The Big Game Hunter: A Forty-Year Defense of Trend Following

Fortune and misfortune turn into one another and give birth to one another; the change is hard to see.

—— Huainanzi, “On the Human World” (Western Han dynasty; translated by the author)

What This Episode Is About

The August 25, 2026 episode of The Meb Faber Show (#647) brings back Jerry Parker. He was one of the first traders trained by Richard Dennis in the Turtle experiment of the early 1980s, founded his own firm in 1988, and has been doing essentially the same thing for close to four decades.

Meb opens with summer winding down and a story about the last time the two of them hung out in Tampa, with wrestler Ric Flair at the next table — then turns to the real subject: what trend following actually is, and what the asset management industry has turned it into.

Listening through, this is less an introduction to a strategy than a defense of one. Parker spends most of the hour drawing boundaries: managed futures is not trend following, the crisis-hedge pitch is corroding the strategy from inside, and the instinct to lock in gains is one of the worst ideas the backtests ever exposed. He also admits, with some humor, that the posts of his that get quoted back at him are usually written late at night or early in the morning when he is fired up — “overstatements, usually, just to get people going.”

The Main Points

1. Managed futures is a large bag; trend following is one compartment in it. Parker is blunt about this. The managed futures industry contains mean reversion, carry, pattern recognition, short-term trading, machine learning — trend following is a subset. It used to be one hundred percent of what CTAs did; now for many it may be half, or three quarters. So when people hear “managed futures” and mentally substitute “trend,” they have already misread the label. His own definition is much narrower: let profits run, take small losses, diversify like crazy. That is the whole thing.

2. A win rate under forty percent is a feature, not a defect awaiting repair. In pure trend following the per-trade win rate runs around forty percent or lower, and roughly five to ten percent of trades in a year carry all the money and pay for everything else. He has been running this formula since 1983. Tell someone you have a strategy where a tiny minority of positions produce all the gains and it sounds broken — but you have also just described the market-cap-weighted index.

3. The index is the world’s most naive trend follower; it simply never exits. Meb makes the case: cap weighting means you own more of a thing as it rises and less as it falls, until it eventually drops out. That is trend following at its core. Parker accepts the analogy and then supplies the crucial difference — the index does not enter and exit markets. Something in a downtrend just gets a smaller weight, and when everything falls together, nothing changes. He offers a self-deprecating example: he bought Tesla right before its first big move, and the index added it near the top. “So it’s sort of a trend following thing, but it doesn’t do a very good job of it. And yet it still does pretty well.”

4. The sign is actually reversed between the two. Buy-and-hold research says roughly four percent of stocks generate all of the index’s excess return, while the other ninety-six percent do not beat T-bills over a lifetime. In trend following, Parker says, it is the other way around: about ninety-six percent of the markets they trade end up contributing to the bottom line, and only a small share do not. And even that small share is provisional — cocoa contributed nothing for fifteen or twenty years, then had one enormous move that turned it into a contributor. Two very different distributions, routinely conflated.

5. The “crisis alpha” pitch is a kind of Stockholm syndrome. This is the hottest stretch of the episode. The industry’s assigned role for a CTA is: your only worth is what you do for me when stocks crash. Some managers have genuinely adapted to that — deliberately choosing worse overall performance in exchange for looking good during a selloff, because that is what wins allocations. Parker points out that the logic collapses on itself: those same allocators are never willing to put more than five or ten percent into it, so even a fund purpose-built for the job cannot move the needle on their portfolio. His position is to refuse the trade, run the best strategy he can, and let the allocations follow.

6. Strict on losses, generous with gains — and human instinct has it backwards. Every position has a predetermined stop and a trailing stop. Losses should be small, fast, automatic. But the winning side needs room, which means sitting through large open profits shrinking into small ones, or occasionally into losses. Parker recalls that Dennis’s research method was to take whatever everyone believed was true and test it — and it was almost never true. He singles out “never let a profit turn into a loss” as one of the worst ideas ever. “The computer basically says that’s exactly where you want to be.”

7. They deliberately do not optimize the portfolio, and deliberately keep correlated markets. People hear “hundreds of markets and futures” and assume enormous risk; Parker says the opposite is true. When building the portfolio they do not overweight the markets that have historically trended well or underweight the ones that have not. Currencies, commodities, stocks and rates get roughly equal allocation, assembled loosely with common sense and a touch of correlation analysis. More counterintuitive still, they will trade crude and Brent, gold and silver — pairs that can run ninety percent correlated. The reason: the outlier may show up in only one of them. He points to 1987, when silver doubled or tripled while gold moved about twenty dollars.

8. The one thing trend following is worst at is the thing you care about most. Late in the episode Meb asks the question everyone wants answered: why didn’t we sell silver at 120? It looks so obvious in hindsight. Parker’s answer is the most honest passage in the hour — there is no rule for selling at those levels that survives a backtest with a respectable sample size. Then this: “Nothing is more important than today’s trade. And yet that’s the only thing trend following is really bad at. It has no interest in today’s trade.” The itch you have right now, he says, I have no way to scratch it. Meb adds the closer: 120 is obvious in retrospect, but wouldn’t you have said the same at 20, 40, 60, 80 and 100?

Going Further

1. “I’m underwater — do I cut or hold?” This question shouldn’t be answered now

The painful part of being underwater is not the loss itself. It is discovering that you have no rule written down in advance. So you relitigate the case every morning: it rallies and you think holding was right, it drops and you think you should have cut. Every day you use the latest price to decide whether to keep believing your original judgment.

