The fifty greatest trends in forty years of futures markets, and what they reveal about luck, selection, and the futility of picking winners
Warren Buffett, the most celebrated stock picker who has ever lived, is disarmingly honest about where his fortune came from. In his 2022 letter to Berkshire shareholders he wrote that the firm’s results were the product of “about a dozen truly good decisions,” roughly one every five years. Most of the rest, he admitted, had been no better than so-so, and a few outright disasters were rescued by large doses of luck. His partner Charlie Munger went further still. Strip out their fifteen best decisions, he said, and what remained was a thoroughly average record.
I find that admission fascinating, because it holds the whole problem of investing in a single breath. Almost everything depends on a handful of outcomes. That part is not in dispute. The question that separates one strategy from another is not whether it is true, but what you decide to do about it.
Buffett’s answer is selection. His edge is being right, in advance and with deep conviction, about which dozen businesses will become the winners. It is an extraordinary skill, and for that kind of edge concentration is not merely defensible, it is correct. The textbook may call diversification the only free lunch in investing, but if you can genuinely identify the flowers, scattering capital into your fortieth best idea only dilutes the few that matter. For the great selector, breadth is the enemy.
I am a trend follower, and I have no such skill, nor do I pretend to. I cannot tell you which market will produce the next great move, and the uncomfortable truth is that nobody else can either. So I do the opposite of selecting. I hold a broad universe of markets, apply the same rules to every one of them, and wait. The edge is not in the choosing. It is in being present, with a position on, when the rare and enormous trends arrive. For this kind of edge, breadth is not dilution. It is the entire mechanism.
So I went looking for the very greatest trends in modern markets. Forty years of daily data, sixty-eight markets, and a simple question. Where did the biggest trends actually happen, how large were they, and could anyone have chosen them in advance? I expected the answer to be full of familiar names. It was not.
How I went looking for the greatest trends
To find the greatest trends honestly I needed a universe broad enough to be fair, a way of spotting trends that owed nothing to hindsight, and a yardstick that worked across markets as different as gold and the yen, and across decades as different as the placid 1990s and the violent 2020s.
The universe is sixty-eight futures markets carrying more than forty years of daily history, spanning currencies, bonds, stock indices, energy, metals, grains and the soft commodities. I measured every move not in dollars or points but in units of the market’s own risk, dividing each day’s move by its recent average daily range. One of those units I call an R, and it lets a trend in 1980s rice stand fairly beside the 2024 cocoa squeeze.
Then, so the answer would not depend on my taste in indicators, I let four completely different trend following methods vote on every market, on every day. One watches for breakouts to new highs and lows. One follows the crossing of a fast and a slow moving average. One simply asks whether a market is higher than it was a year ago. One waits until three averages line up in order. Different logic, different speeds, no shared machinery. A trend, for this study, is a stretch of time where the ensemble of the four leans the same way, and its size is the risk adjusted distance they rode together.
The first thing the method told me was the most reassuring. Every single one of the fifty largest trends was found by all four methods at once. The giants were not delicate signals teased out by one lucky rule.
They were so large and so relentless that four unrelated mechanisms all climbed aboard the same moves.
Robustness, not cleverness, finds them.
One caveat before the findings. The size I quote, the captured R, measures the opportunity a diversified ensemble was sitting in, not the return of any fund. It ignores costs, slippage and position sizing. It is a way of seeing where the great trends were and how big they grew, not a track record. With that said, here is what forty years turned up.
The names that were missing
I expected the league table to be ruled by the great set-piece markets everyone remembers. The first surprise was not what was on it. It was what was absent. Here are the twenty five largest. The full fifty are at the end.
Read the list again. Your eye goes hunting for the names the financial world talks about every single day. Gold. Crude oil. The S&P 500. The Nasdaq. They are almost nowhere to be seen. Of the fifty greatest trends in forty years, only five came from the liquid, headline markets that attract the overwhelming majority of attention and capital. The other forty five were made quietly, in markets most investors have never once considered owning.
So where did they fall?
Metals and soft commodities between them account for thirty two of the fifty. The single biggest trend in four decades was not a commodity at all but a currency, a four year migration in the Brazilian Real. Behind it stood a roll call no one would ever assemble on purpose. Tin. Milk. Rice. Robusta coffee. Zinc. Lead. Not one of them is a market you would have set out to own. And that, rather than any merit of the markets themselves, is the point worth holding on to. The outliers did not appear where the attention was. They appeared where almost no one was looking.
These are not trades. They are migrations.
The surprises keep coming
The absent names were the first surprise. There were others, and each one chips a little further at any idea that you could have seen this coming.
The second. Fifteen of the fifty greatest trends were short trades. It is not only that you could not have named the market in advance. You could not even have assumed the direction. The largest moves were as likely to be collapses as booms, and a process that can only buy would have missed nearly a third of them outright.
