DEADLY TEMPTATION #1

Beat the Market

We told you that you must tune out siren calls to “beat the market.” Now we’ll tell you why. The basic reason is this: You are extremely unlikely to find a manager, an advisor, or a strategy that will enable you to predictably beat the market. It will be even more difficult to know whether you have found one. It will certainly not be worth the cost to try.

THE ODD BEHAVIOR OF STOCK PRICES

In their textbook Principles of Corporate Finance, authors Richard A. Brealey and Stewart C. Myers report the following event in stock market history:

In 1953 Maurice Kendall, a British statistician, presented a controversial paper to the Royal Statistical Society on the behavior of stock and commodity prices. Kendall had expected to find regular price cycles, but to his surprise they did not seem to exist. Each series appeared to be “a ‘wandering’ one, almost as if once a week the Demon of Chance drew a random number … and added it to the current price to determine the next week’s price.”¹

In other words, each change in price seemed to be drawn at random from a bowl full of pieces of paper with percent price changes written on them—“up 1%,” “down 0.5%,” “up 2/3%,” and so on. If that was how price changes were determined, what would be the use of trying to predict securities prices—or the stock market as a whole? Whatever you predicted, something random would happen that might agree with your prediction or might not, but it would be no better than any other prediction.

This phenomenon of random securities price movements was not fully explained for more than 10 years after that. In 1965, the economist Paul Samuelson wrote a paper titled “Proof That Properly Anticipated Prices Fluctuate Randomly.” The proof is really pretty simple. Think about it this way: Everything about a public company (a company with stock listed on a stock exchange) that can be made public is required by law to be made public immediately. Then, thousands of analysts leap on the information and recommend buying or selling the stock based on that information. The stock reaches a consensus price almost instantly. This is what people mean when they say that markets are “efficient”—there’s hardly any time at all between the release of a company’s public information and setting that company’s stock price.

So what will be its price tomorrow? Well, that depends on the information that’s publicized between now and tomorrow. It could be anything—anything, except for one fact we do know about it: it’s information we know nothing about now. So it is unpredictable information.

Unpredictable information is random information. It is as likely to drive the price of the stock up as to drive it down (well, it’s always very slightly more likely to drive it up because companies’ earnings do grow over time, on average). You can’t tell which.

And neither can professional investment managers—no matter how smart they are, no matter how sophisticated, no matter how good their advertising is at selling their management skills to you, and no matter how well their investments performed in the past.

This is just a fact. It’s provable, using real data—as much real data as you’d like. It’s been proven again and again and again, every decade at least and more, going all the way back to the 1930s. You cannot predict which stock, or which mutual fund, or which investment management firm, will do better in the future by looking at its past performance. Many learned books about investment underscore this fact, beginning with Burton Malkiel’s A Random Walk Down Wall Street as well as Michael’s previous book The Big Investment Lie.

THE HISTORY OF ACADEMIC STUDIES

The first widely known academic study of the investment performance of professional investors was reported in the July 1933 issue of the journal Econometrica.² The study is known as the Cowles Report, because it was performed by the Cowles Commission, endowed by investor Alfred Cowles. As reported in The Big Investment Lie:³

One of the first research projects of the Cowles Commission was to study whether it was possible to predict the movement of stocks and stock markets. The preliminary version of the 1933 report, titled “Can Stock Market Forecasters Forecast?” was summarized in a famous three-word abstract: “It is doubtful.”

The Cowles report reviewed the stock recommendations and market forecasts of sixteen financial services, twenty fire insurance companies, twenty-four financial publications, and those of William Peter Hamilton, the editor of the Wall Street Journal. In each case, the recommendations performed worse than the average common stock, by an amount ranging from 1.2 to 4 percent per annum. For example, the report’s summary states:

“Sixteen financial services, in making some 7500 recommendations of individual common stocks for investment during the period from January 1,1928, to July 1,1932, compiled an average record that was worse than that of the average common stock by 1.43 per cent annually. Statistical tests of the best individual records failed to demonstrate that they exhibited skill, and indicated that they more probably were results of chance.”

In other words, the stock forecasters not only performed randomly, but their performance was actually even a little worse than random.

