THE BOOK--Playing The Percentages In Baseball

Phil is exactly right, and Zimbalist is way off. But this seems like an irrelevant question—in the end, either the NHL owns its territorial and ownership rights or it doesn’t. Whether Balsillie was willing to pay $212m or $512m shouldn’t matter, nor should the value of the team if it were moved. We know Balsillie is willing to pay more than anyone else, regardless. What’s being challenged is whether the NHL has the right to block a team from being sold/moved.

Didn’t the Oakland Raiders settle this issue already?

Phil writes: “Why are the values so flat relative to profits, where a team that makes $1 million a year is worth almost half as much as a team that makes $40 million a year ... What that tells us is that, regardless of earnings, there’s a value of $200 million dollars to owning a team, even if it only breaks even every year.”

Actually that $200m probably represents the book value of the company’s assets. If you run a regression on a large cross-section of stocks to predict their current share prices based on a variety of accounting variables, a really good equation is something like: (book value per share x .60) + (next-year earnings x 10.5). This is the residual income approach to equity valuation, which says that the stock is worth its book value plus the present value of its abnormal earnings (a shortcut is to use a multiple of earnings). Abnormal earnings are earnings in excess of the cost of capital employed by the firm. If a hockey team has a book value of $100m and its investors require an annual return of 10% (the “cost of equity"), but the team can earn 15% on its invested capital (ie it earns $15m on equity of $100m), then “abnormal earnings” are $100m X (15% - 10%) = $5m, and the team is worth more than its book value. If it can only earn 10%, then its abnormal earnings are zero ($100m x 10% - 10%) and it’s worth its book value. If it earns less than 10%, then it’s worth less than book. Two decent books that discuss the subject are Applied Equity Analysis by James English and The Quest For Value by G Bennett Stewart III.

Basically, if a chain of grocery stores earns $0.50/share but owns the land its stores sit on, and that land is worth $30 per share, then the company is obviously worth more than an advertising firm that also earns $0.50/share but doesn’t really own any assets other than goodwill. Likewise, a hockey team that owns its own arena, or the land it sits on, is worth more than one that doesn’t, even if they both generate the same earnings.

Also, I’m not sure if I agree with the sensitivity table on pg 31 of Mr Zimbalist’s document. Zimbalist bases his valuation on the present value of EBITDA. The proper approach is to calculate the present value of free cash flow—the money left over for the owners after everyone else has been paid and capital expenditures have been incurred to maintain the company’s assets. In Zimbalist’s table, the present value of EBITDA goes straight into the owner’s pockets, and that’s what the team is worth. Actually, from that EBITDA, the owner would still have to pay cash taxes, maintenance on the arena and other capital expenditures, interest on any debt, etc etc. Also, given today’s economic environment, the discount rate he’s using is much too low. A good part of the reason why the stock market crashed last fall is because the “risk premium” shot up as credit markets froze. In other words, future earnings stayed relatively the same, but investors required a lot more compensation to take on the added risk given the uncertainty at the time. The market has been rising recently in large part because that risk premium has been returning to normal. In any event, one could easily argue that Zimbalist’s discount rates are too low in the current environment, and thus his NPVs (which are already inflated by the use of EBITDA rather than free cash flow) are much too high.

Finally, the price to sales multiple is acceptable if all the companies in the industry have the same net margin (earnings divided by revenue). Sales are much smoother than earnings, and it could be argued that most teams will “revert to the mean” of industry net margins. In other words, a successful team that goes deep into the playoffs might earn a high net margin that year, but in the long run its net earnings will probably revert back to what a typical hockey team earns relative to its revenue.

Bottom line, I don’t think anyone is really right or wrong here—it’s very difficult to come up with an accurate valuation with the murky information available.

Tom—answer is sort of. Raiders move didn’t involve an ownership change, and the case didn’t get up to the high courts, so a different judge could probably overturn that precedent, depending on the case. The other factor here is the bankruptcy, which brings with it a whole different set of precedents.

And what was the deal with the California Golden Seals?  I see they wanted to move to Vancouver, but were blocked (and the courts agreed).

Dackle: I’m not sure I agree with you that you add the multiple of earnings to the book value.  Specifically, you say,

>“Basically, if a chain of grocery stores earns $0.50/share but owns the land its stores sit on, and that land is worth $30 per share, then the company is obviously worth more than an advertising firm that also earns $0.50/share but doesn’t really own any assets other than goodwill.”

