8 ms·

Analysis of all YouTube trending videos in US for 2019

100 points by ammar_x 6y ago

symplee 6y ago
Another excellent analysis of what Trends on YouTube [0] "Is the trending tab rigged against creators? Is late night dominating the charts? Why does it feel your favorite controversial creator never trends. All of that and more, in the first ever data-driven look at the trending tab." What 40,000 Videos Tell Us About The Trending Tab: [0] https://www.youtube.com/watch?v=fDqBeXJ8Zx8 https://www.youtube.com/watch?v=fDqBeXJ8Zx8
- ammar_x 6y ago
  I like that. Excellent storytelling, inspiring. Thank you for sharing.
pvankessel 6y ago
This is really great - and kudos for providing methodological details and code! I love seeing this kind of large-scale descriptive research, it's a real bummer that YouTube is starting to close up access to their API. We did a similar kind of analysis looking at videos posted by popular channels last year, including some analysis of keywords that boosted views - figure you might find it interesting (and I'd love to see if our findings hold up with your dataset!) https://www.pewresearch.org/internet/2019/07/25/a-week-in-the-life-of-popular-youtube-channels/ https://www.pewresearch.org/internet/2019/07/25/a-week-in-th...
- ammar_x 6y ago
  Thank you.. The analysis you've done seems really interesting and gave me some more ideas to implement in future analyses. For keywords that boost views, I'll run a similar analysis on my data and report the results to you here in another comment. Probably this weekend. And I just heard from you that YouTube is starting to close up access to their API. I visit YT API website from time to time but haven't noticed such a thing. It's really a bummer. They should be more open not the opposite. Do you have some source if I want to know more about that?
  - pvankessel 6y ago
    Would love to see what you come up with, will stay tuned! As for the API restrictions, they aren't advertising it but about a year ago they started warning users about forthcoming extensive audits to maintain access, and about six months ago they started reducing access for API keys if you stopped maxing them out for a day or more. Our last API key got shut down for good a couple of weeks ago. We're going to try to fill out the form and get our access reinstated, but I'm not sure how willing they'll be to allow access for research. The form seems intended for client-facing apps. But who knows - Facebook/CrowdTangle/Twitter have been very supportive of legitimate research initiatives, I'm hoping YouTube follows that trend!
    - ammar_x 6y ago
      Hey, I was not able to do the analysis we talked about. The analysis received much interaction after I published it here and was contacted by many people. As a result, I've been too busy. I hope I can do it soon. When I do it, I'll notify you in a comment here. Or provide me with your email.
    - pvankessel 6y ago
      To clarify - the current API has a limit of 10,000 "query points" per day for new API keys (most endpoints cost 1-5 points per query). It used to be 1 million; they've since throttled everyone down and started forcing audits. 10k is still something, but it certainly doesn't allow large scale research.
      - macromaniac 6y ago
        It used to be 50 million, it used to be 500000% more. I would have expected it to increase as bandwidth got cheaper yet it for whatever reason has decreased. https://stackoverflow.com/questions/15568405/youtube-api-limitations https://stackoverflow.com/questions/15568405/youtube-api-lim...
        pvankessel 6y ago
        Wow, I had no idea. 1m was enough to work with, but... wow. Thanks for the reference.
leereeves 6y ago
> The minimum number of views a video had when it first became trending was 53,796 views. How does that happen?
- ammar_x 6y ago
  There is a video that had only 53,796 views on the day it became trending. That video is "Phora - Don't Change [Official Music Video]". If you look at the data, you find that it became trending on the same day it was published. So before accumulating a large number of views, it became trending. Maybe it acquired thousands of views very quickly which sent it to the trending list? Maybe other factor.
  - markdown 6y ago
    Probably easily done if the publisher has a large fanbase and uses other social media channels to send them all to Youtube the moment the video is published.
e737ueuryryy 6y ago
None of this is really that surprising though. Progressive cancel culture is an American phenomenon and was a massive point of criticism when YouTube announced whatever their volunteer moderator program is currently called. Google is a political platform for its employees and the only time we see that not be true is when the activism tries to jump into a domain that requires trustworthy employees like government contract work. Traditional media can overcome this because they have these things called budgets which independent creators lack until they make it big. Traditional media can simply throw money around until they make it on the board. What we're seeing isn't an accident of some algorithm, it's what happens when a company is so saturated with children that it simply can't operate as a creatives-first platform.
wolco 6y ago
From my understanding any channels you subscribe to with new content/video have a good chance to appear there. This is for the generic viewer.
ve55 6y ago
I find it kind of sad looking at what popular Youtube channels looked like 5-10+ years ago, and comparing it to their videos today. As you scroll down the list of videos from oldest to newest, the lengths get longer. No longer are there any petty three minutes or five minute videos, let alone a one minute video. Now videos are all 10-25 minutes, the perfect length for maximum ad revenue. Similar things happen to the thumbnails, which start containing more and more exaggerated, sometimes creepily-so, faces, reacting to absurd images that last for a fraction of a second in the video. The titles change too, becoming more and more outrageous, begging for you to click on them and spend hours watching anticipatory and 'exciting' content. A lot of things have gotten better too, of course, and the amount of content that is available is amazing. But I do really miss when it felt like a place where people went to share videos and socialize, rather than where all the popular channels are ran like companies that follow the same clickbait format news organizations now also follow.
- javajosh 6y ago
  Doesn't the one activity wholly subsidize the other?
  - ve55 6y ago
    They might to an extent, but I don't think it's required or necessary. Ads can go on any type of videos, and if Youtube had less traffic as a result of less clickbait, then presumably the costs to host it would be a lot less too. Apart from this, keep in mind that Youtube TV likely pulls in a ton more money, and even if it didn't, Google/Alphabet has many other larger sources of revenue that could pay for Youtube too.
- graedus 6y ago
  > Similar things happen to the thumbnails, which start containing more and more exaggerated, sometimes creepily-so, faces, reacting to absurd images that last for a fraction of a second in the video. There was a nice little examination of this phenomenon, dubbed YouTube Face, a couple of years ago. Article: https://openspace.sfmoma.org/2018/04/your-pretty-face-is-going-to-sell/ https://openspace.sfmoma.org/2018/04/your-pretty-face-is-goi... HN thread: https://news.ycombinator.com/item?id=16803937 https://news.ycombinator.com/item?id=16803937
angel_j 6y ago
This is a lot of work for a lot of practically meaningless results. These are only examples: The urls and social links in the descriptions. The "the" in the titles. The other findings offer very little insight. Be popular, post videos.
rienbdj 6y ago
Cool analysis but I don’t think any interesting conclusions came from it.
jcahill 6y ago
My web archival group (nonprofit) has international trending data for the same period. I've been meaning to get around to analysis for a while. This article is a good point of reference.
- ammar_x 6y ago
  Glad it helped. If possible, could you share your analysis results with us after you finish?
_ZeD_ 6y ago
I do wonder what would be the results for other countries - like in the EU zone (it would be interesting, moreover, to see if / what are the differences between denmark, germany, italian etc.. youtube trending videos)
- ammar_x 6y ago
  I have the same data for Canada, Denmark, UK, Malaysia, and other 13 countries. I plan to analyze this data soon. So stay tuned.

