Mandelbrot
“A forest fire doesn’t start because lightning exists. It starts because the forest was already full of dry brush.” —David Dredge
Lightning hits the ground thousands of times a day. Most strikes disappear into dirt. But one strike hitting a stretch of accumulated branches, leaves, and tinder becomes a catastrophe.
The event is the same. The outcome depends entirely on where it hits.
That idea stuck with me longer than it should have. Mostly because it forced me to confront something uncomfortable: I spend far more time thinking about “lightning events”: Fed meetings, CPI releases, earnings surprises, than I do about the actual structure of my risk.
We obsess over possible shocks and ignore the positioning that determines how those shocks affect us.
risk lives in the structure, not the spark.
So, I started looking, not at the sparks, but at the structure itself.
What This Taught Me About Trading and Investing
When I first heard the forest-fire analogy, I realized how much of my mindset had been built around the wrong focal point. I spent years scanning the sky for lightning: CPI prints, FOMC meetings, earnings weeks, as if anticipating the next spark was the key to avoiding damage. It felt responsible. It felt analytical. It felt like the adult way to manage risk.
But most of the time, it wasn’t the spark that decided whether I took a hit or walked away unscathed. It was the condition of my portfolio beforehand. Two people can face the same event and walk away with completely different outcomes. The difference isn’t foresight; it’s structure.
When I look back, I can see how fragility crept in through decisions that felt harmless at the time. There were stretches where I let position sizes grow simply because they had been working, or where I convinced myself, I was diversified because the tickers looked different even though they all leaned on the same set of assumptions. I took comfort in option structures that appeared elegant on paper but hid asymmetry I didn’t fully appreciate. I treated margin as something abstract until volatility reminded me it wasn’t. And I often confused calm markets for understanding, mistaking a period of stability for evidence that my approach was sound.
None of this felt dangerous when I was doing it. That’s the unnerving part. Fragility rarely announces itself. It builds slowly and quietly, and because nothing bad happens right away, you convince yourself the structure is sound. I wasn’t intentionally taking outsized risk; I was just not looking in the right places.
Understanding this shifted my attention away from the next headline and toward the underlying conditions that make headlines matter. It’s not that events stopped being relevant. It’s that I started seeing them for what they are: matches. What matters far more is whether I’ve been piling dry brush under my feet.
Endogenous Risk — What It Is and Why It Matters
The more I thought about the dry-brush analogy, the more it clarified something I had been circling for years without fully articulating: most of the risk we face isn’t external at all. It’s internal. The financial system doesn’t usually break because of an unexpected shock; it breaks because the shock lands on a structure that was already fragile.
This is the essence of endogenous risk: risk that develops from within the system itself through the behaviors, incentives, and positioning of its participants. It’s the kind of fragility that doesn’t require a dramatic trigger. A small shift can cascade simply because the network is tightly coupled, crowded, or built on assumptions that only hold in calm environments.
Endogenous risk shows up in the quiet corners: leverage that isn’t obvious at first glance, trades that all lean in the same direction, diversification that works until correlations snap, yield-seeking structures that suppress volatility, and portfolio constructions that rely on conditions staying inside a comfortable range. None of these things look dangerous individually. They become dangerous through accumulation.
The unsettling part is how gradually this risk forms. It often builds in periods of stability: the very moments when markets feel safest. That’s the paradox: the smoother the surface becomes, the easier it is for fragility to take root underneath it.
Recognizing endogenous risk doesn’t solve it. But it changes the question. Instead of obsessing over what the next catalyst might be, I’m trying to understand how my own structure behaves when something, anything, disrupts the calm. Events will always come. The more important work is understanding the conditions that decide whether an event fizzles or spreads.
The Retail Trader’s Dry Brush
Understanding endogenous risk in theory is one thing; recognising where it shows up in my own portfolio has been something else entirely. The concept made sense to me long before I could actually see how it lived in the day-to-day details of my positioning.
Fragility doesn’t sit politely in one corner waiting to be identified. It shifts. It hides. Sometimes it only becomes visible in hindsight. Most evenings, I log into my account uncertain whether I’ve managed to spot the real vulnerabilities, but I try to pay closer attention to what’s quietly accumulating beneath the surface.
