22

We've been told the flag weight formula above 500 has an asymptote of 750. What exactly is the formula being used though?

Improve this question
2

2 Answers 2

Reset to default
17

I'm not sure the exact formula is particularly helpful, but the usual increase (10, 5, whatever) is multiplied by the exponent (of 10) of some negative number, calculated as the difference between your current weight and 500, along with some scaling factor. Such that at 500 you get full points (10^0 == 1), with that dropping fairly swiftly after that.

There is nothing particularly scientific about that, other than it gives the shape we were looking for. I can recall that it would take 1072 correct flags with none incorrect to get to something like 749.5. But getting 700+ is sooner (you flagging-crazy fiends know who you are!).

"How much to reach x" is probably unanswerable except with "from y, assuming no flags get bounced". Re % (comments); that isn't something we can directly report at the moment.

Since computers do lots of rounding, for all I know it might be possible to hit 750.00; but there is a hard max - even if the maths goes wobbly through rounding error, you cannot exceed 750.

Improve this answer
6
  • Oh, "how much to reach x" also depends on which types of flags
    – Marc Gravell
    Mar 17, 2011 at 9:32
  • @Richard - that would probably not be well received. I recommend against it.
    – Marc Gravell
    Mar 17, 2011 at 10:21
  • @Richard - I mean: well-received by maths.SE - that isn't the type of question they crave
    – Marc Gravell
    Mar 17, 2011 at 10:27
  • 1
    you flagging-crazy fiends know who you are! - Are you suggesting that level of flagging is frowned upon? Yes, my flag weight is >700.
    – moinudin
    Mar 17, 2011 at 10:29
  • 6
    @marcog the exact opposite; they are good and correct flags, and do nothing but improve the site. Bring it on!
    – Marc Gravell
    Mar 17, 2011 at 10:41
  • @Marc Good to know :)
    – moinudin
    Mar 17, 2011 at 11:22
3

It is not super-linear, it is asymptotic

@Mehrdad - asymptotic, but not logarithmic – Marc Gravell♦ Mar 10 at 7:27

Under this answer https://meta.stackexchange.com/questions/80170/what-is-flag-weight/80178#80178

The formula is.. well, to be a true asymptote, it must use something like TAN() or an inverted (1/x) series. Plug your own numbers and you can't go far from the truth.

You will never reach 750, for that is the definition of Asymptote

EDIT

Except.. when it is not and asymptote. See here: https://meta.stackexchange.com/questions/84300/flag-weight-750-reached/88856#88856 It appears flag weight ALWAYS goes up by at least 0.1, right up to 750.

Improve this answer
3
  • 1
    Got the wrong terminology, removed super-linear from the question (still the same question). I want to know how many valid flags I need to reach a flag weight of X, as well as have a sense of what percentage of my flags are being marked valid/invalid.
    – moinudin
    Mar 17, 2011 at 8:20
  • 5
    Computers do rounding; on a computer you probably can reach it... eventually.
    – Marc Gravell
    Mar 17, 2011 at 9:15
  • @Marc: Sorry to keep nagging, but this has nothing to do with rounding.
    – Hendrik Vogt
    Aug 22, 2011 at 14:42

Not the answer you're looking for? Browse other questions tagged .