|
On December 13 2011 06:23 chaynesore wrote:Show nested quote +On December 13 2011 06:15 TheDougler wrote:On December 13 2011 06:06 caradoc wrote:On December 13 2011 06:03 chaynesore wrote:On December 13 2011 06:00 Ariovist wrote:
we need more graphs... Ask and ye shall receive.. ![[image loading]](http://i.imgur.com/MjKSy.png) I expanded the x axis to show how much time is left. I find it curious how both votes rates of increase stay in proportion with each other... EDIT: What I want to see now that I'm tracking, is another giant surge from us, to see if it surges for Jennifer at the same time.. If we can do that. We have solid proof guys. We need day9 or something. Lol---- absolutely ridiculous, this is so obviously a bot  Can someone extrapolate the first graph to project the numbers at 5:00 please? (Probably the most bizzare thing I've ever asked, but figured I'd put it out there.) ![[image loading]](http://i.imgur.com/w8U86.png) And as somebody else stated, we do have to take these graphs with a grain of salt. Nevertheless, I'll be recording this information anyway and keeping the information as raw as possible. If we're worried about repeatability, we could have a few other guys keep track as well. Back to the graph though.. note it's still maintaining proportions. Also kind of funny how close they come together by the end. We should be worried I think....
Of course the extrapolation assumes the bot won't just surge ahead in the last 30 minutes, which I predict it will.
|
On December 13 2011 06:37 EnderCraft wrote:
She has only took 1% off of Glider in the past hour or so. I don't think she is going to close at the rate she's going. I predict a win for Glider, Art, Mankind, The universe and ESPORTS!
The margin is 3% I believe, and we have 3 hours left. So we would be pretty screwed if we're losing at 1% per hour.
|
On December 13 2011 06:38 urashimakt wrote:
I just got a text from my mum (a retired primary school teacher) asking if I could vote for "the teacher on youtube" if I have the time. I think I might know where some of these votes are coming from. And no, my mom isn't a bot.
I hope.
Ah, so some of these votes have to be coming from somewhere....
|
On December 13 2011 06:38 urashimakt wrote:
I just got a text from my mum (a retired primary school teacher) asking if I could vote for "the teacher on youtube" if I have the time. I think I might know where some of these votes are coming from. And no, my mom isn't a bot.
I hope.
this doesn't explain the above graph where her votes mirrors ours.
|
***CAR 2 SPOILER: The good guy will be the bad guy at the end. Glider is the one who is cheating for JenniferESL just for the sake of pissing off sc2 nerds and create a shocking but stupid plot twist to this epic story
|
edit: Sorry, since the system actually allows multiple voting, I assumed it's alright if we work on our own bot.
User was warned for this post
|
On December 13 2011 06:41 brachester wrote:
***CAR 2 SPOILER: The good guy will be the bad guy at the end. Glider is the one who is cheating for JenniferESL just for the sake of pissing off sc2 nerds and create a shocking but stupid plot twist to this epic story
LOL Glider wrote the bot voting for JESL, and is using this to promote his channel in a massive, massive ESPORTS movement.
YOU DEVIOUS MAN, YOU!
Ps the portrait you drew me is still grade A awesome <3
You're a winner in my heart.
(d'awwww)
|
On December 13 2011 06:41 Glider wrote:Show nested quote +On December 13 2011 06:38 urashimakt wrote:
I just got a text from my mum (a retired primary school teacher) asking if I could vote for "the teacher on youtube" if I have the time. I think I might know where some of these votes are coming from. And no, my mom isn't a bot.
I hope. this doesn't explain the above graph where her votes mirrors ours.
I'm not gonna say there isn't a bot or cheating going on, because I don't know. But I do know from what basic understanding of statistics I have that graphs are graphs, not diabolical evidence of misdeeds. There is no reason at all to think that those two lines = bot proof unless that's what you're looking for when you take a peek at them.
|
almost 13 k ahead. Lead has dwindled. idk what to do
of course I'm voting. But I want to do more. I feel so lonely.
|
On December 13 2011 06:43 figq wrote:
How to make a bot with Firefox (votes once at about every 10 seconds):
Also as much as I love Glider and want him to win, I don't feel like we should absolutely resort to cheating the system to win ^^;;
|
On December 13 2011 06:43 Torenhire wrote:Show nested quote +On December 13 2011 06:41 brachester wrote:
***CAR 2 SPOILER: The good guy will be the bad guy at the end. Glider is the one who is cheating for JenniferESL just for the sake of pissing off sc2 nerds and create a shocking but stupid plot twist to this epic story LOL Glider wrote the bot voting for JESL, and is using this to promote his channel in a massive, massive ESPORTS movement. YOU DEVIOUS MAN, YOU! Ps the portrait you drew me is still grade A awesome <3 You're a winner in my heart. (d'awwww)
Such a baller plan haha.
|
@ Figq remove that. We don't want to ascosiated with that kind of trash -.-
|
I dont understand something about the graph though, considering it is a bot, how can it know the rate at which votes for Glider are coming? I mean, for sure it can do that after refreshing the page and seeing how many votes he got from the last 'wave' but then it would be delayed in time, and this graph shows that it changes almost exactly as our graph at the same time, is it really possible without having an access to the actual site(poll?).
|
On December 13 2011 06:43 figq wrote:
How to make a bot with Firefox (votes once at about every 10 seconds):
Use another browser for Internet while this is going on. Don't switch to full screen SC2 though; that stops the macro.
