4 ms·
> you have blinders on. Tucker Carlsens videos have tens of millions of views on Twitter. Twitter misrepresents views and impressions. The "views" you see unde
by PartiallyTyped 3y ago
> you have blinders on. Tucker Carlsens videos have tens of millions of views on Twitter.
Twitter misrepresents views and impressions. The "views" you see under the video are impressions. Presumably reached 24ish M views, 110ish M impressions on first episode.
Episode 3 has, presumably, 90M impressions and < .6M likes, < 0.170M retweets etc. The metrics (likes and retweets) are not particularly good for a supposedly high engagement video, are they?
- mjfl 3y agoYou think 600,000 likes is low engagement?
- PartiallyTyped 3y agoNo. I think that in the context, 600k likes is low engagement. We are speaking about 2-3 orders of magnitude difference. For comparison, MKBHD video [1], 1.2 orders of magnitude difference between views and likes; Adam's video [2], 1.4 orders; 40 minute long video essay on House Md [3] (1.5M views with subscriber count of 0.15M) , 1.7 orders of magnitude. Carlson's engagement has 2.17 orders of magnitude difference, while being as long as [1], longer than [2], and far shorter than [3]. All of the videos above are supposed to be far less "entertaining" and are expected to have far less engagement value, yet outperformed Carlson's thing. [1] https://www.youtube.com/watch?v=kvN5_GXlg2Y https://www.youtube.com/watch?v=kvN5_GXlg2Y [2] https://www.youtube.com/watch?v=gK5PzV35UPA https://www.youtube.com/watch?v=gK5PzV35UPA [3] https://www.youtube.com/watch?v=xwbB7XkjwHw https://www.youtube.com/watch?v=xwbB7XkjwHw
- codetrotter 3y agoThe voting culture on YT and Twitter could be different. Everyone kept going on about pressing the like button and smashing the subscribe button on YouTube for years. I would not be surprised if people on YouTube use the like button more actively than people on Twitter. Besides, is 1.7 orders of magnitude (from one of your examples) even significantly different from 2.1 orders of magnitude enough to draw any conclusions at all?
- PartiallyTyped 3y ago> Besides, is 1.7 orders of magnitude (from one of your examples) even significantly different from 2.1 orders of magnitude enough to draw any conclusions at all? 10^(2.1) / 10^1.7 gives us about 2.5x lower engagement in Carlson's video over some unknown guy's 40 minute long video on House Md.
- PM_me_your_math 3y agoThis is mostly nonsense. Carlson's show averaged 4.3 million viewers on fox. His latest episode on twitter has almost 16 million.
- PartiallyTyped 3y agoI wouldn't really trust the numbers twitter is putting out. They have no obligation to any shareholders not to lie. Can you, in good conscience, accept that 1 in 8 people saw this tweet [1]? http://web.archive.org/web/20230616134055/https://twitter.com/tamsully12/status/1661360147074097152 http://web.archive.org/web/20230616134055/https://twitter.co...
- jquery 3y agoYeah I'll take Twitter's metrics with a metric ton of salt. They cannot be trusted. A "view" could be two seconds of autoplay with the video half-in-view when it shows up in the "for you" tab. Not real engagement.