A Masterclass on Leadership | Stanley McChrystal | Knowledge Project 132

KEYPOINTS

• 01

  Risk is threats times vulnerabilities, and you control mostly the vulnerabilities.

  McChrystal used to define risk as probability times consequence, then insure against it or dodge it. Now he treats it as threats multiplied by vulnerabilities. Threats are largely beyond our control, and we’re poor at predicting where and when they’ll come, so trying to swat every one away is ‘a fool’s errand.’ Vulnerability is where we have agency. He came to this after noticing that checklists and risk matrices hadn’t given him good outcomes, and that 9/11, the financial crisis and COVID-19 were all foreseeable but badly handled.

• 02

  Build resilience like an immune system: detect, assess, respond, learn.

  An immunologist he met at Yale told him we ingest a potentially lethal microorganism an estimated 10,000 times a day, yet rarely get sick. We’re in trouble only when the immune system is weakened. Organizations work the same way. His team identified ten ‘risk control factors’ that can be exercised like muscles: communication, a clear and shared narrative, the ability to act and overcome inertia, timing, adaptability, diversity of perspective, overcoming bias and leadership. Assume many threats will get through your detection and prepare to deal with them on the doorstep.

• 03

  Signals get lost in silos and in noise.

  Before 9/11 the government held the relevant intelligence, but it was a tiny fraction of everything flowing through the system, and it was split across departments. Like the blind men and the elephant, each saw only part. Agents who knew about suspicious flight lessons couldn’t link them to a hijacking plot, and he doesn’t think it’s fair to call them simply wrong. The lesson is to connect organizations to complete the picture while accepting you’ll never see it clearly. More data and AI help, but they won’t be a cure-all.

• 04

  Mismatched time horizons distort decisions.

  In Iraq and Afghanistan, the CIA built networks for the long haul, the military came in large numbers for short tours and pushed for quick results, and the State Department saw its job as reporting reality rather than changing it. Mike Bloomberg told him only elected executives tend to fund long-term infrastructure like New York’s post-Hurricane Sandy defenses; legislators on two- or four-year cycles rarely do. Commanders on two-year tours feel the same pull. He offers a parenting analogy: you’re not ‘having a baby,’ you’re having a person for at least 18 years. His son is 38 and lives next door.

• 05

  Certainty is impossible; set the decision time by working backward.

  As a new lieutenant, McChrystal asked his father how to tell who would be a good combat commander. You can’t in peacetime, his father said. In combat, the weak ones keep asking for more information, trying to drive uncertainty to zero, and freeze. McChrystal’s own flaw is deciding too fast, so he surrounds himself with careful ‘hand-wringers.’ His method: work out how long it will take to pass the decision down and carry it out, and count back to the last moment you can still decide effectively. Use the time until then to gather information. A mentor taught him to announce two phases. In phase one, everyone gives input. In phase two you execute, and only genuinely new information reopens the decision.

• 06

  Delaying a hard call is itself a decision.

  When someone says they’ll give a struggling colleague three more months, ask what you’ll know then that you don’t now. Usually nothing. Hard decisions don’t get easier with time; they get harder. In counter-terrorism work, officials along the chain would ask irrelevant questions and, by the time answers came back, the intelligence had gone stale. They never approved or rejected anything, so their hands stayed clean, yet the outcome was still a failure. Bureaucracies punish bad decisions and rarely reward good ones.

• 07

  If big failures end careers, people stop taking risks.

  The Army in his era often ended an officer’s rise over a single mistake, and Navy captains pay with their careers. Yet people learn most from getting ‘scuffed up.’ Executives say they want risk-taking, but ask them where the people who failed big are now and there’s usually silence. Everyone ‘goes to school’ on that. Judge decisions by two tests: were they consistent with your values, and did they have a reasonable chance of success? A bad outcome doesn’t prove a bad decision. Don’t let a lucky result make you think you’re a genius.

• 08

  Leaders must close the gap between the map and the ground.

  By the time reports reach the top, they’re ‘homogenized’; symbols on a map don’t get tired or scared. Drone video from 10,000–20,000 feet gave commanders a false god’s-eye view: you can’t feel how dark or cold it is or hear the bullets. When a staff sergeant emailed, ‘with all due respect, I don’t think you understand what’s going on out here,’ McChrystal flew down the next day. Outside Kandahar he spent a day patrolling a maze of six-to-seven-foot mud walls used for grape vines, where the enemy could fire down a single corridor and vanish. He called it the hardest terrain he’d ever seen tactically. He couldn’t drop the mission, but he understood it, and his troops knew he did.