100 points by ammar_x 6y ago

Another excellent analysis of what Trends on YouTube [0] "Is the trending tab rigged against creators? Is late night dominating the charts? Why does it feel your favorite controversial creator never trends. All of that and more, in the first ever data-driven look at the trending tab." What 40,000 Videos Tell Us About The Trending Tab: [0] https://www.youtube.com/watch?v=fDqBeXJ8Zx8 https://www.youtube.com/watch?v=fDqBeXJ8Zx8

ammar_x 6y ago
I like that. Excellent storytelling, inspiring. Thank you for sharing.

pvankessel 6y ago

This is really great - and kudos for providing methodological details and code! I love seeing this kind of large-scale descriptive research, it's a real bummer that YouTube is starting to close up access to their API. We did a similar kind of analysis looking at videos posted by popular channels last year, including some analysis of keywords that boosted views - figure you might find it interesting (and I'd love to see if our findings hold up with your dataset!) https://www.pewresearch.org/internet/2019/07/25/a-week-in-the-life-of-popular-youtube-channels/ https://www.pewresearch.org/internet/2019/07/25/a-week-in-th...

ammar_x 6y ago
Thank you.. The analysis you've done seems really interesting and gave me some more ideas to implement in future analyses. For keywords that boost views, I'll run a similar analysis on my data and report the results to you here in another comment. Probably this weekend. And I just heard from you that YouTube is starting to close up access to their API. I visit YT API website from time to time but haven't noticed such a thing. It's really a bummer. They should be more open not the opposite. Do you have some source if I want to know more about that?
- pvankessel 6y ago
  Would love to see what you come up with, will stay tuned! As for the API restrictions, they aren't advertising it but about a year ago they started warning users about forthcoming extensive audits to maintain access, and about six months ago they started reducing access for API keys if you stopped maxing them out for a day or more. Our last API key got shut down for good a couple of weeks ago. We're going to try to fill out the form and get our access reinstated, but I'm not sure how willing they'll be to allow access for research. The form seems intended for client-facing apps. But who knows - Facebook/CrowdTangle/Twitter have been very supportive of legitimate research initiatives, I'm hoping YouTube follows that trend!
  - ammar_x 6y ago
    Hey, I was not able to do the analysis we talked about. The analysis received much interaction after I published it here and was contacted by many people. As a result, I've been too busy. I hope I can do it soon. When I do it, I'll notify you in a comment here. Or provide me with your email.
  - pvankessel 6y ago
    To clarify - the current API has a limit of 10,000 "query points" per day for new API keys (most endpoints cost 1-5 points per query). It used to be 1 million; they've since throttled everyone down and started forcing audits. 10k is still something, but it certainly doesn't allow large scale research.
    - macromaniac 6y ago
      It used to be 50 million, it used to be 500000% more. I would have expected it to increase as bandwidth got cheaper yet it for whatever reason has decreased. https://stackoverflow.com/questions/15568405/youtube-api-limitations https://stackoverflow.com/questions/15568405/youtube-api-lim...
      - pvankessel 6y ago
        Wow, I had no idea. 1m was enough to work with, but... wow. Thanks for the reference.

