To anyone who wasn't just following that ridiculously epic thread about voting for Glider, you missed an awesome community experience - but to those who were there, I was the dude posting the graphs. My original reason for keeping this data was to analyse WTF went wrong after the fact because things were looking qutie grim a few hours ago, but we won!! As long as YouTube don't pull a fast one on Glider on the 19th when they announce, we shouldn't need it at all. Nevertheless, I'm making it available just in case, and for those who would like a closer look to maybe understand or try to reverse engineer what exactly was going on with those votes from the opposition.
DOWNLOAD LINK: [url blocked]
UP TO 4PM @5min Sampling
4PM - 5PM @1min Sampling Note the gap represents only a little bit of data removed because it presented as an outlier on the ratio plot and skewed the axis ridiculously. It's available in the spreadsheet though.
10mins @15s Sampling
I hope it's okay for me to make this thread, I just figured it might be better to be easily available rather than having to search through all the hundreds of "FUCK YEH!!! WE RULE" posts towards the end of the main one, lol. I'd also suggest though that all theories and discussion of botting / data analysis be kept to this thread? And if there is any other data people have collected (I saw a few people putting good stuff out there but I was too busy with the graph sampling to do much else), I'll edit it into the OP
In the face of cheating or not though, we won! Congratulations to Glider and the Sc2 community abroad. I only came into this scene with the launch of Sc2 so I'm relatively new here but it's true what everyone says - this is the nicest community there is. Absolutely unbelievable effort guys. What a great day :D
I voted! Awesome guy, he really deserves the feature; also, how big of a deal is getting featured for a day on youtube anyway--i mean, I had never even heard of this before I saw glider's blog.
Ok, so for those who weren't following the thread, take a look at the "Growth Ratio by 5 Minute Interval" graphs (4-5 shows it the best). This shows that the votes for JenniferESL seem to mirror the votes for ThePortraitArt fairly closely. On top of that, people refreshing the page noticed that Jennifer's votes seemed to move up in groups of 50 votes at a time, where Glider's seemed o be moving up fairly normally by a couple votes at a time. This lead most people to believe that someone was botting votes for Jennifer.
On December 13 2011 10:28 CrazyF1r3f0x wrote: I voted! Awesome guy, he really deserves the feature; also, how big of a deal is getting featured for a day on youtube anyway--i mean, I had never even heard of this before I saw glider's blog.
He gets featured for a month. Think about the millions of people that go to youtube everyday and i'm sure you'll realise why its such a big deal.
Edit: Actually getting mixed answers for how long he gets featured >.<
After about 40k votes for the Jennifer ESL thing, I was getting really suspicious. Her subscriber base isn't particularly large and she kept somehow managing to keep up with our community's votes. Anyways, good work everyone and grats to Glider!
On December 13 2011 10:28 CrazyF1r3f0x wrote: I voted! Awesome guy, he really deserves the feature; also, how big of a deal is getting featured for a day on youtube anyway--i mean, I had never even heard of this before I saw glider's blog.
He gets featured for a month. Think about the millions of people that go to youtube everyday and i'm sure you'll realise why its such a big deal.
Edit: Actually getting mixed answers for how long he gets featured >.<
So if I read this right, the similarity between the voting patterns suggests a bot or something similar?
That would make sense from a youtube point of view. What is the point in having a competition if one of the competitors absolutely blitzes the field. How humiliating for the other guys and sort of makes the competition a waste. I can see how youtube benefit from having a close result.
That last graph showing Jennifer ESL growth is definitely suspect of botting look at the growth compared to ThePortaitArts growth the JenniferESL's growth in the last 2 graphs is completely unnatural.
Look how it spikes up and down like that that's really weird.
On December 13 2011 14:40 Probulous wrote: So if I read this right, the similarity between the voting patterns suggests a bot or something similar?
That would make sense from a youtube point of view. What is the point in having a competition if one of the competitors absolutely blitzes the field. How humiliating for the other guys and sort of makes the competition a waste. I can see how youtube benefit from having a close result.
I don't see how Jenny could have botted though.
Very intelligent post. Seems like a plausible scenario to me. Certainly more likely than the Jennifer ESL botting herself, and also not winning.
On December 13 2011 14:40 Probulous wrote: So if I read this right, the similarity between the voting patterns suggests a bot or something similar?
That would make sense from a youtube point of view. What is the point in having a competition if one of the competitors absolutely blitzes the field. How humiliating for the other guys and sort of makes the competition a waste. I can see how youtube benefit from having a close result.