The third is stranger, and it is the one most likely to lead you astray. The same markets keep reappearing. Tin does not show up once but six separate times, in distinct trends spread across the decades. Class III Milk appears five times. Cocoa appears five times. The pull is immediate and powerful. Perhaps these markets simply trend better than the rest, and the real lesson is to favour them, to tilt the portfolio toward the proven performers and quietly drop the rest.
It is the most tempting conclusion in the world. It is also wrong.
Why a trend follower keeps no favourites
Start with the mathematics. The enormous moves that fill this table are the tails of a leptokurtic distribution, the fat tailed behaviour that surfaces, sooner or later, in very nearly every market that exists. The capacity for a violent, sustained trend is not a special trait belonging to tin or cocoa or the Brazilian Real. It is a property of the distribution itself, and every market in the universe carries it. A quiet market today is not a market without tails. It is a market whose tail has not yet arrived.
And in any case, the sample is far too small to rank anything. Fifty trends, drawn from sixty-eight markets across forty years, is a vanishingly thin slice of evidence, nowhere near enough to crown any market a reliable trender. Tin appearing six times feels like a pattern the instant you see it. But scatter fifty rare events across sixty-eight markets and clusters exactly like that are simply what chance produces. The eye sees a signal. The honest statistician sees noise.
Which is why a trend follower keeps no favourites. Not as a discipline held against temptation, but as a plain consequence of the mathematics. If any market can produce the next great trend, and if no sample however long can tell you which one it will be, then ranking markets is not insight. It is a story told after the fact. The only defensible position, and the one the data actually supports, is to treat every market as equally capable of becoming the next outlier, to weight them by risk rather than by reputation, and to hold them all. It is the same conclusion the famous names pointed to, reached from the opposite direction: you cannot select the winners in advance, not because selection is merely difficult, but because there is nothing stable there to select.
“The outlier is a property of the system, not a property of the market.”
Anatomy of the giants
Numbers describe these trends. Charts let you feel them. Each of the five below shows the daily price as an open, high, low and close bar chart, with the entry and exit of all four systems marked and the leg each system captured drawn as a line. They are not offered as markets to seek out. They are simply what a great trend looks like, and a reminder of how the plain discipline of holding a position let the systems stay aboard for months and years.
The largest trend in the whole study was that Brazilian Real. Through the middle of the 2000s it ground steadily higher for more than a thousand trading days, lifted by a hunger for yield and a long commodity boom. The slow models climbed aboard near the start and simply held, riding it for four years and giving back only the final reversal as it unwound into the 2008 crisis. This is the true signature of a great trend. Not a violent spike, but a long, low volatility migration that rewards patience above all else.
Tin is one of the smallest and least watched contracts on the London Metal Exchange. From the middle of 2020 a supply crunch drove it from roughly fifteen thousand dollars a tonne to over fifty thousand, a rise of more than one hundred and twenty seven per cent, captured here at one hundred and twenty four R. All four systems were long within months of the start. Almost no investor would ever choose to hold tin. A trend follower holds it not because tin is special, but because a trend follower holds everything, and is therefore standing right there when any market, however unfashionable, does this.
The cocoa trend of 2023 and 2024 was one of the most extraordinary commodity moves of the modern era, as West African crop failures sent the price vertical. The systems entered near the lows in 2022 and held through a year of quiet accumulation before the explosive phase arrived. Notice that the straight line joining each system’s entry to its exit captures the net move, not the spike. The systems banked the trend and handed back part of the blow off top, which is the honest texture of trend following. You do not sell the high. You hold the trend, and you leave when it turns.
Here is tin again, two decades earlier, in an entirely different supply cycle. Not because tin is special, but precisely because it is not. Nothing in 2001 marked it out as the market to hold, and nothing about this trend made the one twenty years later any more foreseeable. The only way to have been there for both was to hold the whole universe, playing no favourites, through all the quiet years in between.
Zinc finishes the base metal story. Across 2005 and 2006 a global inventory squeeze drove it up more than one hundred and forty per cent in under eighteen months, a trend clean and ordered enough that even the most demanding of the four methods committed early and stayed. Like the others, it was obvious only once it had happened.
Why a handful of trends is everything
There is a deeper reason all of this matters, and it is not simply that the great trends are large. It is that the whole of the return is hidden inside a very few of them. Across the more than two thousand winning trends the process found over four decades, the best ten per cent accounted for forty one per cent of all the risk adjusted gain, and the best one per cent for nine per cent. A small number of trends carry the entire result.
This is the fat tailed character of markets seen from another angle, and it carries a brutal consequence. If a handful of trends produce most of the return, and if you can never know in advance which markets will host them, then missing them is not a small cost you can quietly average away. It is the difference between the result and nothing at all. The whole enterprise rests on being present for the outliers, wherever they choose to appear. Which leaves only one question that matters. If you cannot favour the right markets, can you at least narrow the field by choosing a sensible few?
