EPTTop 100 streamers of 2013 (+ Top 50 of December) - Page 3
| Forum Index > SC2 General |
|
Isualin
Germany1903 Posts
| ||
|
Kevin_Sorbo
Canada3217 Posts
On January 06 2014 07:47 Isualin wrote: I don't understand how can you stream 4k hours in a year. It just doesn't add up in my head. You have to stream, sleep and eat, nothing else. If he streamed sc2 for all that time he must be the most passionate guy in the scene the most shocking part is when you consider he hasnt gotten into gm after all that  | ||
|
Squat
Sweden7978 Posts
| ||
|
Noocta
France12578 Posts
On January 06 2014 07:47 Isualin wrote: I don't understand how can you stream 4k hours in a year. It just doesn't add up in my head. You have to stream, sleep and eat, nothing else. If he streamed sc2 for all that time he must be the most passionate guy in the scene Honestly, it's not passion at this point. PainUser has some notorious problems with insomnia and probably uses streaming as a way to keep himself busy. Not very healthy at all tho. It's mindless play. | ||
|
BoX
United States214 Posts
| ||
|
Destiny
United States280 Posts
For example. Let's say I announce on twitter that I'm going to be doing a special 3v3 stream marathon for 8 hours where I play with CombatEx and Deezer. Let's say I stream those 8 hours and carry 9,000 viewers for all 8 hours the entire time. Let's say I don't stream for the entire month after that. I would effectively be first place on the first chart, simply because I only streamed during my absolute peak hours. I would much rather stream 150 hours during the month carrying 3k viewers (average) than 8 hours during the month carrying 9k viewers, however, because there is far far far more revenue involved in the second one. There's also the undeniable affect that there is a certain novelty tied to streaming, such that if you only stream for a few hours in a given month, you will always get more notoriety than if you had a constant, steady streaming schedule. It just irks me to see the top three streamers from 2013 streamed 10 hours, 8.5 hours, then 8 hours, and then next on the list is me with 147.75. /butthurt Regardless, cool information as always! | ||
|
Koshi
Belgium38799 Posts
| ||
|
Squat
Sweden7978 Posts
It is interesting though that the top 3 streamers are all(or were, in Stephano's case) high level competitive players(no offense Steven), who stream more as a way to relax or to interact with fans more than a source of revenue, or so I would assume. I do like the idea of you playing a 3v3 with Deezer and CombatEx though. Maybe bring in a drunk 2GD to cast and you'd make magic happen. | ||
|
MtlGuitarist97
United States1539 Posts
On January 06 2014 08:51 Koshi wrote: I think the first graph represents better who I would like to see play than the second. Yeah, although the second is better representative of the (generally speaking) more hard-working streamers and more consistent streamers. | ||
|
Ace Frehley
2030 Posts
But you should discount at least 300 hours he spent sleeping on stream, lol /jk | ||
|
Conti
Germany2516 Posts
On January 06 2014 08:40 Destiny wrote: This guy is really great with data collecting, but the one thing that always bothered me is that his first graph is marked as "top 50 streamers", and then his next one is marked "top 50 streamers by viewer hours". Obviously I have incentive to be hugely bias here, but I feel like the second graph is more important than the first. The second one is much more indicative of revenue generated + time spent streaming than the first. For example. Let's say I announce on twitter that I'm going to be doing a special 3v3 stream marathon for 8 hours where I play with CombatEx and Deezer. Let's say I stream those 8 hours and carry 9,000 viewers for all 8 hours the entire time. Let's say I don't stream for the entire month after that. I would effectively be first place on the first chart, simply because I only streamed during my absolute peak hours. I would much rather stream 150 hours during the month carrying 3k viewers (average) than 8 hours during the month carrying 9k viewers, however, because there is far far far more revenue involved in the second one. There's also the undeniable affect that there is a certain novelty tied to streaming, such that if you only stream for a few hours in a given month, you will always get more notoriety than if you had a constant, steady streaming schedule. It just irks me to see the top three streamers from 2013 streamed 10 hours, 8.5 hours, then 8 hours, and then next on the list is me with 147.75. /butthurt Regardless, cool information as always! I got Destiny butthurt. That has to be some kind of life accomplishment, right?  