Credit goes to Louis Parkinson in this video for this advice (paraphrased to my understanding):
When projecting, don't treat attempts as if they are worthless, because they aren't. They have a cost. They cost skin, energy, time, motivation, and potentially the patience of spotters/belayers, and all of these are in limited supply. So recognize the value of each attempt and spend them wisely: only make quality attempts. Toward that end, two rules:
- Exactly 5 attempts. Fewer than that, and you're maybe giving up too soon, but more than that, and you're achieving diminishing returns.
- Don't just jump back on after a failed attempt. Don't get back on until you can answer the questions a) What did I do which caused me to fail? And b) what will I do this time instead?
A few notes:
- "Attempts" is a bit ambiguous. If it's redpoint attempts on a long trad pitch, you probably have fewer than 5 quality attempts in a session. If it's working out beta for a short sequence and you have 6 beta ideas, it probably makes sense to try all 6 betas rather than rigidly stopping at 5. Use your brain, observe yourself, and be aware of the point where the return-on-investment for an attempt begins to decrease. I think the point of the number 5 is to bring to light that this number is significantly lower than is intuitive to most people.