In the 1964 book The Random Character of Stock Market Prices,⁴ edited by MIT professor Paul Cootner, two papers by Holbrook Working and M. F. M. Osborne reported observing the same randomness in securities prices that Kendall had described to the Royal Statistical Society in 1953. Then in 1966, William F. Sharpe, who would later win the Nobel Prize in economics, performed a study of 34 mutual funds over two 10-year periods. He concluded, “Results actually obtained by the holder of mutual fund shares (after the costs associated with the operation of the fund have been deducted) fall somewhat short of those from the Dow-Jones portfolio. This is consistent with our previous conclusion that, all other things being equal, the smaller a fund’s expense ratio, the better the results obtained by its stockholders.”⁵

A widely cited 1968 article by Michael C. Jensen, then a professor at the University of Rochester College of Business, concluded from a study of 115 mutual funds over the period 1945-1964 “not only that [these] mutual funds were on average not able to predict security prices well enough to outperform a buy-the-market-and-hold policy, but also that there is very little evidence that any individual fund was able to do significantly better than that which we expected from mere random chance.”⁶

All of this evidence caused economics Nobel laureate Mer-ton Miller to declare, “If there’s 10,000 people looking at the stocks and trying to pick winners, one in 10,000 is going to score, by chance alone, a great coup, and that’s all that’s going on. It’s a game, it’s a chance operation, and people think they are doing something purposeful …but they’re really not.”⁷

Nevertheless, occasionally a research study produced some pattern in past data that appeared to be predictable. Because by that time it had become widely accepted in the academic community that the market was “efficient” and unpredictable, when professor G. William Schwert, also of the University of Rochester, studied these patterns in 2003, they were termed “anomalies.” After studying several claimed anomalies that had been observed in historical data, Schwert concluded with what we’ll call the “Schwert rule”: “After they are documented and analyzed in the academic literature, anomalies often seem to disappear, reverse, or attenuate.”⁸ Schwert then continued: “This raises the question of whether profit opportunities existed in the past, but have since been arbitraged away, or whether the anomalies were simply statistical aberrations that attracted the attention of academics and practitioners.”

In other words, Schwert raises two possibilities. One is that the anomaly was a “real” phenomenon—an underpriced stock, for example—that if someone knew about could have been taken advantage of; but as soon as it was studied and publicized, the opportunity disappeared. The second possibility is that the investigator who found the anomaly had really only detected a random pattern in the data, the result of too much sifting and processing of the data—a practice called “overfitting.”

Studies on stock performance have continued to be performed on a regular basis. For example, in 1997, Mark M. Carhart, then at the School of Business Administration at the University of Southern California, studied a sample of 1,892 mutual funds over the years 1962-1993 and concluded, “The results do not support the existence of skilled or informed mutual fund portfolio managers.”⁹ And in a 2010 study of 2,076 mutual funds from 1975 to 2006 by Laurent Barras, Olivier Scaillet, and Russ Wermers,¹⁰ reported in the New York Times,¹¹ the authors, using a method called the False Discovery Rate test, concluded that 99.4% of the managers had no stock-picking skill, while the 0.6% that appeared to have some skill was “statistically indistinguishable from zero.” In other words, it could easily have been a statistical fluke.

The strong conclusion, as one commentator stated it, is that “when you pay extra to ‘beat the market,’ you end up, in the long run, paradoxically underperforming the market by, at least, the amount of your fees.”¹²

THE ANECDOTAL HISTORY OF STAR PERFORMERS

Nevertheless, when people hear the results of these statistical studies, they still don’t believe them. It’s like a smoker who hears the results of the statistical studies that show that smoking causes cancer but doesn’t believe it will happen to him. He’ll say something like, “Well, Homer down the road is 94, and he’s been smoking for 80 years and is still fine.” Homer’s case, however, doesn’t disprove the statistical result. Homer was just lucky. The problem is that people hear all the time about star investors who have spectacularly outperformed the market, and they ask, “Well, what about him or her?”

There are three possible answers. One is that, like Homer, that star investor was just lucky. Another is that, possibly like Homer, he did something that we don’t understand that somehow staved off cancer—or beat the market. But we’ll really never know which. And the third answer is …just wait. Because far more often than not, those star performers subsequently wind up plummeting to the ground. Let’s take the history of these spectacular rises—and falls.

In fact, to start with a well-known example, one of us, Michael, was once exchanging emails with a very prominent financial journalist on the topic of the possibility of beating the market. The journalist was knowledgeable about investing and inclined to agree, but he said, “Yes, but what about Bill Miller?” Unfortunately, Michael’s journalist friend wasn’t quite up on the story of Bill Miller—he had old information.

