Form 7-C, internal risk review, third quarter. Field 14 asks: "List evidence reviewed prior to recommendation." Beneath it, in careful handwriting, a researcher named Tomás Vreeland had written six sources. All six supported the project he was championing. The field didn't ask whether he had looked for evidence against it. Neither had he.
The feeling of thinking carefully and the feeling of defending a conclusion you already hold are, from the inside, identical. That is not a flaw in your character. It is a flaw in the instrument. The mind that wants an answer will find the path of least resistance to that answer, then narrate the journey as due diligence. Evidence gets evaluated for whether it helps, not whether it's true. The process looks like reasoning. It is advocacy.
The feeling of thinking carefully and the feeling of defending a conclusion you already hold are, from the inside, identical.
Philip Tetlock's decades-long forecasting research, detailed in his book Superforecasting, documents this at scale. The worst predictors weren't unintelligent — many were experts. They were wrong because they treated incoming information as something to survive, not learn from. The best forecasters shared one habit: they updated when the data moved, even when updating was embarrassing. They were trying to track reality. Everyone else was trying to win an argument with it.
The hour before a significant decision is when this matters most. You have reviewed your options; you feel ready. Try one thing instead: name, out loud or on paper, the single piece of evidence that would reverse your choice. Not a list of risks you've already discounted — the one finding that would actually stop you. If nothing comes to mind, you are not deciding. You are ratifying. That distinction is where Tomás Vreeland's form went quiet, and where yours can do something different.
You will not always reach the right answer. But you can stop mistaking confidence for accuracy. Those are not the same thing, and knowing that gap exists is the first move toward crossing it.