I’m not sure that’s true.  I’d argue that you value whatever’s higher: the market value of the assets, or the present value of the free cash flow.

If I own a chain of grocery stores where the land is worth $30 per share, and it earns $3 a share, I would NOT value it much more highly than an advertising firm that earns $3 a share.  The only reason I *might* value it higher is that, if business goes bad, the advertising firm is out of business, but the grocery store can close the business and get back its $30 for the land. 

But $3 worth of earnings from a REIT isn’t worth more to me than $3 worth of earnings from Coca-Cola, in general.

I think I agree with the rest of what you said, but I think your “residual income” method is not the same as the quote I disagree with above.

Since Zimbalist is being paid by the new prospective owners, is there any reason to expect his work to be objective and independent (i.e., correct)?  Seriously.

Of course you don’t compute the value of a business which has a large portion of its costs fixed (salaries) by multiplying its revenue by some constant, and that’s it.  That is ridiculous.

But, Zimbalist goes with what Forbes does even though he knows that it is 100% wrong, because that is advantageous to his client, and he can always say, “Well if you can’t trust Forbes, who can you trust?”

Seriously, when a person is being hired to advocate for one side or another, their work is going to be COMPLETELY different than if they are writing independently and objectively, even if they are an expert.

If I am hired by a team or by a player’s agent in an arbitration hearing, do you think I am going to use the same stats and the same arguments regardless of which side I am representing?  No chance.  I will use whatever suits my position.  I might use UZR or I might use Dewan or I might use range factor or even fielding percentage, depending on which side I am representing and which stat supports that side. I might use wOBA or lwts or I might use BA, HR, and RBI, again, depending on what suits my client.

In short, NEVER trust the work of someone who is getting paid to represent one side or the other.

MGL, is there a general cultural set of expectations for this kind of expert witness?  What do lawyers expect and expert witnesses understand about what they’re expected to say?

And any academics reading ... what is the attitude in academia to respected experts whose work is not “objective and independent” in this context?  Is it tacitly accepted that it’s OK to provide a non-objective non-"correct" opinion for money, one that wouldn’t pass peer review in a more objective forum?

I am very curious about this.  To me, I think it’s OK (but not wonderful) to suggest arguments to the lawyers, in order that they make those arguments themselves.  However, I think it crosses the line to sign my name your name to them if they’re one-sided and you know that.

Phil, that is a good question.  I don’t know the answer.  It is a complex question and the waters are very muddy.  In the legal field, of course there are many ethical constraints on lawyers and prosecutors and they both often come as close as possible to those lines and sometimes cross them.  The idea is that if both sides are zealous advocates for their clients, with the understanding that they are certainly not going to be objective or even “fair”, that the truth will emerge, or at least will be discerned by an objective observer. I think that is a fair ideal.

In this case, if one were to read the testimony from the other side’s expert witnesses, one should, theoretically, be able to arrive at the truth or close to it.  For example, you provided a rebuttal to Zimbalist. If we give him an opportunity to reply to you and then you to him again, etc., we ought to be able to come up with something pretty close to the truth (which might be that we have to call it a draw, which sometimes happens).

But to read or use only one side’s arguments, when that side is admittedly biased, is a bad idea for arriving at the truth.  Which is why I can’t stand when someone forms an opinion on something, especially something complex, after only reading one side of the argument.  As I have said many times, it is EASY for an expert to make something look like X, when that something is complicated.  And people do that all the time.  For example, global warming is or is not a fiction because look at these arguments (one sided ones) from all of these experts.  They sure sound convincing, don’t they?  Sure they do. And so do the arguments from the other side!  Especially to a lay person, who is easily fooled by an expert, or a pseudo-expert opining about a complex issue.

I actually think there are two ways of arriving at the truth and one is better than the other. One is to have an unbiased and objective argument, from an expert of course.  The other is to have two, opposing biased and non-objective arguments.  The second is generally better. That is why we don’t have objective, independent arbiters in our legal system.  Some countries do.  Our system - the adversarial one - seems to work better.  As well it should.  And of course you sometimes have to put up with one side winning because they provided false or misleading arguments that either went unchallenged or were not properly rebutted by the other side.  That is especially true if the two sides are not balanced in power and resources.  That is one reason that we give great deference to the criminal defendant in our legal system - because the state generally has many more resources than the defendant, especially an indigent or even average (economically) one.

mgl/9,

I absolutely agree with you that a lawyer should be able to do anything and everything (except lying under oath) to further his client’s interests.  And I agree that the other side should do the same (except for criminal prosecutors), and you get to the truth by hashing out the arguments.

