Hi everyone, I had an idea a few minutes ago. While I was watching the new Obama endorsing Hillary video on YouTube I thought to myself, "These numbers seem odd". At least as pertains to the left side, the up votes on the video. Watching the two numbers, up votes and down votes, I thought to myself that these changes don't seem to look right. That they looked suspicious in way.
So what I did was get a clock and sat at my computer with the clock. I kept the video up so that I could refresh the page over a time total of 12 minutes, checking every 1 minute for the changes on both up votes and down votes. (I refreshed the page every 1 minute exactly).
After recording my data I went through and checked for the differentials (differences) between the changes (Δ) in the data. What I found was that the up votes were dropping into constants (1,1,1), (3,3,3) with as little as ΔΔ and ΔΔΔ. Whereas the down votes were staying variant even down to ΔΔΔ.
Why is this important? Well you see, constants are where a curve has flattened. If we are dealing with constants we are dealing with a straight line. This means the curve is being interrupted.
So in as little as ΔΔ you can see that the up votes on that video are becoming interlaced with straight lines, and consequently something behind the scenes is interrupting the curve.
This is not surprising when the video currently shows me that there are 6,386 comments and yet I am unable to load any.
Math can be fun. Cheers.
(check next image)
(image)
Aw shit it needs to be flipped, hold on. Lol.
Your a good guy
>>76709742
Can u link video this is happening for
>>76710482
https://www.youtube.com/watch?v=S9W0F2mz1jc
Here ya go! :)
>>76709742
That's cool what you did and everything but why bother? I figured the likes and dislikes were fixed long before I even saw this post.
I mean with crooked Hillary and obongo it's pretty much a given that they'll skew things like this
>>76710691
I'm up at 4 in the morning with not much going on. Thought I'd just do some math. :/
I should have included the video in the original post but I'm not super expert level at posting yet.
op you have schizophrenia
also despite being a pink board it is against the rules to post porn on /pol/
>>76710846
Nah, schizophrenia is quite different than basic calculus and some number theory. But hey, for some people math is difficult!
>>76710846
it's not schizophrenia, that's just math. The fact the standard curve is interrupted into a line shows that they're funneling upvotes into the video. lel, no one likes shillary online
>>76711110
Someone gets it! :d
>>76711059
>>76711110
>math
nigga just buy a calculator who gives a shit about all that mess, aint nobody got time for that
>>76711231
:DDD
>>76710135
Who is this semen demon?
>>76711231
>>76711201
>posting the only Singaporean anime girl
Why
>>76709742
Huh. Ok I sort of get the taking the deltas of deltas repeatedly, but what does it mean? What kind of tampering would this catch? Wouldn't they just be adding upvotes randomly if they didn't want to get caught? Can these techniques catch the "mixing" of different distributions somehow?
I guess it would be done by third parties using bots so as to not involve Google, so maybe they don't care about not being detected as much as they should...
This legitimately could be a coincidence. Math major here and the first thing they tell you in any decent analysis class is that 'runs' (consistencies in data) do not happen randomly.
Also if you graph all of the changes in it, the line is still fairly flat, it's below a 5% margin of error, also the changes in the stacked data don't even hit sigma 2, the chances of random data having runs on a sigma 2 delta is about 1/8, so pretty common, and not unreasonable. (for the record the physics fags require a sigma of 5 to make sure something is correct, the chances of something at sigma 5 being 'random' is something like 1/3.5e8).
Interesting experiment, but this happens more than we'd like to believe.
There were several anons running analytics on "With Open Gates" when it first came out and the likes and dislikes were being altered over consistent intervals with near-predictable outcomes.
>>76711768
Yep. It could be a coincidence. To really get a hold on whether or not there are inconsistencies I think I would need more data. But I didn't feel like sitting here recording data for 5 hours etc. Lol.
>>76711768
>>76711611
I thought that was his point, that the repeating deltas show a change in the upvote distribution but at the same types of intervals.
I do chem, not math fa.malam
https://www.youtube.com/watch?v=PFxFRqNmXKg&feature=youtu.be
>>76712102
>the repeating deltas show a change in the upvote distribution but at the same types of intervals.
Yep.
Does anyone have a link to the smear ads Obama ran against Hilary back in the day?
