This question resolves to the median response (excluding non-numeric responses)
4
840Ṁ58resolved Jun 16
100%1.3%
74.1892389
43%Other
3%
0
3%
100
3%
50
3%
-999999999
3%
999999999
3%
0.0001
6%
69
3%
420
4%
58008
1.1%
70
1.1%
68
1.1%
101
1.1%
99
1.3%
69
1.2%
69
1.2%
98
1.2%
87
1.2%
73
All responses that are not numeric will be ignored. A "numeric" response is one matched by the regex `-?\d+.?\d+`. 1.05 is numeric, -4.00 is numeric, 5 is numeric, "the square root of two" is not numeric.
If multiple equivalent responses are the median (e.g. "2" and "2.0" and "2.00"), I will resolve to the one containing the fewest characters (e.g. "2").
Jun 14, 11:23am: Messed up the regex, correct regex is `-?\d*.?\d*` (and the empty string isn't numeric). At least one digit, optional minus sign and decimal point, no other characters. `.5` is numeric even without leading zero, `5` is numeric.
Jun 14, 11:24am: If there are an even number of responses (e.g. 1, 2, 3, 4), I will resolve to 50% the upper median and 50% the lower median (50% 2, 50% 3). Bet amounts are ignored, all responses are counted equally even if they are at 0%.
Jun 14, 11:27am: If there are an even number of responses and the medians are equivalent (e.g. 1, 2.0, 2.00, 3), I will resolve to 100% the one containing the fewest characters (e.g. "2").
If there is a tie in character length (e.g. "0.5" and ".50"), the winner is the ASCIIbetically first one (".50").
This question is managed and resolved by Manifold.
Get 
1,000 to start trading!
1,000Sort by:
I even wrote a script lol
regex=new RegExp(/^-?\d*\.?\d*$/); responses=Array.from(document.querySelectorAll('div.pb-2 > div:nth-child(1) > div:nth-child(1) > div:nth-child(2) > div:nth-child(2) > div:nth-child(2) > span:nth-child(1) > span:nth-child(1)'), t => t.textContent).filter(t => regex.test(t)).map(t => parseFloat(t)).sort((a,b)=>a-b); responses.slice(Math.floor(responses.length/2)-1,Math.ceil(responses.length/2)+1)
@AdrianKelly to be clear i'm not actually going to check things against the regex, if i don't notice that something is invalid then it's going to be treated as valid.
@jfjurchen ok, thanks for the clarification! Unfortunately the API can't read the responses in free response questions yet
@AdrianKelly Based on what I said it's 69 #1. It would make more sense as 69 #3 but I don't want to change the rules too much mid-market.
@jfjurchen I'm aware, I was just hoping that other people wouldn't spot that (and thus would miscalculate where the median is). :D
@AdrianKelly guess I'll sort by that then, final tiebreaker is the one whose serialized number comes first.