I don't see how Jenny could have botted though.
Very intelligent post. Seems like a plausible scenario to me. Certainly more likely than the Jennifer ESL botting herself, and also not winning.
The problem is that if this is true then it is also fraudulent. They are deliberately misrepresenting the results to gain more clicks on their site. Not only does this increase the eposure of the tournament but also advertising revenue. That could be construed as fraud.
On December 13 2011 14:40 Probulous wrote: So if I read this right, the similarity between the voting patterns suggests a bot or something similar?
That would make sense from a youtube point of view. What is the point in having a competition if one of the competitors absolutely blitzes the field. How humiliating for the other guys and sort of makes the competition a waste. I can see how youtube benefit from having a close result.
I don't see how Jenny could have botted though.
I was also thinking the same thing while analyzing the similarity of the voting influxes; its very probable youtube actually rigged the competition to make it a closer race.
Taking a quick glance of the graphs shows the pattern to be very odd to say the least.
I really don't see how the voting patterns are that odd? All it really says is that both candidates were approximately equally popular and gaining votes close to each other, which isn't all that suprising if both were similar.
Of course, if you really wanted to prove something was up you could try a retrospective correlative regression analysis on the curves and see if the similarity was significant. You would not expect the voting patterns to be similar in such a way as soon as one increases so does the other. The fact that they never diverged by more than 1% is really strange. Could be coincidence but it certainly looks fishy.
Okay.. Maybe I'm just awful with math, but where in those graphs is the evidence that something fishy was going on? All I see is a bunch of fluctuating lines.
Just to edit in, checking their youtube channels: JenniferESL Channel Views: 3,372,397 Total Upload Views: 24,443,381 Subscribers: 70,068
ThePortraitArt Channel Views: 1,061,695 Total Upload Views: 10,338,029 Subscribers: 66,142
Similar number of subscribers, but many more channel views and total upload views for Jennifer. Granted she apparently doesn't have a community backing her like TL / Reddit for Glider, but some of the people saying that there's no way she could get that many votes just seem to not understand the fact that there's like a billion people in the world learning English and that her channel might have broader appeal than Glider's art.
Of course, I think Glider is way more awesome, voted for him, and am very happy he won :-p I just get tired of reading all this conspiracy theory type bullshit when there doesn't seem to be any real evidence. Again, if there's something in the graphs, please explain it to me, because I don't really see it. The points where her incoming votes drop to 0 for brief periods seems a little weird, but that's about it.
Did you gather that data yourself? If no, is there any chance to get the data for the past days? (considering you said it was looking grim at first and those graphs show nothing of the sort)
note the featured video (JenniferESL).. and that all the other competitors videos were, at best buried under navigation. The "featured playlist' isn't even this months playlist (which DOES exist.)
On December 13 2011 15:21 dormer wrote: Okay.. Maybe I'm just awful with math, but where in those graphs is the evidence that something fishy was going on? All I see is a bunch of fluctuating lines.
Just to edit in, checking their youtube channels: JenniferESL Channel Views: 3,372,397 Total Upload Views: 24,443,381 Subscribers: 70,068
ThePortraitArt Channel Views: 1,061,695 Total Upload Views: 10,338,029 Subscribers: 66,142
Similar number of subscribers, but many more channel views and total upload views for Jennifer. Granted she apparently doesn't have a community backing her like TL / Reddit for Glider, but some of the people saying that there's no way she could get that many votes just seem to not understand the fact that there's like a billion people in the world learning English and that her channel might have broader appeal than Glider's art.
lastly look at this graph: http://i.imgur.com/zRNUe.png (read up on how certain things are programmed, and you can see her channel mimics my votes EXACTLY + w/e real votes she got on her own)
---------- Look at channel view gained/subscriber growth since 12/5 (the day voting began) ----------
On December 13 2011 15:21 dormer wrote: Okay.. Maybe I'm just awful with math, but where in those graphs is the evidence that something fishy was going on? All I see is a bunch of fluctuating lines.
Just to edit in, checking their youtube channels: JenniferESL Channel Views: 3,372,397 Total Upload Views: 24,443,381 Subscribers: 70,068
ThePortraitArt Channel Views: 1,061,695 Total Upload Views: 10,338,029 Subscribers: 66,142
Similar number of subscribers, but many more channel views and total upload views for Jennifer. Granted she apparently doesn't have a community backing her like TL / Reddit for Glider, but some of the people saying that there's no way she could get that many votes just seem to not understand the fact that there's like a billion people in the world learning English and that her channel might have broader appeal than Glider's art.
