Top 50 streamers in September 2014 - Page 5

On October 02 2014 03:14 The_Masked_Shrimp wrote:
Those statistics don't make any fucking sense if you don't put in parallel the geographic dispersion of the audience and the time at which the streamers stream on average. I don't even get the point of getting trouble doing all this work because without the former information that is just useless and meaningless. (well it's great to have all that data, don't get me wrong, but don't asspull conclusions and pretend it is what this data means)

Example out of ass:
If A gets more V*H than B, when 80% of the audience is north american, and A mainly streams between 6pm and 11pm EDT while B streams mainly between 3am and 4am EDT then you can't say shit about who is more popular between A and B.

And despite that there are all those people arguing with arguments which are not even correlated to what they use it to argue for.....

The only way for any discussion in this thread to possibly have made sense would have been if all the streamers from the list always streamed at the same time for the same amount of time. (and the V*H does not even slightly compensate for that, because stream hours in different locations are not equally popular themselves).

Seriously....

What this data is good for is to see the time evolution of the average viewer number for SC and for each player individually, and even then it would need to also have a track of the evolution of the streaming schedules of players to be accurate. So Stephano average viewer drop between now and February does not necessarily mean SC or Stephano are less popular. It could just be that Stephano changed the time at which he is streaming to a less popular time slot or that the popularity of the time slot itself could evolve during the year.

When you present statistical data and conclusions based on it, it is your responsibility to give warnings and statements about the possible flaws of its interpretation. If you don't, profane people will believe in false gods and engage in bloody crusades based on irrelevant knowledge like in this thread, wasting both their time and trust.

Even in grad school of Math & Physics most students (just personal experience) are still unable to comment statistical data properly, so be careful with it. And whatever your data, when you draw conclusions ALWAYS assume that there is at least an unkown factor which has been missed and is supposed to affect the results in some way.

On October 02 2014 03:54 Conti wrote:

The data is even less meaningful than that, if you look at it that way. I would not only have to include the times of day the players stream in regards to their corresponding time zone (which is not even static, since players move from/to Korea all the time), I would also have to take into account the days they stream (weekend/non-weekends) and, in addition to that, take into account which other streams/events were streaming at that time (stream while WCS is on? That's gonna hurt your V*H rating).

Maybe I should mention something along those lines in the disclaimer in the OP, though I would say that this is fairly obvious, and that this is quite obviously not an objective list of popularity. But if I would take all these things into account (and have fun doing that month for month), it would still not be an objective list of popularity, because popularity is inherently subjective. I am assuming that people here are smart enough to know that.

On October 01 2014 22:44 SuperHofmann wrote:
Give me a reason to watch him and not a good progamer, please

On October 01 2014 22:35 SC2Toastie wrote:
Isn't it well known and publicly accepted that Winter's stream is filled with bots, which may or may not be his (If I read my sources correctly, it's out of his control, but who knows).
Anyways, I find it slightly odd to have a stream KNOWN to be almost exclusively bots in the top 50 streamers list.

User was warned for this post

On October 02 2014 03:14 The_Masked_Shrimp wrote:
Those statistics don't make any fucking sense if you don't put in parallel the geographic dispersion of the audience and the time at which the streamers stream on average. I don't even get the point of getting trouble doing all this work because without the former information that is just useless and meaningless. (well it's great to have all that data, don't get me wrong, but don't asspull conclusions and pretend it is what this data means)

Example out of ass:
If A gets more V*H than B, when 80% of the audience is north american, and A mainly streams between 6pm and 11pm EDT while B streams mainly between 3am and 4am EDT then you can't say shit about who is more popular between A and B.

And despite that there are all those people arguing with arguments which are not even correlated to what they use it to argue for.....

The only way for any discussion in this thread to possibly have made sense would have been if all the streamers from the list always streamed at the same time for the same amount of time. (and the V*H does not even slightly compensate for that, because stream hours in different locations are not equally popular themselves).

Seriously....

What this data is good for is to see the time evolution of the average viewer number for SC and for each player individually, and even then it would need to also have a track of the evolution of the streaming schedules of players to be accurate. So Stephano average viewer drop between now and February does not necessarily mean SC or Stephano are less popular. It could just be that Stephano changed the time at which he is streaming to a less popular time slot or that the popularity of the time slot itself could evolve during the year.

When you present statistical data and conclusions based on it, it is your responsibility to give warnings and statements about the possible flaws of its interpretation. If you don't, profane people will believe in false gods and engage in bloody crusades based on irrelevant knowledge like in this thread, wasting both their time and trust.Edit: it sounds a bit like I am bashing the OP but no, there is no out of place conclusion in OP post. However there is no indication for people on how to interpret it correclty.

Even in grad school of Math & Physics most students (just personal experience) are still unable to comment statistical data properly, so be careful with it. And whatever your data, when you draw conclusions ALWAYS assume that there is at least an unkown factor which has been missed and is supposed to affect the results in some way.

TL;DR statistics are not to use lightly, please edit OP with proper reading advices and remind people that none of the data can lead to popularity conclusions.

+ Show Spoiler +
Holy fuck you are entitled.


Conti presents the numbers and people draw conclusions. Stupid people draw stupid conclusions. Get over it

On October 02 2014 16:18 The_Masked_Shrimp wrote:


Yes I am, get over it.

At least I developed my point and gave suggestions relevant to the actual thread.

On October 02 2014 04:19 Mistakes wrote:


Yeah it makes sense. I would say that most people who are Master+ want to see Master+ play. Regardless of commentary and whatnot. But I understand that there are far more people below Master than not. A stream I can link would be to CatZ' stream. Partially why I dislike or have a hard time understanding Winter's numbers is when someone like CatZ who is actually a very good and VERY creative player, does stream daily, and has a better looking stream, but also does "Inside My Head", "Music Mode" (where he sings and is goofy), "Tryhard Mode" (no talk, best games possible) and mixtures of each depending on what subs/donators/viewers want to see. Yet I often see him at around 300 viewers. Even lately he has been doing showmatches on his stream vs known pros. (Nestea a couple days ago, DRG last night) either Bo7 or Bo9.

I guess I can say I understand why Winter's numbers are where they are, but what I can't understand is when someone has arguably a better stream (at least in my opinion), is a better player, and still manages to get far less viewers. Maybe it is that marketing degree. Lol. Ah well, enough Winter talk I suppose.

The numbers are interesting and it's really cool to see. I often miss some players because of time zone differences, only to see they've streamed ~50 hours and I completely missed all of it. xD

On October 02 2014 00:54 jtp118 wrote:
it looks like Protech leads a rich and fulfilling life

On October 02 2014 22:13 iMOOrtal wrote:


How is this not warned lol.... biggest troll in the whole thread. So Winter has an army of admins defending people accusing him of V*Bing, but actual attacks on people go free? Zealously where are you? Ned Stark? Winter is coming?

Also please put a disclaimer saying people can't question Winter, I thought this was maybe the only place we could talk about the numbers but I guess not. Although it seems like it is the place to personally attack SC2 players character, which if it is, I will proceed to do if this bro isn't warned. I guess we can't talk about V*B with Winter, but we can insult him personally? Sounds good to me.

On October 02 2014 22:15 Wombat_NI wrote:

An attack/alternatively it could drawing attention to Protech's numbers which indicate that he would have to stream about 12 hours every day for a whole month.