7:26 Lena: You know, Miles, thinking about that "sure thing" mentality we all fall into, it’s actually a bit dangerous, isn't it? When we frame decisions as "will this work" or "won't it," we’re essentially setting ourselves up for a trap. I was reading a piece on "The Illusion of Certainty," and it argued that binary thinking actually destroys our ability to learn from experience.
7:48 Miles: It really does. Annie Duke, the poker player, calls it "resulting." It’s that tendency to judge the quality of a decision based solely on the outcome. If you make a move that has a 70 percent chance of success and you hit that 30 percent failure, people call it a "bad decision." But in reality, the process was sound; you just encountered the downside of the distribution.
8:10 Lena: Right! It’s like if you run a red light and get through safely, binary thinking says, "Great decision, I got home faster." But probabilistic thinking says, "Terrible decision, you just took a massive risk for a tiny gain." The outcome was good, but the process was a disaster.
1:56 Miles: Exactly. And the reverse is true, too. You can do everything right—gather the data, check the base rates, consult the experts—and still fail. If you don't think in probabilities, you’ll look at that failure and say, "I’m a failure, my strategy was wrong," and you might throw out a perfectly good system just because of one bad roll of the dice.
8:45 Lena: That’s where "calibration" becomes so vital. I was fascinated by the idea that you can actually track how "calibrated" you are. Like, if I say I’m 60 percent sure about something, and I track 100 of those predictions, do 60 of them actually happen? If only 40 happen, I’m overconfident. If 90 happen, I’m actually underconfident and playing it too safe.
9:08 Miles: It’s a literal measurement of how well your internal "certainty meter" matches reality. And most of us are wildly overconfident. We say things are "almost certain" when the actual odds are maybe 75 percent. In the professional world, that gap is where the biggest mistakes happen—investing too much in a "sure thing" that was actually a coin flip.
9:28 Lena: And that’s why the sources suggest replacing vague words like "probably" or "maybe" with actual numbers. One person’s "probably" might be 60 percent, while another person’s is 90 percent. If we’re sitting in a meeting and I say, "This feature will probably increase conversions," we might think we’re agreeing, but we could be miles apart on the actual risk we’re taking.
9:49 Miles: It’s a communication breakdown waiting to happen. Using percentages forces precision. It forces everyone to show their cards. If I say I’m 70 percent confident, you can ask me, "Okay, why not 90? What’s the 20 percent you’re worried about?" That’s where the real insight is—in the gap between "pretty sure" and "certain."
10:08 Lena: It also helps with what's called "outcome bias." If we’ve documented that we knew there was a 30 percent chance of failure, and then it fails, we don't have to panic. We can say, "Hey, we knew this was a possibility. We planned for it. Our process is still intact." It saves so much emotional energy and prevents that knee-jerk reaction to "fix" something that isn't actually broken.
10:33 Miles: It’s about building a "decision journal." I’ve started doing this lately. Before I make a big call, I write down my reasoning, my confidence level, and what information would actually change my mind. It’s the only way to beat hindsight bias. You know, that "I knew it all along" feeling we get after the fact?
10:51 Lena: Oh, we’re all guilty of that! Once we know the outcome, our brains rewrite history to make it seem inevitable. We forget how much uncertainty we were actually feeling at the time.
1:56 Miles: Exactly. The journal keeps you honest. It shows you exactly what you were thinking *before* you knew the answer. And that’s how superforecasters get so good. They aren't smarter than us; they just have a more accurate record of their own mistakes. They can see the patterns in their own overconfidence and adjust.
11:20 Lena: It reminds me of the "Interview Problem" I saw in one of the articles. We usually treat an interview as a definitive "hire" or "don't hire." But an interview only has maybe a 60 or 70 percent predictive power for how someone will actually perform. If we think of it as a "70 percent bet," we might still hire them, but we’ll also set up a clearer 90-day review or a backup plan, because we’ve explicitly acknowledged the 30 percent chance it might not work out.
11:48 Miles: It’s about "de-risking" the decision. When you stop pretending you’re 100 percent certain, you actually become more resilient. You stop putting all your chips on one square. You start thinking in "distributions" of outcomes.
12:01 Lena: And distributions are so much more realistic than "point estimates." Like, instead of saying, "This project will take eight weeks," which is almost always wrong, you say, "There’s a 10 percent chance it takes six weeks, a 50 percent chance it takes ten weeks, and a 10 percent chance it takes sixteen."
12:17 Miles: That "P10/P50/P90" framework is a game-changer for planning. It accounts for the asymmetry of the world. Projects rarely finish early, but they often overrun by a lot. If you only plan for the "likely" eight weeks, you’re essentially ignoring the entire "tail" of the distribution where things go sideways.
12:36 Lena: It’s about being "realistic without being paranoid," as one source put it. You plan for the P80 outcome—the one that happens 80 percent of the time. That way, you’re building trust with your team and your clients because you’re actually hitting your targets, rather than constantly apologizing for "unexpected" delays that were actually quite predictable if you’d looked at the distribution.
12:58 Miles: And it filters into everything—risk assessment, resource allocation, even how we judge people’s competence. We move away from "Is this person good?" to "What is the distribution of their performance in this specific context?" It’s a more compassionate and more accurate way to see the world.