You're once again making some really dumb assumptions. The biggest of all is that there are 50 other games on which to bet.

Or, to bring it back to finance, there have to be 50 other investment worthy companies that fit within the subset of the types of companies that he has an edge in analyzing. I have to mention investment worthy because he has to choose companies to buy and hold whereas he may find 50 companies that fit into his analytical sweetspot, but his edge with those might be to know that those are ones in which he shouldn't invest.

Your example assumes every company is fair game and thus can all be treated as random variables. This makes you a loon.

i figured someone would come up with the "blah blah there's no reason to believe the distribution is normal" argument. but if you look at the distribution of returns on stocks, you'll see that the distribution is close enough to normal that sklansky's argument, "If someone can find 15 stocks a year that beats it by 20% it almost guarantees that he could find 50 that beat it by 10%.", is simply irrefutable.

if you don't have a basic understanding of statistics (crushinfelt), this really isn't an argument you should be involved in.

given the overall distribution of returns, suggesting that there are anywhere near as many stocks that are observably mispriced in such a way that they'll beat the market by 20% as there are such stocks that will beat the market by 10%, is just laughable.

You're like a parrot that repeats what it hears but doesn't actually know what it's saying. My argument had nothing to do with the distribution of returns of stocks and your lack of reading comprehension led you to make the same mistakes as the OP.

Buffet isn't looking at every stock on the exchange and then cherry picking the ones whose stock is going to appreciate by 20% more than a randomly chosen stock (nonetheless ones that will appreciate by more than the average stock would over the ensuing 3 year period).

He's looking at a very specific subset of companies whose business lines match what he's familiar with (that alone eliminates thousands of companies). From there he is able to pour through every detail of the companies' activities and then move on to how it fits into their competitive scene. At that point he is then able to make the decision to invest or not invest in a company. It's so unbelievably far from taking a baseball model and running the stats of more games through it and accepting a 10% edge rather than a 20%. It's laughably different. The OP is essential making the assumption that a winning baseball bettor's edges would automatically translate into having an edge in betting on hockey and basketball and curling and tennis and bowling and ping pong and whatever ****ed up foreign sports there are throughout the world.

"He's looking at a very specific subset of companies whose business lines match what he's familiar with (that alone eliminates thousands of companies). From there he is able to pour through every detail of the companies' activities and then move on to how it fits into their competitive scene. At that point he is then able to make the decision to invest or not invest in a company."

And when he chooses "not" some of the time he thinks its close. And for you to think that the ones he thinks are close will magically only meet the indexes is telling.

For this point to matter you must assume that they wouldn't which is also telling.

I find your insights laughable.

implicitly your argument has everything to do with the distribution of returns of stocks.

Buffett says the market is a voting machine in the short run which leads me to believe that he does not believe the market is as efficient pricing stocks as your assumptions require. However if the assumption is that the market is pretty damn efficient at pricing stocks then you are, of course, 100% correct.

I would point to the fact that a significantly large percentage of professional money managers significantly underperform the market. In a very efficient market it would not make sense that large numbers of pros would systematically underperform. It seems to me if a large percentage of professional managers are underperforming the individual investor should not feel safe that a randomly picked stock is fairly priced.

If Buffett were required to pick and hold 30 stocks out of a universe of 100 he might only find 3 he likes. The assumption is the remaining 97 stocks should perform as good as the market so his remaining 27 would at least equal the market on average and he would retain a small edge. However, if he took out the 3 good ones that outperform the market then the remaining 97 must perform at the market level MINUS the gain from the other 3, because the 100 stocks in total equal the market. If you have stocks that outperform you must have stocks that underperform.

The assumption is that Buffet should be able to tell from the stocks he choses not to invest in which will underperform. This may be true but if he could why not systematically short these stocks and hedge by buying market performers and earn better than the cash rates he is receiving on his mountains of uninvested cash? The reason is simple, he can't pick out the future underperforming stocks with any edge. His edge comes from finding a select group of companies that can generate above average returns it is not in being able to decisde better than anyone else the theoretical value of each and every stock.

The truth is that for my idea to work I don't need the EFM at all. I only need that the market goes up on average. That is all that is needed for a randomly picked stock to have plus EV. In other words it is not necessary for my one factor expert to "feel safe" that his stock is otherwise fairly priced. It is only necessary that it is on average fairly priced. Its like all of you are forgetting that sometimes the one factor expert will make more than he expected because the stuff he didn't know about will ADD to his profit. The unknowns work both ways.

You earlier tried to make your case by postulating a guy who thought Pet.com would go up solely because of its new advertising campaign. Who then got slaughtered. But that example both cherrypicked and also implied he might not be a real expert. But if you make the example more reasonable it illustrates my point well. A company unveils a major new eagerly awaited advertising campaign. Perhaps the stock is up in anticipation. You are a star at a different advertising agency and tell me the ads suck. That is not likely to be a better reason to sell short than what some MBA tells me? If you agree, are you saying that my uncertainty about whether the stock was fairly priced should dissuade me (unless I am worried that other advertising experts got there before me.)

I'm just thinking that the value of that information should be greater then the amount by which that stock COULD be over valued. If all stocks on average are fairly valued this should not be a problem. The greater dispersion of pricing around fair value in the stock market could create higher risks though. Also if an investor is biased toward seeing overvalued stocks as undervalued you might have a problem.

It is impossible to know how the market is pricing your one variable. You may think the stock is over pricing an ad campaign but in reality it may actually think the ad campaign is as bad as you do but it is pricing in a potential take over that you are unaware of. This actually superior information may not be yet fully priced in to the stock but priced in enough to make you belive the stock market is over valuing the ad campaign.

