate purely to information is that it can be difficult to prove that you’re wrong. If the proprietary software in my iPhone is garbage, I know because my phone sucks. If Forbes’ confidential model is screwed up, nobody can really be certain – at best, you can do something like what I’ve done, which leaves open the possibility that Forbes is right about everything else. As a result, Forbes fills the informational void and makes a pile of money.
I rarely, if ever, cite Forbes’ numbers on this site precisely because I’ve seen them contradicted by better sources too many times. Journalists who cover the NHL for a living should know better – David Staples, to pick just one, has said: “I assume that Forbes reporters are connected, sits down in New York with NHL number crunchers, and make sure their numbers are correct, as it’s in the NHL’s interest for these numbers to be correct. But that’s just my hunch.” It requires nothing more than paying close attention to that data which does occasionally come into the public sphere to conclude that seems very unlikely to be the case. Also, the fact that it’s not in the NHL’s interest to have this information published. Getting that right would help too.
It sucks not knowing precisely how much revenue each NHL team pulls in. Pretending that Forbes does, when the numbers we can check don’t check out, doesn’t fix that.Email Tyler Dellow at firstname.lastname@example.org