It all comes down to the definition of a "top streamer". If you want to find out which streamer most likely made the most money, or which streamers can or do live from their streaming, then the second list is obviously the one you should look at. However, what if you want to find out which players are trending right now in their popularity? That's when you can go to the first list. There's no question that Scarlett draws viewers right now, whether it's her stream or a tournament she's playing in. Same with Stephano, and to a lesser extend, Jaedong. All three are still hugely popular, and I'd say the first list represents that quite well. The lists answer different questions. And you're absolutely right that you can "cheat" the first list in various ways. You can also hurt your own numbers for the first list by streaming in undesirable times (during tournaments/WCS, during times bad for Europeans, etc.). These are very clear problems with sorting the streamers just by their average viewership. I try to counteract this at least a little bit by requiring at least 5 hours streamed in a month (Otherwise, HuK and Innovation would have made it into the top 10 for December, too). I might raise that number to 10 hours, but if I would publish last month's list without Scarlett, Stephano and Jaedong anywhere, half of the posts in this thread would ask for why these players are missing even though they've clearly been streaming. I hope you can see my dilemma here. Currently I'm thinking of raising the limit to 8 hours, on the logic that this is basically one work day, and everyone who streamed more than that has, theoretically, streamed on at least two different occasions, giving a somewhat more realistic average viewership. But that's obviously not a perfect solution, either. However, the viewers*hours list is not perfect, either. Some streamers have made it into that list by simply leaving their stream on while they sleep, getting 20-50 viewers for free, so to speak, for 8-10 hours. In addition (and this is an entirely subjective opinion of mine) I think the second list "rewards" players who stream a lot a bit too much, compared to players who get good viewer numbers. A player who averages 500 viewers and plays for 200 hours has a higher V*H rating than a player with 3000 viewers playing for 25 hours. That just feels.. odd. Again, I'm being quite subjective here. I started out my threads with only the top list, then later added the second upon suggestions from other people. And since I had to differentiate the two somehow, I added "sorted by viewers*hours" to its name, and never bothered to change the name of the first list. Maybe I should just call the first list "Top 50 streamers sorted by viewers", instead of implying that it is the "main" list. By now I do not consider one list better or worse than the other, but of course I do have to chose one list to display first, and it just so happened that the one sorted by average viewers is the one that made it. Perhaps I should start a poll and find out if people think I should change that. | ||
|
Destiny
United States280 Posts
I can definitely see the appeal of that, but it's hard to measure that with even the first list. It's definitely true that Scarlett's stream drew a ton of attention when she streamed, same as Jaedong. But that was the first time they'd streamed, and there was a huge novelty/hyper surrounding the event. Say Scarlett did the exact same stream the next day and held just a couple thousand concurrents less, she'd effectively be lowering her "Viewers" number due to the averaging of the figures. Of course, since she's so high on that list it doesn't affect her much at all, but in closer ratings it could have some effect. It'd be neat/interesting if there was some way to get an idea of how many unique viewers someone is able to get, as I feel like that could be a good indicator (in conjunction with other stats) for the relative popularity of a person. >I try to counteract this at least a little bit by requiring at least 5 hours streamed in a month (Otherwise, HuK and Innovation would have made it into the top 10 for December, too). It seems like 5 hours is kind of arbitrary, though? Again, you said that if you cut it off at 8 or 9, a lot of people would be asking about the big three, so I can understand the problems. >I hope you can see my dilemma here. Heh, my post was just an observation, I don't think it's worth it for anyone to get insanely upset at any of the lists here. >That just feels.. odd. Again, I'm being quite subjective here. Yeah, I definitely agree with that, when you get to extremes you can see the problems with going by only viewer hours. While it might urk someone who averages 4k to have someone above them who streams for only 8 hours a month averaging 8k, I can definitely see a problem with someone who averages 2k getting beat out by someone who averages <300-400 just because of the sheer number of hours they stream. As I said earlier, it'd be interesting if you could get the uniques who watch a stream, as I believe there's interesting stuff you could do with access to that number. Anyway, regardless, great work on the data mining! | ||
|
mihajovics
179 Posts
| ||
|
Monochromatic
United States999 Posts
That way players who stream for a short amount of time would get lower then those who streamed for far longer, yet the amount of hours matters less the higher it gets? | ||
|
JustPassingBy
10776 Posts