Concentration That Doesn’t Announce Itself
I’ve told myself many times that my portfolio is diversified because the tickers look different. But whenever markets actually stress, the behavior tells a different story. A lot of my positions seem to lean on the same macro assumptions, even if they weren’t chosen with that intention.
I don’t always know which similarity matters most, but I’m starting to notice that “different symbols” doesn’t automatically mean “different risks.”
Correlation I Take for Granted
Diversification has a reassuring logic to it, but the more I look at how correlations behave under pressure, the more I realize how conditional the whole idea is. Things that drift apart in quiet markets often snap together when volatility rises.
I don’t have a formula for predicting when that will happen. I just know it happens more often than I assumed.
Yield That Might Be Hiding Something
I’ve used income-generating strategies because they feel steady and disciplined. Covered calls, structured notes, premium-selling: they calm the portfolio in normal periods.
But every time I think about Dredge’s point on endogenous risk, I can’t shake the suspicion that these smooth-looking lines are often built by selling the one thing you need in a crisis: convexity.
I’m still figuring out how to judge which yield is earned and which is manufactured.
Leverage That Doesn’t Look Like Leverage
I’ve never taken big, obvious leverage. But I still catch myself wondering if I’ve taken on leverage in subtler ways: through options, through concentrated exposure, through complexity I haven’t fully unpacked.
Some risks don’t come with a margin interest line item. They just behave like leverage when it matters most.
Position Sizing That Creeps Up
One thing I’m still learning to monitor is how easily a position grows beyond what I intended. A winner becomes comfortable, comfort becomes confidence, and suddenly the size no longer matches the level of understanding.
I can’t say I’ve solved this. I’m just trying to be more aware of how quietly it happens.
Strategies That Depend on “Normal” Conditions
Every portfolio has trades that only make sense if volatility stays low or correlations behave or liquidity remains available. I still hold some of these. I don’t think anyone can avoid them entirely.
But I’m more conscious now of how many strategies, including most of my own, only work as long as the environment stays inside a narrow band.
And environments don’t always ask permission before shifting.
What I Got Wrong About Diversification
I have often treated diversification as a kind of default safety mechanism. Spread your bets, mix your exposures, let correlations do their smoothing work. And on the surface, it is. But the more closely I look at how diversification behaves in real markets, the more it feels like something I understand from my CFA textbooks but not fully in practice.
The first crack in the idea came from watching correlations behave badly. I used to assume they were stable characteristics of assets, as if gold, bonds, equities, commodities all had fixed relationships to one another. Then I started paying attention to how quickly those relationships change when volatility picks up. Assets that drift independently in quiet markets suddenly move in unison during stress. Two things that look unconnected on a back-test can become indistinguishable when liquidity disappears. I don’t have a formula for predicting when correlations will collapse; I just know now that they aren’t the anchor I thought they were.
Another thing that’s become hard to ignore is how much diversification depends on the environment. Historical data makes it seem like uncorrelated assets are naturally uncorrelated, but so much of that separation is born from a specific period: low inflation, predictable central bank behavior, stable global liquidity. Once any of those conditions change, the old patterns lose their reliability. I’m still trying to understand which environmental shifts matter most, and I’m still not sure how to judge them in real time.
There’s also this quieter issue of crowding. When too many people diversify using the same playbook, the whole structure can behave like one large position. You end up with different instruments tied to the same risk factors, even if they look distinct on the surface. I don’t have a precise method for measuring crowding, but I’m much more aware now of how strategies that appear unrelated can actually be part of the same story.
And then there’s the part I missed for years: diversification reduces volatility, but it doesn’t automatically reduce fragility. A portfolio can look calm while still containing a vulnerability that has never been tested. That calmness used to reassure me; now it mostly prompts questions. Is the structure truly robust, or is it simply untested? Am I diversified, or am I holding positions that will only separate when it doesn’t matter and converge when it does?