Because of this, I don’t believe anything at all that a lawyer says when he’s acting for his client.  Nothing.  If he says his client is male, I want to see a picture of the testicles.  If he says 1 plus 1 equals 2, I want to plug it in to my calculator to make sure.

I respect that lawyers are doing a necessary job, but it doesn’t mean I have to like what they do.

Anyway: my point was more about expert witnesses.  If someone offered to pay me (or not) to advance only one side of a case, I would decline.  If they wanted to pay me to tell them both sides so they could present their client’s side better ... well, in that case, I might go for it if it didn’t prevent me from speaking freely about it regardless.

To me, if a researcher, whose reputation depends (somewhat) on being unbiased, signs his name to a biased statement (even if it’s technically true) ... well, anything he says from then on is suspect to me.  I’m wondering if that’s generally true in academia, or if everyone tacitly understands that, yeah, that’s how you make an extra buck on the side, and everyone knows you stop advocating when you go back to your real research.

I just don’t like it much.  From a lawyer ... necessary evil.  From a non-lawyer ... just evil.

Alan Nathan has served some as an expert witness in cases where bat makers have been sued by people who have been injured by batted balls.  My impression from Alan was that his role in those cases was to present the facts without bias.  Obviously, the bat maker wouldn’t call him if they didn’t think the facts helped their case, though.

Right, that kind of thing is OK ... I’m thinking of cases more like, as mgl said, when a witness will say something “even though he knows that it is 100% wrong, because that is advantageous to his client.”

Does that happen often?  Is it considered ethical? 

I’m assuming that by “100% wrong,” mgl doesn’t mean the witness is lying about the facts, just that he’s presenting an argument he feels to be invalid.

Phil, think about my example of if I (or you) were representing a player or a team in an arbitration hearing.  Sometimes there are two ways to couch an argument and neither side is necessarily wrong.  Like the MVP debate that we have here occasionally.  I can easily represent one side or the other and not feel a twinge of guilt either way.  Now, if I were truly being unbiased and objective, I would give both sides and express my concerns, misgivings, and uncertainty, but if I were being paid to represent one side or the other…

Before I went to law school, Phil, I thought the same way as you. However, my perspective changed even though I never ended up practicing (so you cannot call me biased - I can give a rat’s ass about lawyers).  Because the system works best when each party un-objectively and zealously represents their party’s position and their interest, I would be doing society a disservice if I did not act in such a way.  That is true in a civil matter as well.

Even if I am pretty sure my client is guilty (he doesn’t tell me), I must and I should zealously represent him and try and get him off (I may suggest a plea bargain, if that is available of course, but that is up to the client).  I need to do that for several reasons.  One, to preserve the system in the future.  Two, because I might be wrong about my client and wouldn’t that be a hoot if he were innocent and I did not defend him with everything I have.  And three, if I don’t and my client is guilty, he might get his conviction reversed on appeal (for ineffective assistance of counsel).  Basically, no matter what, we want the jury to make the best decision possible and that involves me zealously defending my client and the state zealously prosecuting him.  And we are both bound by legal and ethical constraints, as we should be.

If Zimbalist wants the court go get at the truth, he must represent his client in a biased and non-objective fashion.  Otherwise the other side will likely win.  If he tells the “truth” (I am not suggesting that he lie of course) and hems and haws and is objective and unbiased, the other side will win almost 100% of the time, and that would not be fair to anyone involved.

Do you realize that if criminal lawyers did not use all the tricks in their bag that you might consider unethical, how many innocent people would go to jail?  Is that want you want?  Yeah, in a perfect world, the truth is all that is necessary.  But we don’t live in a perfect world.  Say, for example, my client is accused of beating up his wife and he is innocent.  And let’s say that he is a mean lout and everyone knows it.  And let’s say that I decide to put on the stand a couple of people that are willing to say how nice he is, for whatever reason they are willing to say that.  Am I being dishonest or unethical?  The prosecutor is free to put on witnesses who will testify what a lout my client is.  And if he does, and I don’t rebut that, and my client gets convicted because the jury is prejudiced by that testimony from the prosecution, that is a miscarriage of justice.  And let’s say that my client is guilty and they jury should know what a lout is he is in general.  Well, who am I to know or think that he is guilty and not put those friendly witnesses on the stand.  Lawyers, contrary to popular belief, do now generally know whether their clients are guilty or not.  They merely represent them to the best of their ability.

