Almost none of the common failures are "did not know the formula".
Preparing for the wrong process
The most avoidable one. A market-maker round, an aptitude screen, and a research interview select for different things. Some large firms run several entirely separate pipelines under one brand, and generic preparation serves none of them well. Find out which process you are in before you plan the month.
Guessing on a penalty-scored screen
Several assessments subtract for wrong answers. On those, a guess has negative expected value and skipping is correct - yet candidates guess anyway, because unanswered questions feel like failure. See the probability OA explained.
Know the scoring rule before you sit the test. If you do not know it, ask.
Optimising for speed before accuracy
On a timed arithmetic screen, people push their pace until the error rate climbs, then keep pushing. Accuracy first, then speed - a fast wrong answer scores worse than a slower right one on nearly every scored screen.
Going silent
Covered at length in thinking out loud. Silence is not neutral; it removes the evidence you are being marked on.
Defending a wrong answer
The single worst signal. If the interviewer pushes back and they are right, update. Trading is a job about changing your mind when the evidence changes, and an interview is a live test of whether you can.
Equally: do not abandon a correct answer because someone frowned. Both failures are the same failure - responding to social pressure instead of evidence.
Studying finance for a firm that does not test it
Some firms state publicly that they do not test finance or economics knowledge. Reading about market structure for those processes is time not spent on probability, which they do test.
Thin claims on your own CV
Research-heavy processes press hard on what you personally did on a project. A vague claim on someone else's work collapses in about ninety seconds, and it colours everything after it.
Neglecting sleep in the final week
Unglamorous and real. These processes filter on arithmetic accuracy under time pressure, which is exactly the faculty sleep deprivation degrades first.