I'm not disagreeing, just playing devils advocate. I read Fabricand's book with great interest. It seems to me the horse race comparison in the stock market is with a group of stocks in the same business with similar track records with one trading a little richer than the others. One would not expect this stock to be more highly valued and perhaps it is experts with material information beginning to price in a variable unknown to the market. Most outside experts might consider the stock over valued based on market benchmarks of profitability, sales and / or assets because they are unaware of this superior information.

"It is impossible to know how the market is pricing your one variable. You may think the stock is over pricing an ad campaign but in reality it may actually think the ad campaign is as bad as you do but it is pricing in a potential take over that you are unaware of."

But you could use common sense to decide its a remote possibility. In this example it might be because you have good reason to believe no other experts saw the ad. Or better yet you don't care because you did your own private study that shows that people get turned off by ads that use aardvarks but you have no reason to believe that anyone else knows it.

I agree that acting on information an investor believes to be superior in a fairly effecient market can't hurt. However, I feel the odds of any one person possessing MATERIAL information that is not already discounted to be quite small.

Theoretically a stock is a machine that discounts future cash flows. When the Shuttle exploded the market had correctly identified within 15 minutes the company responsible and discounted the stock nearly precisely the amount of damage to the company. This happened although no one trader knew what caused the crash and it in fact took a 6 month investigation to determine the cause.

The problem with inside information is if it really is material and non public you are breaking the law by acting on it and if it isn't material and non public its worthless.

This sounds a little urban legendish. Can you cite the source? How do they know that the market just realized Shuttle launches were going to stop for a while, and discounted the company's stock for the value of losing that business?

Also, I think I have a counter example to this. Assume you are an analyst researching a retailer. You develop a list of stores that you think offer a statistically valid cross sample of a retailer's total sales, and cultivate relationships with the managers at each of those stores. One week you call on all of the managers and find every one is negative about the retailer's future, due to sales trends and changes in how the stores are being run. They don't give you specific sales numbers for the quarter or anything remotely illegal.

You know have non public information. It's legal. It's material. And it's likely very valuable.

I think it was Gordon Gecko if my memory serves me correct

It was an example used in the The Wisdom of Crowds.

It is just hard for me to believe that this type of reversal of fortune has not already worked its way in to the price of the stock. Even if it hasn't, the odds of an individual investor ever having this type of information on a particular company is infinitesimal.

How does this information work it's way into the price of the stock if market researchers aren't trading on this information? I.e. it works it's way into the price because the proactive investors who collect this information use it profitably.

And we are talking about Warren Buffett in this thread, you could call him an individual investor but he certainly has resources to do research like this. Edit: And he does, but it doesn't appear in any organised manner, I think he just uses Phil Fisher's scuttlebutt approach. Carl Icahn has enormous resources to do this stuff as well.

Watch the first season of Wall Street Warriors, the hedge fund guy does this kind of research on a fad jeans manufacturer, and his shop is almost a one man effort (he has a trader, a CFO to manage finances, but it seems he makes all the investing decisions).

And the interesting thing about it is that there are cases where investors can have more information about a companies prospects than top management at the company does. Sometimes the grumbling/turnover at the bottom hasn't reached the top, or the management group is so psychologically committed to a course of action they dismiss/ignore bad news, or sometime punish bearers of bad news so that underlings tend to hoard the information hoping it will go away.

I used to run a decent sized product development organization at a software company and was always amazed at the crap that was going on at the bottom of our organization that me, the CEO and CFO found out about last.

I'll get the book, sounds interesting. But this example doesn't sound convincing.

The stock price of a company close to bankruptcy that is at risk to have a significant part of it's business delayed or canceled due to the Shuttle explosion should react more than financially stronger suppliers to the program. Also if Morton Thiokol derives a bigger percentage of it's business from the shuttle program than other suppliers.

When you go to a grocery store do you calculate the amount of goods each person has in front of you, the speed of the cashier and the likeliest shortest wait line? I don't and I don't think it makes much of a difference because the market has done the calculation for me. All lines will ususally take aproximately the same amount of time.

I seriously question the value of research. It seems to me psychology matters more in investment decisions, the ability to buy quality when others are fearful. It seems to me stocks swing around theoretical value depending on sentiment and the gain from buying when sentiment is pessimistic seems to be more than can be gleaned from out researching people on the company's prospects. The real value of research is to identify company's, as Buffett says, that have a durable competitive advantage and can compound earnings without fear of much competition.

Not sure of the relevance, but no, all lines don't take the same amount of time. Some cashiers are faster, some customers have more goods, etc, and many people don't pay attention.

Research matters. If you were shorting a fad retailer, the only way to know whether the fad is fading before the market reflects it is to do proprietary research. The only way to develop lists of companies that have durable competitive advantages is to do proprietary research. The only way to figure out before others that those advantages are fading is to do...wait for it... proprietary research. An example would be KO, whose barriers to competition are fading bit by bit.

If the market is so efficient you don't have to do research, who's doing the research to make the market so efficient?

The fact that so much research is being done means the individual investor has about zero chance of doing any meaningful research. Check those lines out in the stores some time and you will find out that over time just by randomness the line you pick will have the same mean wait time as that of all the other lines not chosen and there will be little variance in wait times.

Russia attempted to set prices through market research with disastrous results. The free market mysteriously prices products in efficient market clearing manner with no research. The same principal applies to stock pricing.

Durable competitive advantage can be gleaned from seeing a long record of profitability coupled with factors that protect a company from competition. These company's are best bought when investors are fearful. The same strategy can be used in income real estate or buying private companies.

These are my personal opinions and I do not hold them up as fact.

I am not sure that information qualifies as material. If it does it is borderline.

Material: having real importance or great consequences.

Knowledge of a sudden downturn in the fortunes of a public retailer == extremely material.