The longer it takes to show a difference the greater the chance (statistically speaking) that the difference is due to chance rather than a drug treatment effect, and that is not helpful
Care to rephrase that?
Not really, but here is a different explanation. In Trial A the KM curves track each other closely for three months then separate with the drug curve showing an advantage for survival for some period after that. Trial B looks exactly the same except that the curves do not separate until one year after treatment. While both sets of curves show a survival benefit, Trial A will have better statistics.
If a difference in outcomes is observed immediately following administration of a drug treatment, then it is likely, but not certain, that the observed change is due to the drug. The longer it takes for an observed outcome to emerge, the greater the chance the improvement is due to something other than the drug. That is why long tails are harder to interpret statistically and why a long tail effect is less helpful to shareholder chances for FDA approval than a near term benefit would be. Patients are grateful for any benefit.