[Retric]

Smaller tasks have less accurate estimates. If you say something takes 1h and it actually takes two days that's ~16x as long or a week long task taking 4 months. All it takes as a small corner case or odd bug that you did not consider and suddenly your estimates are worthless.

The only thing that makes small estimates seem better is the worst case tends to be less time, but overall they don't help.

[shagie]

An estimate is not a single number. Rather, an estimate says “it may take 1h if everything goes well, it may take 16h if things go bad.” This is then followed up with “at 1h, I’ll update with either a done or a revised estimate.”

Giving single numbers for an estimate with uncertainty is bad for all parties.

Additionally, with experience one can get better at giving estimates with less uncertainty for well known tasks and recognize the uncertainty for other tasks.

[Retric]

That's not relevant. Low / high estimates can be written in this form: X hours +/- Y%. 1-3 days seems wide but it's just 2 +/- 50%. 1 hour or 2 days ~= 1 day 1 hour +/- 90%.

AKA Y is much higher when you deal with low hour estimates.

This is most noticeable when you have lot's of tasks. If you think 200 tiny tasks may take 1 month or just over 1 year that's an almost useless estimate.