2. Analytical edge: the same data, a better reading
Level 7: The four possible edges
If nobody can know before you, and you cannot know before anybody, one possibility remains: that we all read the same document and you understand it better. That is the analytical edge. It is the most respectable of the four, the one that sustains the active management industry, and also the easiest to believe you have without having it.
The price does not contain the fact, it contains the expectation
The beginner's error is to think that analysis means forecasting a company's future. It does not. Today's price already embeds a collective forecast of that future. Your forecast is only worth money if it differs from the one built into the price, and differs in the right direction.
Michael Mauboussin has spent decades insisting on turning the exercise around: instead of estimating growth and comparing it with the price, start from the price and work out what growth, what margin and what duration of competitive advantage would be needed to justify it. The question stops being 'how much will it grow?' and becomes 'is what you have to believe in order to pay this reasonable?'. That is far more answerable, and it exposes immediately whether you have something to say or merely an opinion that is already in the price.
Watch out for a nuance that catches a lot of people: the published analyst consensus is not the market's expectation, it is a survey. What has to be beaten is what the money is discounting, and the two often do not match.
The three honest ways of reading better
When someone has a genuine analytical edge, it almost always comes from one of these three things:
- Understanding the business better than the consensus. Really knowing what makes a customer stay or leave, how costs behave when volume doubles, who holds pricing power in that chain and what happens if a competitor enters. That is sector knowledge, not financial knowledge.
- Valuing the intangible properly. Accounting treats money spent on research, on brand or on acquiring customers as an expense of the year, when in many businesses that is investment building a durable asset. Reported profit ends up depressed and book value captures none of it. In a market where intangible investment has come to weigh more than tangible investment, reading that distortion well is one of the few legitimate sources of disagreement left.
- Thinking over a horizon the average analyst does not cover. The bulk of professional work is concentrated on the current quarter and the following year. Years four to ten are barely studied seriously by anyone, and that is where most of a company's value sits.
The paradox of skill
Now the uncomfortable part. Charles Ellis described the modern market as a game where professionals no longer compete against amateurs but against each other: trading in the hands of private investors went from being a meaningful share of the market in the middle of the last century to a small fraction of the total.
Out of that comes what Mauboussin calls the paradox of skill: when everyone's absolute skill rises and also converges, luck weighs more in the relative outcome, not less. The dispersion of results between managers has narrowed decade by decade. It is not that they have got worse: it is that there are hardly any weak rivals left to beat.
The periodic comparisons between actively managed funds and indices point the same way with boring consistency: over fifteen or twenty years, around nine out of ten US large-cap funds fall behind their index once costs are taken out. That does not prove the analytical edge does not exist. It proves it is rare, and that its market price is high.
How to tell whether you have it: the arithmetic nobody wants to do
Here is the hard test in this module, and it is purely logical.
Suppose your real edge is 2 percentage points a year over the index and that your result deviates from the index by about 15 points a year, up or down, which is normal in a concentrated portfolio. The signal is 2 and the noise is 15. Since noise only dilutes with the square root of the number of years, for your track record to separate that edge from chance with reasonable confidence you would need something on the order of two centuries of data. With 3 points of edge and 10 of deviation, the sum falls to about 45 years: still more than an entire investing lifetime.
The conclusion is not that having an analytical edge is impossible: it is that you are not going to be able to prove it to yourself from results. Five good years prove nothing. That is why the only control available is to assess the process: write down what you expected, over what horizon and why the market was wrong, and check afterwards whether you were right for the reason you gave or for another one. Being right for the wrong reason is just as dangerous as being wrong.
Real case: twenty years of almost no accounting profit
A large e-commerce company reported virtually non-existent accounting profit for almost two decades, with losses in some years. In 2014, for example, it reported a net loss of around 240 million dollars on sales of close to 89 billion, while cash flow from its operations was comfortably positive and growing year after year. Anyone looking only at profit saw a company that did not make money. Anyone who understood that a large part of that spending was building logistics capacity, technology and a recurring customer base saw something very different.
That was a genuine analytical edge, and it was available in public information. But beware the easy moral: over that same period hundreds of companies also spent their way into losses and built nothing. The edge was not in applauding losses. It was in telling which ones were investment and which were money being destroyed. That is what almost nobody knows how to do, and that is why it pays.