If no one were to represent OJ or even the guy that (presumably) rapes and kills a little girl and “tries to get them off” then when you are accused of a similar crime and you are innocent, who is going to be left to defend you?  No one!  Lawyers shouldn’t decide who they think is guilty or not and then use that to determine who they should defend vigorously or not (assuming, again, that the client does not want to plead guilty). How can they?  Is that what you want?  That is the only answer to the question of, “How can a lawyer defend someone who is obviously guilty and committed a heinous crime?” The only answer to that is for a lawyer to decide who is guilty and who is not and then tailor his defense to that opinion. That would be a horrible system and you would scream bloody murder if you were accused of a crime and everyone though that you did it.  That happens all the time of course, which is why we have hundreds of convicted murderers and rapists who have been exonerated with DNA evidence.

And when a lawyer “gets someone off” that they just know is guilty, people get really pissed off at the lawyer.  Of course, lawyers better get off lots of guilty people!  If they don’t, there are lots of innocent people who are going to go to prison.  Which is one reason why you zealously defend everyone whether you think they are guilty or not.  As the famous English jurist Blackstone said, “Better that ten guilty persons escape than that one innocent suffer.” That is the way our system is supposed to operate.  Some people disagree with that and some countries have systems where it is a lot easier to get a prosecution.  Presumably those countries believer that it is better to sacrifice a few innocent people in order to get as many guilty ones as possible behind bars.  I suppose that is a reasonable position as well. I don’t happen to ascribe to it.

"If Zimbalist wants the court go get at the truth, he must represent his client in a biased and non-objective fashion.”

This should be true for the team’s lawyer, but not Zimbalist.  Zimbalist should give his honest opinion.  He may marshall evidence in a way that makes the strongest possible case for that position (as an academic might also do), but he should only state fundamental conclusions that he really believes in.  The bias should be provided by the lawyer, who would decline to submit Zimbalist’s testimony if it doesn’t help his client. 

Now, all expert witnesses obviously don’t approach it that way.  Telling lawyers what they want to hear may get you hired as a witness more often.  But that’s not the only incentive a witness has:  if he wants to remain respected in his field—which also impacts his value as an expert witness—there are limits to how far from honest analysis they can stray.  And many experts probably value the respect of their professional peers, apart from any financial incentives.  My guess is that there are doctors, for example, who play it pretty straight as expert witnesses.  But also some who will say what they’re paid to say.

mgl/13:

Actually, I agree with you 100% in criminal cases ... my exception was for criminal *prosecutors*, not defense lawyers. 

I agree with you in civil cases too, that it’s the lawyer’s job to present the best case possible regardless. 

When you say that before you went to law school “you thought the same way as me” ... I think you think the same way as me NOW. 

Phil, sure, I should not have assumed you feel that way about lawyers who defend criminals, especially ones who are accused of heinous crimes and/or when we think we KNOW that the criminal is guilty.  Many people do feel that way, and, as is often the case, I think it is out of ignorance of the value of and necessity for the lawyer having to zealously defend ALL clients and for every accused to be able to have good representation.

It is true that prosecutors have a different set of responsibilities and that their job is not to “zealously convict a defendant.” However, probably the only reason that is the case is because it is assumed that the deck is stacked against the defendant in the first place (such as most juries go in automatically assuming that the defendant is guilty) and that the state has many, many more resources available to it than the average defendant.  If those things were not the case, then the system might be such that both sides were put on an equal footing in terms of their responsibilities.  Generally, the state is not supposed to have an “ax to grind” in a criminal prosecution, although in reality that is not true at all. The state is supposed to see to it that justice is served whereever that may take it. In most cases, for various reasons (job pressure, human nature, many prosecutors learn to hate criminal defendants and criminal lawyers, etc.) the state ends up vigorously and zealously prosecuting a defendant in the same way that a defense lawyer does, even though they are not supposed to.

Guy, yes, I agree that in a perfect world, the expert witness is supposed to tell the lawyer that hires him the truth and let the lawyer decide what to do with it. Of course in the real world that does not happen at all. What happens is that there are plenty of “guns for hire” that will do or say whatever the lawyer wants.  And a lawyer (both sides actually) gets to cherry pick their expert witnesses, so that if there is any controversy about the issue in question or there are several ways to couch the arguments (some in favor of one side and some in favor of the other side), then it is easy for the lawyer to choose the expert who happens to argue from the perspective that helps the lawyer’s case.  Even if there are 1000 experts who believe X and only 1 who believe Y (and those that believe Y are “whack jobs"), I can hire the one who believes Y if Y helps my case.