On January 06 2014 09:46 Destiny wrote: >However, what if you want to find out which players are trending right now in their popularity? I can definitely see the appeal of that, but it's hard to measure that with even the first list. It's definitely true that Scarlett's stream drew a ton of attention when she streamed, same as Jaedong. But that was the first time they'd streamed, and there was a huge novelty/hyper surrounding the event. Say Scarlett did the exact same stream the next day and held just a couple thousand concurrents less, she'd effectively be lowering her "Viewers" number due to the averaging of the figures. Of course, since she's so high on that list it doesn't affect her much at all, but in closer ratings it could have some effect. It'd be neat/interesting if there was some way to get an idea of how many unique viewers someone is able to get, as I feel like that could be a good indicator (in conjunction with other stats) for the relative popularity of a person. >I try to counteract this at least a little bit by requiring at least 5 hours streamed in a month (Otherwise, HuK and Innovation would have made it into the top 10 for December, too). It seems like 5 hours is kind of arbitrary, though? Again, you said that if you cut it off at 8 or 9, a lot of people would be asking about the big three, so I can understand the problems. >I hope you can see my dilemma here. Heh, my post was just an observation, I don't think it's worth it for anyone to get insanely upset at any of the lists here. >That just feels.. odd. Again, I'm being quite subjective here. Yeah, I definitely agree with that, when you get to extremes you can see the problems with going by only viewer hours. While it might urk someone who averages 4k to have someone above them who streams for only 8 hours a month averaging 8k, I can definitely see a problem with someone who averages 2k getting beat out by someone who averages <300-400 just because of the sheer number of hours they stream. As I said earlier, it'd be interesting if you could get the uniques who watch a stream, as I believe there's interesting stuff you could do with access to that number. Anyway, regardless, great work on the data mining! Just a suggestion, how about calling the first list "top 50 streamers by average viewers", the second list "top 50 streamers by viewers*hours", and letting each reader decide which list, if any at all, he/she considers to represent the "top 50 streamers". And thanks to conti for the list, always interesting to read.  | ||
|
mrRoflpwn
United States2618 Posts
Not surprised that Idra is at the top of overall viewers though, he is a full time streamer now. | ||
|
Minigun
619 Posts
On January 06 2014 13:40 mrRoflpwn wrote: Painuser really needs to find something to do with his life. He clearly has no place in starcraft esports anymore. Its just the hard truth :/ Not surprised that Idra is at the top of overall viewers though, he is a full time streamer now. God forbid he does something he enjoys... | ||
|
Conti
Germany2516 Posts
On January 06 2014 09:46 Destiny wrote: >However, what if you want to find out which players are trending right now in their popularity? I can definitely see the appeal of that, but it's hard to measure that with even the first list. It's definitely true that Scarlett's stream drew a ton of attention when she streamed, same as Jaedong. But that was the first time they'd streamed, and there was a huge novelty/hyper surrounding the event. Say Scarlett did the exact same stream the next day and held just a couple thousand concurrents less, she'd effectively be lowering her "Viewers" number due to the averaging of the figures. Of course, since she's so high on that list it doesn't affect her much at all, but in closer ratings it could have some effect. Yes, that's definitely true. But I could argue that, if Scarlett would stream every day for 5 hours instead of once every three months, people would get used to her presence. A sort of oversaturation would kick in. The same would happen if she would appear in practically every tournament. Rarity can be one of the many factors of a player's popularity, and the average viewer numbers, incidentally, reflect that, too. I'd say that Scarlett would be slightly less popular if we could watch her (on streams and in tournaments) all the time. It's a minor point in the end, and I'm not trying to say that the average viewer numbers are an exact measurement of a player's popularity. But it is still far better than the V*H category at the end of the day. On January 06 2014 09:46 Destiny wrote: It'd be neat/interesting if there was some way to get an idea of how many unique viewers someone is able to get, as I feel like that could be a good indicator (in conjunction with other stats) for the relative popularity of a person. Unique viewers would be a much cooler number to have, absolutely. Not just for player streams, but for tournaments especially. But twitch does not provide them, unfortunately, and most likely never will. I'm not quite sure how they could provide the numbers, anyhow. Provide the number of unique viewers per streaming session at the end of a stream, I suppose? On January 06 2014 09:46 Destiny wrote: >I try to counteract this at least a little bit by requiring at least 5 hours streamed in a month (Otherwise, HuK and Innovation would have made it into the top 10 for December, too). It seems like 5 hours is kind of arbitrary, though? Again, you said that if you cut it off at 8 or 9, a lot of people would be asking about the big three, so I can understand the problems. Yeah, the 5 hours are entirely arbitrary, like any other cut-off number. I suppose I at least have a half-assed argument for raising it 8 hours, but in the end, there's no objective way to find the right amount of streaming hours required here. And not having any at all just leads to even more obvious problems (I remember MarineKing having streamed for 0.75 hours in one month, theoretically making it to the top 5..) Thanks for the feedback, though! It's always appreciated. | ||
|
Aeromi
France14462 Posts
| ||
|
Conti
Germany2516 Posts
On January 07 2014 03:40 Aeromi wrote: Golden is teamless. Indeed he is. And here I thought I checked all the Koreans. Thanks! | ||
| ||
