Advanced math problem (number/percentage extrapolation)

[Counter-Strike Player]

Extrapolate a set of "voting" numbers that fit the following series of percentages:

45.13%

33.83%

12.57%

5.09%

1.82%

1.54%

 

Wolfram has no idea. I know there's a way to assign numbers based on weighted averages but my math classes have passed some time ago...

[Counter-Strike Player]

edit: the estimated max number in the set (that correlates to 45.13%) should probably not exceed 100,000 if that makes it easier. Although it could be higher if the data fits better.

[Rolling Potatoe]

45.13% 4513

33.83% 3383

12.57% 1257

5.09%   509

1.82%   182

1.54%     54

so yeah, that adds up to 10000, wich is under 30000

[Counter-Strike Player]

 

45.13% 4513

33.83% 3383

12.57% 1257

5.09%   509

1.82%   182

1.54%     54

so yeah, that adds up to 10000, wich is under 30000

 

 

I meant a single number, not the total. And actually I could be wrong, max number could be up to 100,000 (or higher if the data fits better.)

[Rolling Potatoe]

I meant a single number, not the total. And actually I could be wrong, max number could be up to 100,000.

yeah, but that doesnt really change anything. I dont really get what the point of this ``problem´´ is though (If the answer shouldnt look so lazy, multiply everything by 7 or something).

[Counter-Strike Player]

yeah, but that doesnt really change anything. I dont really get what the point of this ``problem´´ is though (If the answer shouldnt look so lazy, multiply everything by 7 or something).

 

I'm trying to find a set of actual number of votes that correlate to the listed percentages with a good confidence fit. I'm thinking the only way to solve it is brute forcing every combination based off say, a total of around 200K votes.

[Rolling Potatoe]

I'm trying to find a set of actual number of votes that correlate to the listed percentages with a good confidence fit. I'm thinking the only way to solve it is brute forcing every combination based off say, a total of around 200K votes.

Yeah, the numbers I posted fit 100% to the percentages, I still dont see what youre trying to do to make it any more complicated.

The problem you posted is kinda pointless, if you are looking for the minimal number of total voters that would be an actual problem albeit an easy one.

[Counter-Strike Player]

Yeah, the numbers I posted fit 100% to the percentages, I still dont see what youre trying to do to make it any more complicated.

The problem you posted is kinda pointless, if you are looking for the minimal number of total voters that would be an actual problem albeit an easy one.

 

You know I'm realizing this is not "solvable" considering there's too little information and I'm trying to out-think myself into an answer where many exist. I don't know the total number of votes and was hoping that an extrapolation method could be possible to give me an "exact" number when pretty much any number is possible when you start with any estimate number of votes. The RES addon for Reddit can calculate the total number of downvote/upvote votes because there's other variables it can work with, like "percentage down-voted" "percentage up-voted" and "net upvoted."

Oh well, thanks for trying.

[Rolling Potatoe]

You know I'm realizing this is probably not solvable considering there's too little information and I'm trying to out-think myself into an answer where many exist. I don't know the total number of votes and was hoping that an extrapolation method could be possible to give me an "exact" number when pretty much any number is possible when you start with any estimate number of votes. The RES addon for Reddit can calculate the total number of votes because there's other variables it can work with, like "percentage down-voted" and also "percentage up-voted."

Oh well, thanks for trying.

 

Yep, you cant possibly get the total amount of votes from percentages.

You could calculate the minimal number of votes assuming those are accurate number (which is easy)

or assuming those numbers are rounded (youd need some programm for that of course....or alot of patience)

or assuming those are rounded numbers