The episode’s answer is mechanical, but its real value is that it moves the question to the correct point in time. For Parker, the exit exists the moment the position is opened: a stop placed at what he calls an optimal distance from entry — not too short-term, not too long-term. That sounds vague until you notice what it rules out. A stop is not placed where your pain threshold sits; it is placed outside the market’s normal range of movement. The first is a measurement of your mood and drifts daily. The second is a measurement of the market and is the same today as yesterday.

The more portable half comes next. What listeners actually ask, he says, is not when to get out but when to get back in — and his answer is: at the next breakout, the next new high. The existence of that re-entry rule is what makes the stop executable, because the real psychological cost of a stop is never the loss itself. It is the fear of selling at the exact bottom. Give that fear an explicit re-entry condition and it is structurally dismantled: you did not sell the asset, you sold this leg of the trend.

Conversely, if your position has neither a written exit condition nor an answer to “at what price do I concede I was wrong,” you are not making an investment decision. You are re-betting daily. You can reject the entire method here and still keep the question.

2. “Has this stopped working?” — the line between performance and process

The strategy lagged a little in 2025 (Parker is careful to note it was down one or two percent, not fifty), and Meb asks how he thinks about that. His answer detours somewhere interesting: he recalls Meb asking a professor how long it takes to know something is broken, and getting an absurdly long answer. That was Kenneth French. Sixty-some years.

The number reads like a dodge, but it makes a serious point: within normal market noise, using performance to judge whether a method has failed requires a sample far larger than anyone’s patience or career. Which means your one-year or three-year results carry almost no information about whether your approach is sound. You think you are running a scientific test; you are reading noise.

So what do you judge instead? Parker’s substitute is process: your feelings about how things are going should have no impact on the trades in the portfolio. He describes what he did during the recovery — nothing. He sat and watched. The same trades I made when I was losing money, I made when I was making it. The work was done in the research, years ago, plus the occasional small tweak.

There is a useful corollary for individual investors. If short-run performance carries almost no signal, then the only thing you can genuinely audit is whether you followed your own rules — and that produces a usable sample quickly. How many times did I break a rule this month? Which situations caused it? That is the data worth collecting. Following the same logic, Parker cites Cliff Asness on the practical precondition: size it right, don’t over-lever, don’t overtrade — because you have to be the last person to give up on a strategy you believe in, and you only make it to the end if your position sizing didn’t force you out along the way.

3. “Why do I take small gains and sit on losses?” — the asymmetry is installed backwards

This is the most common self-accusation among individual investors, and the episode offers a structural explanation rather than another instruction to overcome your nature.

Trend following is deliberately asymmetric: strict, fast and automatic on the loss side; loose, tolerant and spacious on the gain side. Parker is explicit that he doesn’t like volatility and doesn’t like losing trades, “but those are two different things” — volatility on winners must be endured, losers must be exited quickly.

Most people ship with the same asymmetry pointed the wrong way: strict on gains (any giveback triggers the urge to lock in) and loose on losses (there is always one more reason to wait a day). The right tail gets clipped by you, the left tail gets extended by you. And the right tail is where the entire return lives.

Parker extends the criticism somewhere less commonly discussed: many products are engineered to flatter that inverted instinct. Clients want smooth curves, high Sharpe ratios, something closer to a normal distribution — so the product is tuned toward that, cutting off the fat right tail and reducing skewness. That destroys everything trend following is made of, he says, and telling clients there is no opportunity cost, that you can have it both ways, “is seductive.”

This travels well beyond trading. Whenever someone offers you something that has it both ways — high return and low volatility, high growth and a wide margin of safety, flexible and stable — the question is not whether they are lying. It is which compartment the cost was moved into. Usually it still exists; it has simply been relocated somewhere you cannot see from where you are standing: liquidity, the tails, or time.

Worth Looking Into

• The Meb Faber Show, episode 647, August 25, 2026, hosted by Meb Faber with guest Jerry Parker
• Public accounts of the Turtle trading experiment (Richard Dennis and William Eckhardt, beginning 1983)
• Hendrik Bessembinder’s research on the small share of stocks producing all of the market’s excess return, referenced repeatedly in the episode’s buy-and-hold comparison
• Kenneth French’s public discussion of how long a sample must be before a strategy can be declared broken
• Cliff Asness’s writing on extended periods of underperformance and investor behavior

帶得走的一件事

One idea: write down your strictness about losses and your generosity toward gains separately and in advance — instead of asking one heart, in the moment, to decide two opposite things.

You are a different person when you are winning than when you are losing, and expecting those two people to make consistent decisions is the fundamental design error. Parker’s forty-year practice is not that he became calmer than everyone else. It is that he moved the decision to a moment when he was still calm.

An exercise for today, no investing required:

Pick something you are actively investing in that has no defined endpoint — a relationship, a job, a long-running project, a fitness or study plan. Take a piece of paper and write two sentences:

1. “If this happens, I concede it isn’t working and I stop.” Specific enough that someone else could judge it. Not “if it never improves,” but “if by the end of October they still haven’t reached out to me once,” or “if three consecutive months show no change in weight or capacity.”
2. “If this happens, I allow myself to continue, even when it looks bad.” This one is written for the future version of you who wants to bail early — decide now under what conditions ugliness is acceptable.

Put it away and set a date to reread it. You will find the first sentence is the easy one. Almost all of us can picture where we should quit; very few of us have decided in advance where we should endure. And in the things that actually change a life, nearly all the return arrives during the ugly middle.

Raw Markdown

This article is an educational discussion of investment method. It is not advice to buy or sell any individual security, offers no target prices, and does not analyze any current holding. Investing carries risk; make your own decisions or consult a qualified professional.