The Wikipedia entry on Bill Miller¹³ provides a very interesting history and analysis of Miller’s investment performance from 1991 to 2005, when he beat the stock market (the Standard & Poor’s [S&P] 500 index) 15 years in a row, becoming a legend. One analyst calculated that the probability of this happening at random was only 1 in 2.3 million—making it all but impossible that Miller’s performance was only luck (but the Wikipedia site reports that another analyst disagreed with this probability calculation).

In any case, it didn’t last. The Schwert Rule kicked in, as it almost always seems to. According to a May 2012 Forbes article, Miller’s subsequent performance over the next five years was an appallingly bad return of -30.05%, compared to the market’s (S&P 500) return of +10.48%.¹⁴ This outcome “left many latecomer investors far worse off than if they had never followed the ‘hot money’ into Miller’s fund in the first place…. To add salt on the wound for recent ‘hot money investors’ Miller’s fund charged a sky high expense ratio of 1.77%.”

This is not an atypical result for investors who think that when they see that an investment manager has done well, then they should invest with that investment manager. And we won’t even mention Bernie Madoff again.

Two investment managers whose names are still legendary are Julian Robertson and George Soros. But few seem to know that while they both experienced long hot streaks, they both performed extremely poorly at the end of the 1990s. Robertson’s funds, for example, underperformed the S&P 500 by 40% in 1999.¹⁵ Since this decline occurred right after his funds had the most money invested in them—$23 billion—this means it is possible that more investors may have lost money investing with Robertson than made money. Yet his reputation as an investing guru lives on.

John Bogle, the founder of the Vanguard Group of mutual funds and a much-admired pioneer of index investing, reports in his book Don’t Count On It¹⁶ the sad story of Gerald Tsai, a star investor of the 1960s. “Tsai was the inscrutable manager who had turned in a remarkable record in running Fidelity Capital Fund—+296 percent in 1958-1965 compared to a gain of 166 percent for the average conservative equity fund.” Nevertheless, he subsequently racked up by the end of 1974 the worst eight-year performance record in the entire mutual fund industry, according to Bogle—a cumulative loss of 70%.

Here’s another example of reversal of investment performance from The Big Investment Lie:¹⁷

A striking example of the point that past performance does not predict future performance is a story in the July 11,2003, issue of The Economist. The Economist’s story relates the astonishing fact that the top ten mutual funds for the three years from the end of 1996 to the end of 1999 were all among the worst-performing 7 percent of mutual funds in the next three years.

If you had invested at the beginning of 2000 in the top ten performers for the previous three years, you would have experienced a brutal come-uppance. In the next three years, you would have lost 70 percent of your investment. Many investors actually did this—a large proportion of them on the advice of professional investment counselors.

At any time there will always be some investment professional who has had an outstanding record of performance—in fact, this has to be true, even if performance really is completely random. If you toss a thousand pennies 10 times, one of them will randomly come up heads 10 times in a row, even though the chances of that happening are only one in a thousand.¹⁸ And there are many thousands of investors.

But if you see performance like that and assume it will continue, take John Bogle’s title as your guide: don’t count on it.

THE TREND IS YOUR FRIEND

People seem to use the little motto “The trend is your friend” to mean that whatever investment has performed well recently will keep on performing well. But that is a mistake.

For example, many people invest in mutual funds. They have a choice of mutual funds to invest in. So they have to choose. These mutual funds always provide information about their historical returns—typically one-year, three-year, and five-year investment returns. Almost invariably, people choose to invest in the fund with the best one-, three-, and five-year returns. Many voices in the industry—financial advisors, representatives of the mutual funds, industry experts, and even impartial mutual fund screeners—suggest this is logical and the best practice. It even feels good to do research and hunt down the best products.

But it makes no sense.

“Now, come on,” you say, “how can that make no sense? If you want to choose a lawyer, you’ll want the one that has the best record of cases won to cases lost. If you want to choose a surgeon, you’ll want to choose the one with the best record of successful operations as compared with unsuccessful ones. If you want to choose a baseball player, you’ll choose the one with the best record of hits, home runs, and runs batted in. How can you not choose the mutual fund with the best investment record?”