Of course, I think Glider is way more awesome, voted for him, and am very happy he won :-p I just get tired of reading all this conspiracy theory type bullshit when there doesn't seem to be any real evidence. Again, if there's something in the graphs, please explain it to me, because I don't really see it. The points where her incoming votes drop to 0 for brief periods seems a little weird, but that's about it.
Thank you! I also hate conspiracy theorists and don't get me wrong the way that the SC2 community has rallied behind ThePortraitArt is very supportive but as a scholar of linguistics I cannot help but side with JenniferESL because in my mind learning an international language is much more important than the spread of a sport (electronic or otherwise). Reddit hates me for that
On December 13 2011 15:54 rewired wrote: I voted like 150 times I think one of the real heroes here was Seeker tho, I don't think he logged off all weekend.
Cheating to win a competition? Classy.
I don't understand why people were so set on this guy winning. Sure the art is amazing but it's only mildly associated to the starcraft community is it not? Then again I may not know what I'm talking about. Anyway it just seems oddly childish that our community wants to win everything even if there are more deserving candidates out there.
On December 13 2011 15:54 rewired wrote: I voted like 150 times I think one of the real heroes here was Seeker tho, I don't think he logged off all weekend.
Cheating to win a competition? Classy.
I don't understand why people were so set on this guy winning. Sure the art is amazing but it's only mildly associated to the starcraft community is it not? Then again I may not know what I'm talking about. Anyway it just seems oddly childish that our community wants to win everything even if there are more deserving candidates out there.
And who are you to judge that (truly amazing)art is not as deserving as learning english? Just let people be.
Those graphs..they are too much of a coincidence. Something was wrong in this competition.
On December 13 2011 15:21 dormer wrote: Okay.. Maybe I'm just awful with math, but where in those graphs is the evidence that something fishy was going on? All I see is a bunch of fluctuating lines.
Just to edit in, checking their youtube channels: JenniferESL Channel Views: 3,372,397 Total Upload Views: 24,443,381 Subscribers: 70,068
ThePortraitArt Channel Views: 1,061,695 Total Upload Views: 10,338,029 Subscribers: 66,142
Similar number of subscribers, but many more channel views and total upload views for Jennifer. Granted she apparently doesn't have a community backing her like TL / Reddit for Glider, but some of the people saying that there's no way she could get that many votes just seem to not understand the fact that there's like a billion people in the world learning English and that her channel might have broader appeal than Glider's art.
On December 13 2011 15:54 rewired wrote: I voted like 150 times I think one of the real heroes here was Seeker tho, I don't think he logged off all weekend.
Cheating to win a competition? Classy.
I don't understand why people were so set on this guy winning. Sure the art is amazing but it's only mildly associated to the starcraft community is it not? Then again I may not know what I'm talking about. Anyway it just seems oddly childish that our community wants to win everything even if there are more deserving candidates out there.
And who are you to judge that (truly amazing)art is not as deserving as learning english? Just let people be.
Those graphs..they are too much of a coincidence. Something was wrong in this competition.
I have a degree in linguistics and I live in a third world country with 11 official languages where basic services are a problem for more than 5 million people. As much as I love SC2 learning English is slightly more important here than esports.
On December 13 2011 15:54 rewired wrote: I voted like 150 times I think one of the real heroes here was Seeker tho, I don't think he logged off all weekend.
Cheating to win a competition? Classy.
I don't understand why people were so set on this guy winning. Sure the art is amazing but it's only mildly associated to the starcraft community is it not? Then again I may not know what I'm talking about. Anyway it just seems oddly childish that our community wants to win everything even if there are more deserving candidates out there.
And who are you to judge that (truly amazing)art is not as deserving as learning english? Just let people be.
Those graphs..they are too much of a coincidence. Something was wrong in this competition.
I have a degree in linguistics and I live in a third world country with 11 official languages where basic services are a problem for more than 5 million people. As much as I love SC2 learning English is slightly more important here than esports.
So people would start learning english by randomly finding a youtube channel about it? Youtube is in english or is it localized to all 11 official languages? I doubt people who find english tutorials randomly will stick with them or gain any long term benefit.
Were you also against the rumble in the bronze charity target? Because african children could have used those $10k more?
I think one thing that could help for people who don't understand graphs or numbers is just writing "This graph looks normal" or "This looks weird" [in my eyes the first growth graph is insane, that she NEVER was growing at a slower rate than Glider, actually didn't make sense]
On December 13 2011 15:54 rewired wrote: I voted like 150 times I think one of the real heroes here was Seeker tho, I don't think he logged off all weekend.