For example, if I want to argue that my client, Jeff Francouer, is worth 12 million dollars, I can hire JC Bradbury, even though 1000 other “experts” disagree with his argument.

Of course, if I am on the other side, and I have the resources (money), I will parade in 10 other experts who will refute Bradbury’s arguments and I will try and impugn or impeach Bradbury’s credentials or testimony as much as I can.  So, hopefully it all comes out in the wash in the final inning…

$3 worth of earnings from a REIT isn’t worth more to me than $3 worth of earnings from Coca-Cola, in general.

Phil, I agree with you in theory, but the market doesn’t seem to value companies that way. For example, here are nine S&P 500 companies with trailing 12-month free cash flow between $2.90 and $3.10 per share, along with their share prices. As you can see, the market values FCF of $3/share anywhere between $6.64 and $68.43 (I pulled these values this afternoon so prices are intraday):

Company            FCF/sh   Share price
Exxon Mobil         3.01       68.43
Waters              3.01       50.17
Brown-Forman        2.91       43.51
Hospira             3.02       39.09
BMC Software        2.99       35.27
Scripps             2.92       32.59
Pactiv              3.04       24.60
Leggett & Platt     3.04       18.29
Marshall & Ilsley   2.93        6.64

BTW, I didn’t cherry-pick those to prove a point The main conclusion for me is that the confidence interval is just too wide to pin down a really accurate share price based on free cash flow alone.

You made an interesting point about the relative flatness of market values relative to profits:

Why are the values so flat relative to profits, where a team that makes $1 million a year is worth almost half as much as a team that makes $40 million a year? It could be the Picasso effect. I ran a regression to predict market value based on earnings. The results, rounded:

What that tells us is that, regardless of earnings, there’s a value of $200 million dollars to owning a team, even if it only breaks even every year. That might be Picasso value. Or, it might partially reflect the value of the right to move the team if it starts losing money. It might reflect the fact that owners think that earnings will jump soon—maybe they think a new TV contract will someday be worth a present value of $30 million each, and that’s part of the $200 million. But I think it’s consumption value, Picasso value.

I’m not entirely convinced the $200m is Picasso value. For example, here is the same regression on the constituents of the S&P 500 (as of this afternoon). I used next year EPS consensus as “earnings” because it correlates more highly with share prices than current-year or trailing EPS. (EPS was converted to earnings before running the regression by multiplying by shares outstanding):

Market cap = (11.79 * earnings) + $2,251,860,354.  (r^2 = .929)

So, it looks like a Picasso effect for the S&P 500 as well, although I don’t think that’s the case. I think the intercept stands in for a variety of valuation methods besides earnings forecasts.

If you add in book value and force the intercept to zero, you tighten the valuation and avoid a generic intercept:

Market cap = (10.84 * earnings) + (.25 * book value) (r^2 = .935).

Using this method, the source of the divergence in share prices in the table above becomes clear. Exxon has a book value per share of $22.18 and next-year EPS of $5.89. Plugging those numbers into the equation, I get an estimated share price of $69.39. Leggett & Platt has a book value per share of $10.54 and next-year EPS of $0.94, resulting in a “fair value” of $12.82.

Dackle,

I agree with you that the confidence interval is too wide to pin down share price based on FCF alone.

The main reason is that share price is the present value of future cash flows (or dividends, if you prefer), and using one-year numbers isn’t enough.

Those regressions are interesting ... thanks for running them (in fact, your whole post was a lot of work!).  For book value, you seem to have a point.  But there are many reasons why this might hold.  One is that, perhaps, companies with lots of book value (land, factories, etc.) have been hit harder by the recession than others, and their earnings are expected to rise higher once the good times return. 

My response would be that in theory, long term, as an investor, I don’t really care how my company earns the dividends it uses to pay me—whether it’s factories (high book value) or a culture of innovation (low book value).  So long as the stream of dividends will be exactly the same, I shouldn’t care at all.

The difference might simply be that high book value means more growth in earnings/FCF/dividends from this year’s earnings, which is why you got a high positive coefficient.

I am more impressed by the apparent “picasso effect” term in the first regression.  I think that’s what you’d get if earnings were somewhat random.  If half the earnings were real and the rest were random (in the sense of variance), you’d get exactly that effect; the market would value the “real” earnings at a reasonable multiple (11.79), and value the other half at a fixed price based on the *expected value* of those earnings. 