The reason is simple: because that record bears no—repeat no—relation to how well that fund will do in the future.

“Huh? You mean that if some star investment manager has done well in the past and has become a celebrity, that doesn’t mean he’ll do well in the future?”

That’s right—no, it doesn’t. Just the information that his historical record is good tells you nothing, absolutely nothing at all, about how well he’ll perform in the future. That’s why the U.S. Securities and Exchange Commission (SEC), the regulatory agency that governs much of the investment business, requires mutual funds to put words on its advertising that read something like “Past performance is not necessarily indicative of future results” or “Past performance is not necessarily predictive of future results.” The trouble is, these words aren’t strong enough. The truth is that past investment performance has no ability at all to predict future results. It’s not just not necessarily predictive; it’s not predictive at all.

What does that mean, exactly? Well, suppose you took 20 funds that performed below average in the past five years and 20 that performed above average. Which group do you think will have more funds performing above average in the next five years?

You would assume more of the historical better-performers will perform well in the next five years than the worse-performers. But you’d be wrong. On average, it’s just the same. A fund that performed well in the past will have no greater likelihood of doing well in the future than one that performed badly.

Twice during his career, as part of his job, one of us (Michael) has worked with very large, private databases of the performance of professionally managed investment funds. The first time, he had access to the largest database of professionally managed institutional investment funds in the world (pension funds, endowment funds, foundation funds, union funds, profit-sharing funds, and others), which was privately owned by the brokerage firm A. G. Becker & Co. The second time he had access to the database of investments managed or advised by one of the largest brokerage and investment consulting firms in the United States.

In both cases, he performed studies exactly like the one we just mentioned. He looked to see how professionally managed investment accounts ranked in one five-year time period and divided them into the best and worst performers, then looked to see how they ranked in the next five-year time period. There was absolutely no consistency from one time period to the next. There was no way to tell how a fund would rank in the future by looking at how well it had done in the past.

This phenomenon has been studied by a long series of academic investigators over several decades, using a variety of methods.¹⁹ The results are always the same. Past performance is not at all predictive of future performance. In fact, the only thing that is predictive of after-fee investment performance is the level of fees itself: the higher the fees, the worse the fund will perform.

Unfortunately, we’ve seen people get tripped up by their language again and again. After Michael’s book The Big Investment Lie was published at the beginning of 2007, he did a number of speaking engagements. One time, he shared the platform with another man, who had also written a book. What was the nature of the man’s message? He said, “If your investments are not performing well for you, you should change them.”

What can be wrong with that? So Michael asked him, “Do you agree that you can’t predict the performance of an investment?” Well, he did—Michael knew he did, because he had said so in his talk. “So do you mean that if your investments have not performed well for you, then you should get rid of them because they won’t perform well in the future?”

No, he said, I mean if they’re not performing well for you, why would you keep them?

Michael was exasperated—he couldn’t seem to get clear that “not performing well” meant the same as “haven’t performed well”; and “haven’t performed well” meant it was history, and it wouldn’t help you predict how investments would perform in the future.

This is the tricky problem with trying to get people to understand that you just can’t use past performance to predict future performance. The assumption that you can is embedded deep in our consciousness and even in our language.

CHARTING: MUCH ADO ABOUT NOTHING

Nevertheless, lots of people use charts and statistics about past performance to try to predict future performance. This practice is called “charting,” and here’s everything you need to know about it: it doesn’t work. Or maybe the more accurate thing to say is it doesn’t work statistically.

That is, it doesn’t work just because it works sometimes or for a while. You might play the slot machines in Vegas, and if you say to yourself every time you pull the lever on a slot machine (or whatever you do these days), “Mama needs a new pair of shoes” and you go on a winning streak, it doesn’t mean that in the future you’ll continue to win as long as you keep saying, “Mama needs a new pair of shoes.” It doesn’t work in principle, and it doesn’t work empirically. That is, if someone studied a group of people who said to themselves, “Mama needs a new pair of shoes” every time they pulled the lever, and another group that didn’t say that to themselves, one group wouldn’t do any better than the other. That’s true even if you happened to go on a winning streak once when you said to yourself, “Mama needs a new pair of shoes.”

The same thing is true of mutual funds, and every other kind of investment. They’ll win sometimes, and they’ll lose sometimes. Sometimes they’ll go on a winning streak. But it doesn’t prove that they can keep it up. Whatever they’re doing is the equivalent of saying to themselves, “Mama needs a new pair of shoes.”