And this is why one side can be seen as illegitimate at the other.
|
|
|
Even if we were going to bot the ESL bot seems to be working to match our gains and slowly overtake us. If we were botting we would still lose, the numbers would just be much, much higher
|
On December 13 2011 06:45 Benjef wrote:
@ Figq remove that. We don't want to ascosiated with that kind of trash -.- It's legit to vote multiple times, the poll allows it. There are many Internet polls that promote multiple voting.
edit: I got my first warning on TL for that though. So I guess you are right. *shrug*
|
On December 13 2011 06:33 Cascade wrote:Show nested quote +On December 13 2011 06:17 caradoc wrote:On December 13 2011 06:14 Cascade wrote:On December 13 2011 06:09 chaynesore wrote:On December 13 2011 06:07 pPingu wrote:On December 13 2011 06:03 chaynesore wrote:On December 13 2011 06:00 Ariovist wrote:
we need more graphs... Ask and ye shall receive.. + Show Spoiler +I expanded the x axis to show how much time is left. I find it curious how both votes rates of increase stay in proportion with each other... EDIT: What I want to see now that I'm tracking, is another giant surge from us, to see if it surges for Jennifer at the same time.. If we can do that. We have solid proof guys. We need day9 or something. Is it possible for you to provide a 3 days graph or something? It would be really interesting It would be, but no. I'm just doing this manually by refreshing every 5 mins while watching movies.. Maybe youtube has those stats... I'll keep doing this until the comp is over though Yes, and because of this method, we should be a bit careful before we use this as evidence for bot. Imagine if your reading is a bit early or late, due to caching, you being caught up in movie, youtube page not updating votes perfectly live or something. Then you will see correlated shapes like this. If you accidentally get a too short time interval, both increments will be low, if you accidentally take a too large time, both will be high. just saying, don't be too hot on the trigger from this graph, as the information goes through a lot of steps, and any flaw will give this signal. Not technically true--- notice how the lines never cross-- you would predict that they would very likely cross if there was no botting. You would also predict the existence of cases where portrait art inc/5min would increase while esl inc/5min would decrease, again, doesn't happen. What you would NOT predict if this was two organic processes is a tight correlation at 5 minute intervals, which you see here. The spiky nature of the graph is (partially) related to intervals, but that's all. umm, what are you talking about? >_> Why would crossing always happen without bots, and never happen with bot? With random (non-bot) influx of votes, the statistical errors would be small at this scale, and large time-errors would dominate and give this kind of behavior as you see. But this kind of arguments are wasted on TL I'm starting to realise, so nm me. keep voting guys!!
It's not that crossing would always happen without a bot, just that it would be statistically speaking, extremely likely since the total number of votes is on the same order of magnitude, and our sampling granularity is fairly fine. Since we are assuming an organic process it implies a more or less random distribution of votes at any particular point in time.
I know what you're trying to say though, and you have a point-- but think about it this way-- the number of vote increase per interval varies by as much as 50% of the average. In order for large time-errors to dominate, we would expect that the interval would need to be off by a correspondingly large amount to account completely for the similarity.
I don't like your self-superior attitude either, its not necessary since we're both trying to figure the same thing out. You don't need to be 100% definitevely correct all the time, its not that important, and you don't need to worry so much about saving face-- its a collaborative discussion. (except for maybe now that I'm calling you on it, which I wouldn't be if you hadn't made the snide 'my thoughts are a waste on the rabble of TL comment' in the first place, but whatever.)
|
Some serious weird shit is going on with Jennifer's vote totals. There's no community supporting her.
|
On December 13 2011 06:45 Nerchio wrote:
I dont understand something about the graph though, considering it is a bot, how can it know the rate at which votes for Glider are coming? I mean, for sure it can do that after refreshing the page and seeing how many votes he got from the last 'wave' but then it would be delayed in time, and this graph shows that it changes almost exactly as our graph at the same time, is it really possible without having an access to the actual site(poll?).
There is the total amount of votes data always available, i think it will just slightly put in more votes than its increase/set time.
|
|
|
|
|
|
EPT