The needle in the haystack
Here is the part that is easy to miss, and it sits at the very heart of the matter. Every act of selection is also an act of exclusion. To choose ten markets out of sixty-eight is not simply to hold ten. It is to say no to fifty-eight. And the lesson of everything above is that the next Brazilian Real, the next tin, the next cocoa, is just as likely to be hiding among the fifty-eight you turned away as the ten you kept. You will never even see the trend you excluded yourself from. It will simply happen, in a market you chose not to own, and you will read about it later.
Put a number on it and selection stops looking difficult and starts looking impossible. The single largest trend came from one market out of sixty-eight. To have ridden the five greatest you needed to be holding the four specific markets that happened to produce them. To have ridden the ten greatest you needed eight specific markets. The chance of picking those eight out of sixty-eight by luck is about one in seven point four billion. Riding just the top five, by holding the right four markets, is a one in eight hundred thousand event. These are not the odds of a hard task. They are the odds of a miracle.
The lower curve is the probability of holding every one of the eight markets behind the ten biggest trends, drawn against how many you choose. It hugs the floor until your selection swells to nearly the whole universe. You cannot reliably assemble the leaderboard by choosing. The upper curve, the chance of catching even one of the ten biggest, climbs faster, but it too rewards breadth above all. A ten market book has only about a three in four chance of being in even one of them.
The expected outcomes say it most plainly of all. A book of ten markets is sitting, on average, in about seven of the fifty greatest trends. A book of twenty, about fifteen. The full sixty-eight market universe is sitting in all fifty, with certainty, because it holds every market in which they occurred. A trend follower does not improve the odds of selection. A trend follower abolishes the need to select.
“The process supplies the breadth. The markets supply the winners.”
Two answers to the same question
Return, finally, to Buffett, because it is tempting to cast him as the opposite of a trend follower and that is not quite right. Buffett and the trend follower are answering the very same question, the question every investor is forced to answer because the returns of a lifetime hide inside a handful of outcomes. They simply answer it in opposite ways. Buffett solved the outlier problem by becoming extraordinarily, almost uniquely, good at recognising tomorrow’s outliers in advance. The trend follower solves the same problem by refusing to guess where they will appear, and holding everything so as to be there regardless.
That is why the one word, diversification, means opposite things to the two of them. For Buffett, breadth would dilute a rare gift for selection, and every extra holding is a quiet no to his best ideas. For the trend follower, breadth is the gift. Each additional market is not a dilution but another open door, another chance to be standing in the room when the outlier walks in. It is also why the trend follower keeps no favourites. To favour a market is to begin selecting again, to start quietly excluding, and exclusion is the one thing the evidence will not forgive. Take breadth away and you have not made the strategy safer. You have removed the only mechanism by which it works.
Robustness, not cleverness, found these fifty trends. Breadth, not foresight, will find the next one.
Because somewhere today, in a market no one can name in advance, a move is quietly beginning that will become one of the great trends of the next forty years. It may be a currency few people follow, a forgotten agricultural contract, a metal that rarely reaches the news. Or it may be the largest and most crowded market in the world. Nobody knows which, and nobody needs to. The only question that has ever mattered is not whether you can predict it.
It is whether you will already be there when the market nobody chooses becomes the market everybody remembers.
A note on the numbers. Captured R is an analytical measure of the size of historical price trends. It is not a return, a track record, or the performance of any account or fund, and it does not reflect transaction costs, financing, slippage or position sizing. The fifty trends discussed here are a small historical sample of extreme events, offered to illustrate a principle, not to identify markets likely to trend in future. This piece is for education and discussion only and is not investment advice. Diversification does not guarantee a profit or protect against loss, and trend following can endure long stretches of loss and underperformance. Past performance and historical analysis are not indicative of future results.
Appendix: The full fifty
The complete league table of the fifty greatest trends in the sixty-eight market universe, ranked by risk adjusted size. Ranks one to twenty five are on the left, twenty six to fifty on the right.
Richard Brennan writes on systematic trading, complex adaptive markets, and the philosophical foundations of trend following at atstradingsolutions.com. His books include The Fractals of Finance and Complex Adaptive Markets. The forthcoming Carved by Impossibility completes the trilogy.
Want the theoretical foundation for why trend following works?
The Fractals of Finance: Determinism, Adaptation and the Geometry of Markets
The book explores the full architecture of feedback, fat tails, and fractal structure in financial markets, and what it means for how we trade, invest, and understand risk.
Available now on Amazon in paperback, hardcover, and Kindle.
Want a practical field manual for trading trends and capturing outliers?
The Aussie Turtles Trend Following Guide: A Field Manual for Hunting Outliers adapts the timeless principles of the original Turtle traders into a systematic, rules-based approach for modern markets. Co-authored with Adam Havryliv.
Available now on Amazon in paperback, hardcover, and Kindle.