I haven’t abandoned diversification; it still helps me manage noise, but I’m less confident in the comfort it used to provide. I’m learning that diversification is conditional, not guaranteed, and that the conditions that support it can shift before I notice. I’m trying to move away from the assumption that spreading capital across different tickers solves the problem, and toward a more uncomfortable question:
What happens if the things I believe will protect me all fail at the same time?
What Happens Most vs What Matters Most
One of the quieter realizations I’ve had, while thinking through all of this, is how much of my early approach to risk was built around the middle of the distribution: the part of the market that behaves itself. Most days are normal. Prices drift. Volatility stays within familiar bounds. Strategies that lean on stability tend to look sensible because the world, most of the time, is stable.
The problem is that compounding doesn’t care about “most of the time.” It cares about the handful of moments that fall outside the comfortable range.
This is the part I hadn’t fully internalized. A strategy can win on eighty or ninety out of a hundred days and still end up behind because the losses on the outlier days take back everything the inside-the-range days delivered. The average return still looks respectable on paper, but the actual wealth path, the lived experience of compounding, tells a different story.
People focus on what happens frequently instead of what matters. It’s easy to build a portfolio that performs well in normal conditions, because normal conditions don’t ask very much of you. But when conditions shift, frequency stops being a guide. Magnitude takes over. The one or two large moves shape the outcome far more than the dozens of small ones.
And this is where the dilemma shows up for someone like me. I understand the logic that buying convexity: some form of protection, is structurally superior to selling it. Selling volatility works until it doesn’t. Buying volatility bleeds until it pays. The argument makes sense. But understanding the principle is not the same as managing it in practice.
Buying protection consistently means accepting a drag on returns, and that drag is real. It’s not theoretical. Even when I know that protection improves the long-term path, I still feel the sting of watching capital erode slowly when nothing dramatic is happening. There are days when it feels like paying for fire insurance during a season of rain. Rational, but uncomfortable.
What I haven’t figured out yet is the part that seems to distinguish the genuinely skilled from everyone else: how to carry convexity without bleeding to death. How to balance robustness with realism. How to design a structure that respects tail risk without turning the whole portfolio into a cost center.
I don’t think there is a universal answer. At least, not one that survives contact with different personalities, different goals, and different tolerances for pain. But I’m trying to pay more attention to the gap between “what works often” and “what survives long enough to matter.”
It’s a shift in mindset more than a shift in strategy. Instead of asking whether something is likely, I’m trying to ask what it would mean if it happened at the wrong time. Instead of asking whether a trade looks good on average, I’m trying to understand how it behaves at the edges.
I’m a long way from mastering this. But the distinction feels important.
Most of the market lives in the middle. Compounding lives at the tails.
Understanding that difference hasn’t given me certainty. It’s given me better questions.
Closing Reflections
The more time I spend sitting with these ideas, the more I realize how much of risk management is about learning to see properly. Not predicting better, not optimizing harder, just noticing the structure that sits underneath whatever the market happens to be doing on the surface.
I don’t think I’ll ever have a perfect sense of where the dry brush in my portfolio is. Some of it is visible in hindsight, some of it becomes obvious only when volatility exposes it, and some of it probably lives in places I haven’t learned to look yet. But paying attention to the idea has already changed the way I think.
Instead of asking whether the next event will be good or bad for me, I’m trying to understand how my positioning would react either way. Instead of relying too heavily on correlations or patterns that held in the recent past, I’m trying to notice when those assumptions start feeling convenient rather than true. Instead of treating calm markets as a sign that my structure is sound, I’m trying to remember that calmness can also mean the system hasn’t been tested.
This isn’t a finished framework. It’s just a better question: What part of my structure is vulnerable in ways I haven’t accounted for?
Maybe that’s all you can really aim for. Not perfect foresight, but a clearer sense of the conditions that matter. The spark will come from somewhere, sometime, in some form. That’s always been true. What I’m learning, slowly, is that the outcome depends far more on how I’m positioned long before the spark arrives.
I’m not trying to eliminate uncertainty. I’m trying to respect it. And that feels like a more honest starting point than the belief that risk can be solved once and for all.