leereeves 6y ago

> The minimum number of views a video had when it first became trending was 53,796 views. How does that happen?

ammar_x 6y ago
There is a video that had only 53,796 views on the day it became trending. That video is "Phora - Don't Change [Official Music Video]". If you look at the data, you find that it became trending on the same day it was published. So before accumulating a large number of views, it became trending. Maybe it acquired thousands of views very quickly which sent it to the trending list? Maybe other factor.
- markdown 6y ago
  Probably easily done if the publisher has a large fanbase and uses other social media channels to send them all to Youtube the moment the video is published.

e737ueuryryy 6y ago

None of this is really that surprising though. Progressive cancel culture is an American phenomenon and was a massive point of criticism when YouTube announced whatever their volunteer moderator program is currently called. Google is a political platform for its employees and the only time we see that not be true is when the activism tries to jump into a domain that requires trustworthy employees like government contract work. Traditional media can overcome this because they have these things called budgets which independent creators lack until they make it big. Traditional media can simply throw money around until they make it on the board. What we're seeing isn't an accident of some algorithm, it's what happens when a company is so saturated with children that it simply can't operate as a creatives-first platform.

wolco 6y ago

From my understanding any channels you subscribe to with new content/video have a good chance to appear there. This is for the generic viewer.

ve55 6y ago

I find it kind of sad looking at what popular Youtube channels looked like 5-10+ years ago, and comparing it to their videos today. As you scroll down the list of videos from oldest to newest, the lengths get longer. No longer are there any petty three minutes or five minute videos, let alone a one minute video. Now videos are all 10-25 minutes, the perfect length for maximum ad revenue. Similar things happen to the thumbnails, which start containing more and more exaggerated, sometimes creepily-so, faces, reacting to absurd images that last for a fraction of a second in the video. The titles change too, becoming more and more outrageous, begging for you to click on them and spend hours watching anticipatory and 'exciting' content. A lot of things have gotten better too, of course, and the amount of content that is available is amazing. But I do really miss when it felt like a place where people went to share videos and socialize, rather than where all the popular channels are ran like companies that follow the same clickbait format news organizations now also follow.

javajosh 6y ago
Doesn't the one activity wholly subsidize the other?
- ve55 6y ago
  They might to an extent, but I don't think it's required or necessary. Ads can go on any type of videos, and if Youtube had less traffic as a result of less clickbait, then presumably the costs to host it would be a lot less too. Apart from this, keep in mind that Youtube TV likely pulls in a ton more money, and even if it didn't, Google/Alphabet has many other larger sources of revenue that could pay for Youtube too.
graedus 6y ago
> Similar things happen to the thumbnails, which start containing more and more exaggerated, sometimes creepily-so, faces, reacting to absurd images that last for a fraction of a second in the video. There was a nice little examination of this phenomenon, dubbed YouTube Face, a couple of years ago. Article: https://openspace.sfmoma.org/2018/04/your-pretty-face-is-going-to-sell/ https://openspace.sfmoma.org/2018/04/your-pretty-face-is-goi... HN thread: https://news.ycombinator.com/item?id=16803937 https://news.ycombinator.com/item?id=16803937

angel_j 6y ago

This is a lot of work for a lot of practically meaningless results. These are only examples: The urls and social links in the descriptions. The "the" in the titles. The other findings offer very little insight. Be popular, post videos.

rienbdj 6y ago

Cool analysis but I don’t think any interesting conclusions came from it.

jcahill 6y ago

My web archival group (nonprofit) has international trending data for the same period. I've been meaning to get around to analysis for a while. This article is a good point of reference.

ammar_x 6y ago
Glad it helped. If possible, could you share your analysis results with us after you finish?

_ZeD_ 6y ago

I do wonder what would be the results for other countries - like in the EU zone (it would be interesting, moreover, to see if / what are the differences between denmark, germany, italian etc.. youtube trending videos)

ammar_x 6y ago
I have the same data for Canada, Denmark, UK, Malaysia, and other 13 countries. I plan to analyze this data soon. So stay tuned.