How to Deal With Mistakes (Including the Ones From People You Trusted)

There is no equation in life with a direct solution. Almost every real problem — in health, business, relationships, personal change — requires what mathematicians call numerical methods: an approach that converges toward an answer through iterations, each one correcting the errors of the last.

Mathematicians have known for centuries that some problems cannot be solved perfectly. The integral sine function, for example, has no closed-form solution. You cannot simply compute an exact answer. But you can solve it to any required precision — ten decimal places, twenty, fifty — depending on what your actual application demands.

Most decisions you make in life require far less precision than you think.

The Error Percentage Problem

No information source is error-free. Not books, not doctors, not researchers, not experienced coaches, not long-running channels with demonstrated track records. The question is not whether a teacher makes mistakes — they do — but what percentage of their output is incorrect and whether that percentage is small enough that following their work still produces significantly better results than not following it.

A doctor who is wrong 5% of the time and right 95% of the time is an extraordinary doctor. A system that produces measurable improvement in 90% of those who follow it is an extraordinarily effective system. The existence of error within a system is not proof that the system fails.

The relevant comparison is not "does this method have errors" versus some imaginary perfect method. It is: does this method, with its errors included, produce better outcomes than the realistic alternatives?

What the Perfectionist Does Wrong

Henry Ford observed that any task can be solved when broken into sub-tasks. The perfectionist takes one sub-task out of ten, and refuses to proceed to the others until that single sub-task has been resolved to an impossible standard of accuracy.

The result is that the work that should have been submitted — functional, valuable, perhaps slightly imprecise — sits waiting while the part it was waiting on is polished indefinitely. The numerical answer might have been acceptable at two decimal places. The perfectionist is working on the fifteenth.

This is the same error in reverse: treating a problem that requires approximate-but-good-enough solutions as though it requires exactness, and producing nothing as a result.

How to Evaluate a Source

The only reliable marker of a trustworthy information source is a repeatedly occurring positive result in the population of people who followed it.

Not isolated cases. Not testimonials. Not the confidence of the person presenting. Not their credentials, because credentials can be fabricated and confidence can be performed.

If a doctor's patients consistently recover better than average outcomes would predict — measured across many cases, not one friend's anecdote — that is evidence the doctor is reliable. If a training methodology produces consistent, measurable results across a broad population, that is evidence the methodology works. If a blogger's audience can demonstrate change proportionate to following the advice, that is evidence.

The inability to verify this at scale — the "my friend used that pill and it worked" argument — is statistically meaningless. With small samples, seemingly impossible coincidences are probable. You need meaningful sample sizes before patterns emerge.

Your Own Mistakes

The same framework applies to your own errors.

You made a mistake. This means you solved the problem at the wrong decimal place on that iteration. Adjust the approach, run the iteration again, and continue. "Getting it wrong once means everything has failed" is the cognitive distortion of a person who believes perfect solutions exist.

They don't. They never did. What exists is the feedback loop: do something, observe the error, correct it, iterate. This is the only mechanism by which precision improves.

---