Why Risk Management Is So Important — Even When You're Right

Ask ten traders why risk management matters and most will say some version of "to avoid losing money." That's true, but it's not the real answer.

The real answer is this: risk management is what determines whether being right actually pays off. A correct trading thesis with no risk management doesn't survive long enough to matter. A mediocre strategy with real risk management can compound for years.

A few days ago, a hedge fund proved this in the most expensive way possible — by being right about the idea and still losing almost everything.

The $45 Billion Fund That Was Up 439%

Leopold Aschenbrenner was fired from OpenAI's Superalignment team in April 2024 — officially over an alleged information leak, which he disputes. Two months later, he published a viral 165-page essay, "Situational Awareness: The Decade Ahead," arguing that human-level AI was coming around 2027 and that whoever controlled the physical infrastructure underneath it — chips, power, data centers — would capture enormous value.

The essay made him well-known enough that serious money took him seriously. He launched a hedge fund on the same thesis, naming himself Chief Investment Officer in his early 20s with no prior financial track record. His earliest backers weren't nobodies: Patrick and John Collison (the founders of Stripe), former GitHub CEO Nat Friedman, and investor Daniel Gross.

For a while, it worked spectacularly. The bet wasn't on the obvious mega-cap AI names — it was on the picks-and-shovels layer underneath the boom: AI infrastructure names like Nebius, SanDisk, Micron, and CoreWeave. On the other side of the book, the fund shorted legacy software companies it expected AI to disrupt, including Adobe.

Here's the important part, and the one that's easy to miss in a story like this: his thesis may well be right long-term. This isn't a story about a bad idea. It's a story about what happens when a correct idea meets no plan for surviving the drawdown in between.

What Actually Went Wrong

In May 2026, a new hawkish Federal Reserve chair was sworn in — a signal generally bad for the kind of high-growth, speculative names the fund was loaded up on. Late July brought a broader AI-stock selloff, with the fund's positions falling more than 35% in a single month in several names.

On July 24, Aschenbrenner sent investors a letter. He called the dip "the best entry point for AI investments in over a year," pointed to a possible Anthropic IPO as a future catalyst, and gave existing investors until August 1 to add fresh capital. In other words: asking for more money to hold the position, days before it became impossible to hold at all.

"Being right about a trade and surviving a trade are two completely different problems. Leverage decides which one you're actually playing."

The fund's own account of what happened next, from a statement to investors: the drawdown accelerated because of "increasingly adverse trading in names publicly associated with us" — comparing it to a bank run, where vulnerability begets more vulnerability. Before market open on Thursday, July 30, the fund exited its entire public equity book — both longs and shorts — through a single transaction. Roughly two-thirds of the fund's total assets were public equities at that point. All of it, gone in one trade.

The buyer, confirmed by both the Wall Street Journal and CNBC: Citadel, Ken Griffin's firm. By late morning that same day, most of the affected AI-infrastructure names were up 15-20%. The fund kept one thing untouched — a private stake in Anthropic, worth roughly $5 billion — the one part of the portfolio that carried no leverage.

Aschenbrenner's own statement ended with: "I take full responsibility for these events." That's a rare thing to see from a fund manager, and it points at the actual lesson here — not that AI infrastructure was a bad bet, but that a 40% concentrated position, on 4x leverage, with no protection for the gains already made, was always one bad month away from this outcome.

The 5 Things Real Risk Management Actually Looks Like

It's easy to nod along to "manage your risk" as advice. It's much less obvious what that actually means in practice. Here's the structure, drawn from the same mechanics this fund skipped:

1. Rule 1 — Size positions according to volatility

   Risking the same size every day regardless of how calm or violent the market is right now is how a normal bad week becomes an account-ending one. Size down when volatility is elevated, size up when it's calm — same edge, less exposure to the moves that actually cause forced liquidations.

2. Rule 2 — Set a leverage ceiling and never override it when confident

   4x leverage on a diversified book is a different animal than 4x leverage on a single 40%-concentrated position. The leverage ratio matters far less than what it's applied to — and the ceiling has to hold precisely when the thesis feels most obviously right, which is exactly when discipline is hardest to keep.

3. Rule 3 — Cap concentration so no single position can sink the account

   A 40% position in one name isn't a portfolio, it's one trade wearing a portfolio's clothing. When it works, it works spectacularly. When it turns, there's nothing else in the book to absorb the hit.

4. Rule 4 — Build a plan to protect winners, not just cut losers

   Most traders have a stop-loss mentality for when they're wrong. Far fewer have a plan for protecting gains once they're up 439%. A trailing stop, a scaling-out rule, or a drawdown ceiling that tightens as profit grows all serve the same purpose: locking in what's already been won instead of leaving it all exposed to the next bad week.

5. Rule 5 — Use external rules that don't bend on a hot streak

   This is exactly what a funded evaluation forces on a trader whether they like it or not — max drawdown limits, daily loss caps, consistency rules — imposed by someone other than the trader, so conviction never gets the chance to override discipline. It's a structural fix for a problem that willpower alone rarely solves.

How Prop Firm Rules Force This Discipline For You

Everything above is exactly what a prop firm evaluation builds into the account structure itself. Max drawdown rules, daily loss limits, and consistency requirements aren't bureaucratic friction — they're the same protective mechanics this fund needed and didn't have, just enforced externally instead of relying on discretion in the moment.

That's the same structured-leverage principle behind the NIG Daily Profit Accelerator: leverage built into a defined, rules-based structure, not left to individual judgment under pressure. If you want the fuller breakdown of how margin, drawdown, and leverage actually work mechanically, that's covered in the companion piece: Is Leverage Trading Dangerous? A Real Market Case Study Inside.

The NIG Leverage & Prop Firm Playbook

Free comparison of how different prop firms structure drawdown, consistency, and leverage rules

Free

How I Actually Trade This

If you'd rather have this discipline built into a system than execute it manually yourself, that's exactly what the NIG Daily Profit Accelerator is built around — the same structured, rules-first approach, without needing to read order flow yourself.

Free Tools to Put This Into Practice

Frequently Asked Questions