Cheating to win a competition? Classy.
I don't understand why people were so set on this guy winning. Sure the art is amazing but it's only mildly associated to the starcraft community is it not? Then again I may not know what I'm talking about. Anyway it just seems oddly childish that our community wants to win everything even if there are more deserving candidates out there.
And who are you to judge that (truly amazing)art is not as deserving as learning english? Just let people be.
Those graphs..they are too much of a coincidence. Something was wrong in this competition.
I have a degree in linguistics and I live in a third world country with 11 official languages where basic services are a problem for more than 5 million people. As much as I love SC2 learning English is slightly more important here than esports.
So people would start learning english by randomly finding a youtube channel about it? Youtube is in english or is it localized to all 11 official languages? I doubt people who find english tutorials randomly will stick with them or gain any long term benefit.
Were you also against the rumble in the bronze charity target? Because african children could have used those $10k more?
they might not sub before because they dont have time to learn proper english, but apon seeing shes up for ontherise might decide shes most deserving, we dont know
this whole thread seems to be based around the fact that "she was never growing slower than glider". but its only about 4pm onwards, go back 2 days and glider was behind, so she must of grown slower.
On December 13 2011 15:54 rewired wrote: I voted like 150 times I think one of the real heroes here was Seeker tho, I don't think he logged off all weekend.
Cheating to win a competition? Classy.
I don't understand why people were so set on this guy winning. Sure the art is amazing but it's only mildly associated to the starcraft community is it not? Then again I may not know what I'm talking about. Anyway it just seems oddly childish that our community wants to win everything even if there are more deserving candidates out there.
And who are you to judge that (truly amazing)art is not as deserving as learning english? Just let people be.
Those graphs..they are too much of a coincidence. Something was wrong in this competition.
I have a degree in linguistics and I live in a third world country with 11 official languages where basic services are a problem for more than 5 million people. As much as I love SC2 learning English is slightly more important here than esports.
So people would start learning english by randomly finding a youtube channel about it? Youtube is in english or is it localized to all 11 official languages? I doubt people who find english tutorials randomly will stick with them or gain any long term benefit.
Were you also against the rumble in the bronze charity target? Because african children could have used those $10k more?
People desperate to learn English would do so in any way possible, especially if it's free. I know Youtube is localised to at least the 2 main langauges. That wasn't my point though, I was merely responding to Alpino's comment. If you look at JenniferESL's profile page you'll see, amidst the sea of awful comments from the gaming community, a dedicated group of individuals from around the world who are learning English through her tutelage.
No, of course I was not against Rumble in the Bronze.
Well, if this will make companies more interested in esports, then it was a good cause in my eyes. Yes, learning english might be more important overall, but you should not need youtube to do that, or, let's put it like that, if you do search youtube for something like that, you should be able to find it naturally.
On December 13 2011 15:31 JustPassingBy wrote: Did you gather that data yourself? If no, is there any chance to get the data for the past days? (considering you said it was looking grim at first and those graphs show nothing of the sort)
I did gather this data manually so no there is no way to backtrack the data unless Youtube/Google makes it available. I found the motivation to start tracking these numbers at around the same time TB made his video and created the surge. The timings will all match up in the original thread if you must check, but that is why there is sadly no recorded evidence of this.
On December 13 2011 15:56 zatic wrote: Can you please explain what each of these graphs is supposed to mean instead of just dumping them in a thread?
I collected data and generated graphs. I don't believe it's necessarily my role to skew the OP with the way in which I view the data, which is why I left this out. Additionally, I had no idea what I was collecting until I had collected it, and as I stated, it was more of a backup plan in case something went wrong, such that we might be able to view the trends in voting in the hours leading to the close. Having collected the data, although Glider had won, I decided to share it with the community anyway as many people appreciated the graphs generated throughout the live thread and there is a general consensus that something is awry in the system and some people have expressed interest in analysing this further.
Nevertheless, since you're asking... I find it curious that the growth between both poll options was consistently proportional with one another for a very long time. This struck myself and many others as something other than organic growth, and that perhaps one was a function of the other (i.e. bot). Additionally, the largely erratic spikes in the smaller samplerate graph I generated would perhaps indicate a crude self-correcting whatever? I dunno, I'm an engineer. I'll leave this to the dedicated math/stat types who were talking about regression analysis and all that.
On December 13 2011 18:09 karpo wrote: Shouldn't the op at least add what this is about or link to the original thread. As i missed this "event" the OP is like an enigma of graphs.
