Hacker Read top | best | new | newcomments | leaders | about | bookmarklet login

Yeah, that was what I was assuming as well. I would ask the range, they'd give one, I'd flinch at the top-end of it and inevitably they'd come up 5%ish percent right away. From there I was generally able to wind up 10-20% above top of the original band.


sort by: page size:

anybody have anecdotes about their range vs. expected?

I'd take this even further. Sometimes it's 50/50. Sometimes 60/40, Sometimes 100/0. You just have to comfortable with that's how it is.

Good, this was my takeaway as well. My range was loosely rounded since 716k was something like the 85th percentile.

Lol actually I set my range at 30-50. It’s been interesting !

I don't recall, but I think I just settled on 50 or 60, because I didn't want to skew results too incorrectly.

If you know how the right amount feels, it really does not matter one bit how far off your initial measurement is. You start with an amount too low, and you add more until it feels right. If the amount you start with is 24% too low vs 20% too low bears no significance. Obviously you shouldn't start with an amount so far off as to be too high, but again that would never happen if you are going by feel.

I believe you're off by at least one digit in your range.

doesn't work should be range(1, 101).

I got called on that in an interview once.


I honestly couldn't tell. I'm guessing if I tried again I'd get anything between 4/10 and 7/10.

That's like 5% or so, for the first one. Could easily be significant.

It's also probably worth mentioning that the range of those values is likely less than 10, in that the answer is never going to be greater than 9 or less than 2. That should make any movement have more significance.


It's good to know what range people have tested with, even if it's not a hard cap.

Good question! I used 0.5 out of habit, but I do need to do some more experimenting with this parameter. But yes, intuitively it should probably be low. I'll do some experiments in the morning and see if it works well at 0.

That's excellent news, and thanks for responding! Can you share how you determine this threshold?

Interesting...isn't boardband penetration somewhere around the 35% mark there?

It’s been a long long time. I may be remembering the ratio wrong, or we might have been clipping the range a bit.

When the range is 130-190, the people currently making 130 will take note where the other end is. I think that's where the worry is.

The results around 22 aren't too surprising: most people will probably quickly figure out that the target cannot possibly be above 66.67, but few will realize that this implies that the target will be below 44.44. If you assume the guesses will be randomly distributed from 0 - 66, you get a target of 22, which isn't a bad guess for most of the times this experiment was performed on the proles.

I got it down to 43%. I had my eyes closed most of the time, only opened them twice; saw a 57% and a 43%.

That is an interesting chart. Apparently the breakpoint between medium and high is 70, and some just barely squeaked in under that in the upper 60s. That doesn’t inspire a lot is confidence.
next

Legal | privacy