OK, we understand that this is hard to believe. Investing isn’t like playing the slot machines. In investing, there are things to know, real things. If you know them, you can predict.

What things? What makes you think there aren’t also real things to know when you play the slot machines? Slot machines are machines, so they have gears, connecting chains, or whatever; actually, these days they probably just have electronics. But still, they obey the laws of physics, like any physical device. How can that not be predictable, if you knew enough about how it works?

For some reason, we have no problem understanding that it doesn’t matter how much you know about how a slot machine works—you still can’t predict its result. It’s designed so you don’t know, and it’s designed well.

The same thing is true of a coin toss. Everybody assumes it’s random. You wouldn’t expect a theoretical physicist with a Ph.D. to predict the result any better than anyone else. So why do people assume that if you’re smart enough or know enough or do enough analysis, you’ll be able to predict the change in price of a stock?

Well, all right. We’ll have to admit it. There is a difference. It’s true that if you knew something about a company before anyone else knew it, you might be able to predict ahead of time how the company’s stock will move. If you were the only one who knew that Apple was about to announce a new, paradigm-busting invention tomorrow—or that it was about to announce that iPhone sales were not going very well—you’d have a pretty good idea what would happen to the stock price tomorrow, and you’d buy or sell the stock accordingly.

The trouble is, you don’t know things like that before anyone else knows them. And if you ever do know them, you’re probably a corporate insider, and it’s illegal for you to trade on that information.

The point is that there are many people who work full-time—and more than full-time, in most cases—in the financial industry who collect huge salaries and bonuses to try to get information like that about companies. If someone finds something that others don’t know, it won’t stay private for seconds. It will change the price of the stock so that it won’t be a good deal to buy the stock anymore (or sell it, if the news is bad), because the price will already have changed to reflect the information. It happens too quickly, just because there are so many people who get paid so well to watch out for it.

True, if people who do this didn’t get paid so well just to watch, there would be fewer of them doing it. Prices might not respond so fast to new information. But that’s not the situation now; it’s not even remotely close to the situation now. There are far too many people in the financial industry who get paid to watch out for this sort of thing. Why? Because so many people invest in the mutual fund that did the best over the past one, three, or five years. They’re paying the stock analysts’ salaries and bonuses by paying big fees to a mutual fund that might accidentally have done well in the past but won’t in the future.

Now, suppose you find out something about Apple that you think other people don’t know. How do you know they don’t know it? How do you know that Apple stock hasn’t already been priced to take account of that information? Could you tell whether it has or hasn’t by looking at the price of the stock? The answer is no. So not only are you unlikely to know important information about a company before anyone else knows it, you’re even more unlikely to know it and to know that no one else knows it. And if you don’t know both of those things, you won’t have any advantage in trading the stock.

Forget the whole idea of predicting the price of a stock, not to mention a whole portfolio of stocks like a mutual fund. It’s almost as unlikely that you can do it as that you’ll see a unicorn. It’s even less likely than winning the lottery—but the payoff isn’t as big.

And there’s a cost to just trying to predict the price of a stock or the returns on a mutual fund. If the cost is very low, it doesn’t matter much—you might as well try, if you want. But usually the price isn’t low—just like if you play the lottery a lot, you’ll pay and pay and pay and yes, there’s a chance you’ll win; but the chances are much greater that you’ll just pay, and lose.

Many people seem to believe that if you just trade stocks a lot—or currencies, or options, or some other thing—you can make a lot of money. This is the urban legend of day trading. We can’t tell you how many times we’ve heard this kind of story. Someone we meet or know will mention offhandedly that some friend of theirs, or relative, or someone they know, quit his job and just trades stocks, or options, or currencies, at home. Sometimes they’ll add that this person does very well.

Sorry, it just isn’t true. It’s so far from the truth that it’s a genuine, 100% colossal whopper. People don’t tell the truth about these things. It’s funny how often you discover that a few months later, people who quit their jobs to make money trading securities at home are back at their jobs. If it was working so well, why didn’t they keep on doing it? People who do this kind of trading tell you about their winners but they don’t tell you about their losers. And they don’t tell you what their net gains or losses are; often they don’t even know.

