Victor Vescovo has stood on the summit of Everest, sat at the bottom of Challenger Deep and flown to space. He also runs a private equity business making some of its largest bets. When I interviewed him, the sentence that stayed with me was not about courage.
He described the boundary between mitigated risk and recklessness as "actually a very thin line. And it's even more dangerous because it moves. It moves based on conditions, it moves based on your own capability, it moves based on what other people are doing."
Bent Flyvbjerg, the Oxford professor who has assembled the world's largest dataset on megaprojects, has measured what happens when nobody watches that line. "Only 8.5% of projects meet or exceed their time and budget expectations," he told me. "A mere 0.5% achieve this goal within the allocated budget and timeframe."
Two ends of one problem. The explorer knows the boundary is unstable. The dataset shows what it costs when nobody tracks it.
Founders talk about risk appetite as if it were a fixed personal quantity, like height. It is not, in the same way that deciding well has more to do with process than with the information you wish you had. The people I have interviewed who take the largest professional risks almost never describe themselves as comfortable with risk. They describe machinery: for pricing exposure, staging it, and knowing in advance what would make them walk away. What separates them from the reckless is not nerve. It is arithmetic the reckless never do.
01Do serious risk-takers actually feel like risk-takers?
Almost none do, and those who say they do are usually describing something else.
Stephen Schwarzman has built Blackstone into one of the largest alternative asset managers in the world by making very large bets. When I asked him about it, he was blunt about the folklore. "There is a common myth that entrepreneurs are risk takers," he said. "If you ask entrepreneurs, they tell you they're not. Why would you take a risk on something you didn't think was going to work? Smart entrepreneurs create processes that virtually eliminate risk to the extent that it would be taken."
What that process looks like at Blackstone is unglamorous and entirely copyable. Roughly eight people sit around a table with written materials circulated days in advance. Everyone participates; nobody dominates. Each is required to discuss how the firm could lose money, and how it might not. The group attacks the weaknesses, assigns probabilities, and leaves with more questions than it arrived with. It may run three times before anyone decides.
Note what that design protects against. Not ignorance. Ego. It is built so no single confident voice can carry a decision, the failure mode that destroys firms run by people with a good record.
Perry Chen, who founded Kickstarter, gave me the discipline that makes a fair judgement possible afterwards. "Navigating complexity and uncertainty is critical to good leadership," he told me. "We have to be OK with saying 'I don't know…' and still making a sound decision with the information you have at the time." The standard he is describing is not whether the decision turned out to be right. It is whether it was the best call available given what was knowable when it was made.
That distinction is the whole thing. Organisations that judge decisions by outcomes teach people to avoid decisions. Organisations that judge them by process teach people to think.
02Why does exposure have to be staged?
Because tolerance for risk is a trained capacity, and untrained people cannot tell the difference between a risk they can hold and one that will break them.
Jeb Corliss is a wingsuit pilot who has flown through gaps in mountains and survived an accident that should have killed him. He described building fear tolerance to me in the most ordinary terms available. "Confronting fear is progressive," he said. "If you walk into a gym aiming to lift 300lbs overhead immediately, you're setting up for failure and injury. Instead, you start light, gradually increasing the weight."
David Blaine, whose entire profession is graduated exposure to discomfort, described the same method from the analytical side. "The way you eliminate fear is to approach your situation from every angle, almost as though it were a logic puzzle," he told me. "I like to think of it as approaching like a tortoise, not like a hare."
Neither is describing bravery. Both are describing a sequence. The business translation is the one founders skip: run the small version first, at a scale where being wrong is survivable, then commit the capital that would hurt.
Mike Maples Jr, co-founding partner of Floodgate, put the structure of it precisely. "You want to make the cost of failure low. Make bets that are small enough that failure doesn't hurt you badly but aim them on a trajectory where, if you are right, you're spectacularly right."
A risk you have priced can be sized. A risk you have fallen in love with can only be doubled.
03Where does the line actually move?
In three places, all of them invisible from inside.
Conditions change and the plan does not. The market that made the thesis true eighteen months ago has moved, but nobody has reopened the deck. This is the most common and most boring cause of expensive failure.
Capability changes, and confidence changes faster. Success is the more dangerous direction, because a run of good outcomes is indistinguishable, from the inside, from a run of good decisions.
And other people's behaviour changes what feels normal. When everyone in your sector carries a particular exposure, it stops registering as a risk at all. It becomes the baseline. That is the mechanism Flyvbjerg named when he told me that "the greatest risk often lies in people's biased perceptions of risk itself," which is why his answer is reference class forecasting: not what you think this project will cost, but what projects of this class have actually cost.
Vescovo's own rule is the cheapest protection available. "You should always have the ability to have a plan B: a way to get out," he said. "That's where a lot of people get into trouble in extreme physical situations, or even in investing: not having an exit avenue."
04Three ways the arithmetic gets abandoned
Three things, and they are all failures of the same kind: the downside was never written down.
The walk-away line is set too late, and by the wrong person. Whoever decides what would make you stop should be the person who has not yet spent anything. By month six, with fees paid and the decision announced internally, that person no longer exists. Write the conditions before you start: a spend ceiling, a date, a specific piece of evidence that would falsify the thesis. Then give somebody other than you the authority to enforce them.
Success stops the questioning. Robert O'Neill, the former SEAL Team Six leader, named this exactly. "The dangerous part of our work is complacency," he told me. "I got out of the Navy because I stopped getting adrenaline in gunfights. That's when it gets dangerous. You don't want to get killed because you get bored." The organisational version is the deal nobody interrogates because the last three worked.
The bet becomes an identity. Henk Rogers, who secured the worldwide rights to Tetris, described his own moment of this. "When Arakawa told me someone else was offering him the rights, alarm bells went off," he said. "So, without thinking it through, I got a tourist visa and flew to Moscow. I didn't consider the repercussions or how risky it could be. I just did it." It worked, spectacularly. It is also the behaviour that, in ninety-nine other cases, ends a company. A story that ends well is not evidence of a good decision.
05Is any of this an argument against instinct?
The honest objection is that some of the best businesses in the world would not survive this process, and it is a real objection.
Sir Richard Branson made it to me directly. "If I relied on accountants to make decisions, I most certainly would have never gone into the airline business, most certainly would not have gone into the space business," he said. He is right, and the answer is not to pretend otherwise.
The resolution is to be honest about which situation you are in. Structured interrogation works where a risk is priceable: where there is a reference class, comparable outcomes, and data a group can attack. It works badly where the opportunity is genuinely novel, because there is nothing to compare it to and the process manufactures false precision.
Most founders are not in Branson's situation. They are in a priceable one and treating it as unpriceable, because that framing is more flattering, the same self-deception that shows up in raising money without a comparable company to point to. The test is uncomfortable: can you name three organisations that attempted something structurally similar, and what happened to them? If you can, you owe yourself the process. If nobody has, you are in instinct territory, and the discipline that applies is Maples's: make the bet small enough that being wrong is survivable.
06Recklessness is the absence of a decision
Name the largest live commitment your business currently holds. Can you state its downside in specific numbers rather than adjectives? If not, write it down now, along with what would have to happen for you to stop. Then give the second one to someone who is not you.
The people who take extraordinary risks for a living are not braver than everyone else. They have simply built the habit of describing, in advance and in writing, the thing everyone else prefers to leave vague. Recklessness is rarely a decision to accept a large risk. It is the absence of a decision at all.