That is likely also happening with the NHL teams, so I think you’re right that it’s not all Picasso effect.  However: I still think a lot of it IS Picasso effect, because I think earnings are less random in the NHL than they are in the S&P.  (For one thing, all NHL teams are in the same industry.  For another, none of them can really grow more than others by taking business from another team.)

But I should think about how to figure out how much of the $200MM is Picasso value and how much is earnings expectations.

I think I know a way: take consecutive years of NHL earnings and figure out the correlation from year to year.  Then, use the regression equation to figure out how to predict next year’s earnings.  That tells you how much to regress each season to the mean. 

Suppose you have to regress 30%.  That means that the future cash flow (assuming earnings = cash flow) equals

30% regressed + 30% regressed twice + 30% regressed three times ...

which gives you a substitute measure of expected earnings.  Then, you can rerun the regression and get a better estimate of Picasso value.

I’ll do that when I get a chance.

Hi Phil, yeah, I think the book value is a proxy for “normalized” earnings. Back in March the weight on book value was much higher—I was getting values for TSX-listed companies of around (.70 x book) + (6.0 x earnings). In a way this makes intuitive sense that book value would equal normalized earnings, since book value is a running total of a company’s “lifetime” earnings minus any dividends paid. This year’s book value is last year’s book value + net income - dividends. Book value is often dismissed because assets are based on historical values. I tend to look at it sometimes as assets and liabilities flowing from book value, not the other way around (as is commonly stated)—ie you have to earn the money before you can buy the assets.

I really like your idea of “real” earnings and “random” earnings. It seems like there should be a way of scaling the random (Picasso intercept) component to the size of the firm/team. In other words, the average team might have a Picasso value of $200m, but that might range from $100m for the smallest team to $300m for the largest. In terms of the S&P 500, I don’t know if I’d automatically hand out a value of $5/share to every company. You’d think the smaller ones would have a lower “base” value of maybe $0.50. But then you come back to the issue of what the scaling factor should be based on—arguably it should be book value (as sales are the only other obvious candidate and they’re less correlated with earnings and share prices).

I don’t really care how my company earns the dividends it uses to pay me—whether it’s factories (high book value) or a culture of innovation (low book value).  So long as the stream of dividends will be exactly the same, I shouldn’t care at all.

Agreed, although you likely won’t receive a dividend from the low-book firm, for example Jubilee Jim Balsillie’s Research In Motion—book value of around $12.50, no dividend, and trading at around C$81. On the other hand, if RIM earns $4/share this year, then book value will rise to $16.50. If the price/book ratio remains constant at about 6.5x, then the share price would rise to around $107. Whereas if they paid $3 in dividends, book value would rise to $13.50 and the share price to only $88. So, no dividend would result in cash flow to the investor via a capital gain of $107-$81 = $26. If RIM paid a $3 dividend, cash flow to the investor would be $3 plus a capital gain of $7 ($88-$81) = $10.

Anyway, has Forbes published those franchise values over a number of years? It might be possible to isolate the Picasso effect by regressing five years of earnings data onto market value. Also those teams might be valued by “normalized” playoff and regular-season success. I’d imagine that an average team that reaches the Stanley Cup finals generates substantial earnings but would still be valued as an average team that might only go one round in the future.

The other issue with those numbers is that there really isn’t enough information to do a proper valuation. When merger and acquisition transactions are announced, management of the acquiring firm will often quote the EV/TTM EBITDA multiple they paid. I saw EBITDA in Mr Zimbalist’s document, but you’d need to know debt, cash, and EBITDA over a number of years to really value the teams more accurately.

Hi, Dackle,

I tried a rudimentary attempt at getting rid of the randomness.  It doesn’t change much.

http://sabermetricresearch.blogspot.com/2009/09/re-estimating-nhl-teams-picasso-value.html

As for RIM ... yes, since they don’t pay dividends, the book value goes up.  But they reinvest that retained cash to increase earnings next year, so that should be reflected in the P/E without having to take book value into account.

Book value is only useful if it leads to higher earnings.  And it’s just another measure of the firms assets, which are put to use earning more profits, so it all comes out in the wash.

I’m not disagreeing with you that book value is important: just that, in theory, book value doesn’t matter if you know earnings.  In practice, it might have value in helping to predict earnings better, or putting a floor on them, or some such.