Sorry man, been afk/sleeping for a bit. I'm on a god damn mac at the moment and it's driving me insane so I'll touch the OP up a little bit later with some FAQ / basic arguments/analysis/opinions and stuff if people want later. Maybe add some polls. I dunno. I'm not too worried about this data now. Like I said it was more of a precaution and was going to be something good to fall back onto after the fact if things turned out unfavourably - just sharing for those interested regardless.
On December 13 2011 15:54 rewired wrote: I voted like 150 times I think one of the real heroes here was Seeker tho, I don't think he logged off all weekend.
Cheating to win a competition? Classy.
I don't understand why people were so set on this guy winning. Sure the art is amazing but it's only mildly associated to the starcraft community is it not? Then again I may not know what I'm talking about. Anyway it just seems oddly childish that our community wants to win everything even if there are more deserving candidates out there.
And who are you to judge that (truly amazing)art is not as deserving as learning english? Just let people be.
Those graphs..they are too much of a coincidence. Something was wrong in this competition.
I have a degree in linguistics and I live in a third world country with 11 official languages where basic services are a problem for more than 5 million people. As much as I love SC2 learning English is slightly more important here than esports.
So people would start learning english by randomly finding a youtube channel about it? Youtube is in english or is it localized to all 11 official languages? I doubt people who find english tutorials randomly will stick with them or gain any long term benefit.
Were you also against the rumble in the bronze charity target? Because african children could have used those $10k more?
Why so uptight jutranjo? His point, and I think it's clearly a reasonable one, is that it might be better for something that can change people's lives gets more exposure, as opposed to just another thing like many others on youtube that people can look out and go "oh, cool".
Not that I didn't vote for glider. Congrats!
edit: I don't understand how anyone can care about the innards of a popularity contest on the internet, but whatever.
The most plausible explanation to me, if you must assume there was some kind of illicit activity, is that the numbers were being boosted for both in such a way that eventually they both received the same number of votes, but granted at varying rates. Thus, the legitimacy of the voting is preserved, but by adding in a dramatic baseline orders of magnitude more than the true voting rate, it creates the appearance of a hot race.
Was I reading those graphs correctly in that it said, at times, the contestants were each receiving upwards of 700 votes in 5 minutes? That would blow through the TL viewership of glider threads in less than an hour. Could it really be true that random people check out this youtube contest and then make a vote, to supply such a high voting rate?
On December 13 2011 15:21 dormer wrote: Okay.. Maybe I'm just awful with math, but where in those graphs is the evidence that something fishy was going on? All I see is a bunch of fluctuating lines.
Just to edit in, checking their youtube channels: JenniferESL Channel Views: 3,372,397 Total Upload Views: 24,443,381 Subscribers: 70,068
ThePortraitArt Channel Views: 1,061,695 Total Upload Views: 10,338,029 Subscribers: 66,142
Similar number of subscribers, but many more channel views and total upload views for Jennifer. Granted she apparently doesn't have a community backing her like TL / Reddit for Glider, but some of the people saying that there's no way she could get that many votes just seem to not understand the fact that there's like a billion people in the world learning English and that her channel might have broader appeal than Glider's art.
---------- Look at channel view gained/subscriber growth since 12/5 (the day voting began) ----------
Again, what's the incriminating part? She's gaining views and subscribers, but not a ton more compared to her previous growth.
On December 13 2011 17:57 Zlasher wrote: I think one thing that could help for people who don't understand graphs or numbers is just writing "This graph looks normal" or "This looks weird" [in my eyes the first growth graph is insane, that she NEVER was growing at a slower rate than Glider, actually didn't make sense]
What doesn't make sense, though? First of all, she does drop below Glider's growth at one point, so if you're trying to say that there was some kind of bot automatically keeping her ahead of Glider, that doesn't seem to be the case. Also, the fact that she's growing more just means she's getting more votes, possibly due to... More people voting for her? What about if you compared Glider's growth to the other two channels? I bet he'd never drop below them, either. Because he was cheating? No, because he had more views, more votes, and more support (and because he's way more awesome, of course).
On December 13 2011 15:54 rewired wrote: I voted like 150 times I think one of the real heroes here was Seeker tho, I don't think he logged off all weekend.
EPT![[image loading]](http://i.imgur.com/lkFAl.png)
![[image loading]](http://i.imgur.com/R2oxr.png)
![[image loading]](http://i.imgur.com/hy3GU.png)
Someone link me plox. voted, congrats man, by far the best choice for vote, fully deserved!
