Someone in the room had done the homework. They had pulled the record on every organization that had attempted the thing now sitting on the table, the same restructuring, the same platform migration, the same ambitious reskilling push, and the record was not ambiguous. Most of them had failed. Not failed dramatically. Just quietly missed the thing they set out to do, over budget, past deadline, with the original goal relabeled as phase one and the phase one nobody mentions again. The slide was clean. The room looked at it for a moment. And then the most senior person said the sentence that ends the inquiry before it begins.
"Yes, but our situation is different."
Everyone relaxed, because the sentence is true. Every situation is different. Yours has a different leadership team, a different culture, a different market, a different reason for believing that this time the pattern will break. The sentence cannot be refuted, which is precisely what makes it dangerous. The room did not ignore the evidence. It acknowledged the evidence, admired the evidence, and then exempted itself from the evidence in a single clause, and moved on to the part of the meeting it had wanted to reach all along.
The exemption nobody has to earn.
There is a difference between not knowing the odds and deciding they apply to someone else. The first is ignorance, and it is fixable. You put the base rate in the room, you show the failure pattern, you make the reference class visible. The second is not fixable by information, because the information is already there, on the slide, acknowledged. What the room does with "our situation is different" is not a failure of data. It is a decision, made so fast and so comfortably that nobody registers it as a decision at all, to treat a known pattern as a story about other, slower, less capable organizations.
And notice who those other organizations are, in the room's private account of why they failed. They are always a little behind, a little less committed, a little worse led. They lacked executive sponsorship. They underfunded it. They did not have the talent. Every explanation the room offers for someone else's failure is an explanation that conveniently does not reach us, because we are sponsored, we are funded, we are talented, which is to say we are the people in the room saying so. The reference class gets dismissed on the grounds that we are not really in it, judged by criteria we grade ourselves on.
This is the quiet machinery of it. The base rate describes what tends to happen to organizations attempting this. The room does not believe it is one of those organizations. No room believes it is average. The people who climb to the altitude where these decisions get made did not get there by assuming they were ordinary, and the confidence that carried them up is the same confidence that now assures them the pattern on the slide is a description of the ordinary, from which they are exempt by temperament and track record. The trait that earned them the authority to decide is the trait that quietly corrupts the decision.
And the culture rewards the exemption. Watch what happens to the person who says the base rate suggests this probably will not work. They are not thanked for calibration. They are filed as the one who lacks ambition, who does not believe, who is not a builder. The person who says our situation is different is the visionary. So the payoff is not symmetric. Bet against the pattern and win, and you are the leader who saw what others could not. Bet with the pattern and be right, and you are merely the person who talked the room out of greatness. Given those incentives, of course the room exempts itself. It would be strange if it did not.
Watch, too, which examples get reached for when someone finally defends the attempt. They name the one organization that tried this and won. The famous turnaround. The celebrated transformation. Held up as living proof that it can be done. And it can be done. One, out of some uncounted many, did it. But the room is not measuring itself against the average attempt. It is measuring itself against the single survivor it can name, because the survivor wrote a book and gave the keynote, and the failures wrote nothing and disappeared. The evidence that it is possible is real, and it is also the most selectively sampled evidence in the building.
The pattern was never about them.
Here is the part that should be uncomfortable. The base rate was never a prediction about those other organizations. It was a prediction about the thing itself, about what this kind of change does to any organization that attempts it, including the capable ones, especially the capable ones, because the capable ones are the only ones confident enough to try it at this scale. The failures in the reference class were not the incompetent exceptions to a rule that favors the strong. They were the rule. Each of them also had a room. In that room, someone also pulled the record, and someone also said our situation is different, and they were also sponsored and funded and talented, and they also believed it, completely, the way you believe it now.
You are not looking at a list of organizations worse than yours. You are looking at a list of rooms that sounded exactly like the one you are sitting in.
So the evidence does not get argued with. It gets outranked. "Our situation is different" never engages the failure pattern on its merits. It simply asserts a jurisdiction the pattern does not reach, and because the assertion cannot be falsified in the moment, it wins the moment. By the time the specific ways in which the situation was not different finally surface, the slide has been forgotten, the sponsor has rotated out, and the result has been relabeled, once again, as phase one.
What it would take to stay in the class.
Rebuilding this is not about respecting data more. Everyone already claims to respect data. It is about making the exemption expensive to claim. A room that wants to declare itself the exception should have to say, out loud and on the record, what specifically makes it different, in terms concrete enough to be checked later, and then name what it would expect to see in six months if it turned out to be wrong after all. Not a feeling of difference. A testable one. The moment you force the exemption to be specific, most exemptions evaporate, because most of them were never a reason. They were a mood. The few that survive the demand for specifics are the ones actually worth betting on.
This is the judgment almost no governance process is built to catch, because governance scrutinizes the proposal and never the room's belief about itself. That belief is where SSUNDAR does its work. In a simulation, under a clock, with the reference class sitting visible on the table, you can watch whether a leadership team engages the pattern or waves it away, whether it can state precisely why it is different or only insist that it feels different, whether the confidence in the room is earned by the specifics or simply inherited from the altitude of the people holding it. The failure pattern is cheap to show anyone. What costs something is watching your best people decide, in real time and without noticing, that the pattern is about everyone but them.
Every organization on that slide was also certain it was the exception. That is how it ended up on the slide.