Many day traders and at-home traders who believe they can make money this way are prey for hawkers who sell “systems” or training courses at tuition fees of thousands of dollars to trade stocks, or options, or currencies, or commodities. The systems they sell can’t be verified to work, using any scientific procedure for verification; they just look complicated and technical enough to work.

Many people who trade for themselves at home pay for systems like these and pay a lot of brokerage fees. But they almost never make much, if any, money themselves. Of course, they often lose. What they’re doing is very much like gambling at Las Vegas—their results are random, and the house takes a substantial share of what they bet. Of course, like people who gamble at Las Vegas, sometimes they make a killing—but usually not for long. You won’t find anybody who started with $10,000 and spent four years trading at home who emerged with a million dollars.

UPS, DOWNS, AND DITCHING CONVENTIONAL WISDOM

Earlier we pointed out that there is no trend. But now we’ll say there may be an exception. What would a trend mean, anyway?

Suppose the chance that a particular stock will go up tomorrow is 50%. Now suppose on Tuesday the stock goes up. What are the chances it will go up on Wednesday? If Tuesday’s result and Wednesday’s result are independent of each other—that is, if there is no trend—then the chance it will go up on Wednesday is 50%, just like the chance on Tuesday. The fact that the stock went up on Tuesday hasn’t changed anything. Whether or not the stock goes up on Wednesday has nothing to do with whether it went up on Tuesday. Most studies and measurements have found that there was no tendency to trend. When you assess the chances of what will happen tomorrow, it doesn’t help you to know what happened yesterday.

But if there were a tendency to trend, then the fact that the stock went up on Tuesday would change the probability that it will go up on Wednesday. It will make it a little higher, say, 52%. More recent studies have, in fact, found some evidence of momentum in short-term price changes. That is, what goes up has a slight tendency to keep going up, and what goes down has a slight tendency to keep going down.

The momentum effect has been documented over short time periods, such as from month to month. But “reversion to the mean”—which means essentially “what goes up must come down,” the opposite of the momentum effect—has been observed over longer periods such as a few years. In fact, it must be observed over longer periods if there is a momentum effect over short periods.

Reversion to the mean says that at some time, momentum will not continue but will reverse. Observations of financial crises suggest that momentum is gradual and long-lived, but reversal is sharp, sudden, and unpredictable. You won’t get good long-run performance by relying on momentum.

What causes that momentum effect? That has not been studied very well, partly because the fields of finance and economics got stuck in “mainstreams,” which only study the same model over and over again with slight changes here and there. But the idea is something like this.

People, including professional investment managers, buy or sell securities not just by watching their prices but by watching one another. If they think other people are going to buy and drive the price up, then they’ll try to buy first, so they can take advantage of the rising price.

There’s an understandable tendency to think, “If I managed to run fast enough to the broker to buy before the price went up too much, then I should—shouldn’t I?—be able to run fast enough to sell later before the price goes down too much.” In other words, I’ll get in before the price goes up too much; then I’ll just yank my money out before the price drops.

What makes you think you can yank your investment out faster than the price drops? Obviously not everyone can succeed in doing that at once. You might succeed sometimes, but if you succeed only part of the time, you’ll be better off not even trying.

Then why is “Monitor your investment performance” such deeply embedded conventional wisdom? The standard procedure is to compare the return you have experienced in your investments to the return on a market index. But for what purpose? If beating the market index is something that can’t be done by skill but only by luck, what could possibly be the purpose of monitoring whether you did it or not?

If you were gambling at Las Vegas or Macao and monitored the performance of your bets on, say, a slot machine, what would be the purpose? Would it be to determine whether you should switch slot machines? If you did that, you’d probably be going from slot machine to slot machine and not getting any better results. Well, that’s what happens when you monitor your investment results.

When you play the slot machines, the only thing that really matters is how much money you have left and whether you really want to bet any more of it. The history of your performance doesn’t matter. It’s similar with your investments. What matters is how much money you have now and what investment strategy will increase or safeguard it in the future. The history of your performance is irrelevant.

That’s why we told you, when you hear the noise saying you must monitor your funds’ past performance and yet you are not monitoring your funds’ past performance, tune it out!

SUMMARY OF DEADLY TEMPTATION #1

1. Beating the market is impossible except by chance: future investment performance cannot be predicted, except for the impact of fees.

2. Monitoring past investment performance is pointless.