Datasets:

mpasila

/

DarkViperAU-Essays

like 0

Necrit disregards the many other ways a reaction can harm an original video than the singular one that he listed, that being that people could see the original video but then decide to skip over it in order to wait for a reaction to come out.

One risk with reactions is how they may impact your video and channel in the algorithm, something fairly unpredictable. It is certainly true that someone might skip an impression for your original video in order to wait for a reaction to it, meaning the algorithm to think them less interest in your videos in general. But the algorithm could also just not recommend the original video and instead recommend the reaction and the reactor’s content. Worse is that some of the people who end up watching the reaction will afterwards skip over a recommendation for the original video if they're given it. I mean why would they click it, they've already seen the original video just on a different channel.

This would lessen the click-through rate for the original video and convince the algorithm that either the video isn't very good or at least it isn't to that particular viewer's interests.

Having enough people see the reaction and then later skip over impressions to the original video will obviously impact how far the algorithm will spread the original video. At the end of the day, we have no ability to see how well a video would have done in the algorithm if there wasn't a duplicate competing with it. Those who are in the original therefore just have to take it on faith that their video is doing better competing with a copy of it on another channel than it would have done exclusively if it was just on their channel. There are of course reports of spikes in viewer numbers, no change or sharp decline, but all assess short-term impact and each has to grapple with the uncertainty of never being able to see the alternate reality where the reaction didn't exist.

Maybe you saw a small spike in viewers because of a reaction, but unbeknownst to you, that video then gets 10% less views every day for the rest of its life. Maybe you saw no change, but perhaps that video would have started to decline if not for the additional interest generated by that reaction.

Perhaps you saw an initial drop in viewers because of the reaction, but perhaps that reaction was keeping your video longer in the algorithm so over time you ended up gaining more viewers than you lost initially.

Unfortunately, creators just vary so much in terms of what content they produce and under what conditions it releases. There is also some variation in the method and the extent to which reactors re-upload other people's content. The result is that a generalizable controlled test of the impact of a reaction on the original video is likely impossible. At the end of the day, one of the major issues with the creator-reactor dynamic is that if you've seen the original video, you can still hypothetically find value in the reaction as it has new material that you haven't seen. If you've seen the reaction, however, you have no reason to watch the original as you've seen all its material. That the original creator did 100% of the work to make the original and 99% of the work to make the react video doesn't matter to viewers.

This obvious imbalance, I think will always give more credence to the idea that there is greater harm than benefit for the original video. But when I say this, I feel I need to stress that the dynamic between a single original video and its reaction is not my main issue with reaction content.

My interest will always be on the total impact of reaction content on the entire market of creators.

The third level. I actually regret calling these levels now, I guess I should've called them problems or something, but whatever.

The entire basis for Necrit's analysis is his belief that if he was losing viewers to reactions, that every additional reaction made of his content would increase the amount of viewers that he's losing. This is because more and more viewers will begin to skip over his videos and wait for the future reactions to watch instead.

This however is a fairly silly assumption to base your entire video off.

In the period where Asmongold reacted to 7 of Necrit's videos, not only were these reactions sometimes months apart, but during that same period, Necrit uploaded around 100 videos. This low ratio of videos to reactions would make it fairly silly for any viewer to have a belief that there was a reaction certainly forthcoming.

Furthermore, implicit here is that Necrit believes that viewers are sitting around contemplating how to min-max their time watching YouTube videos.

I find this very strange because I think viewers just watch the appealing videos that are recommended to them by the algorithm.

There seems to be little reason to believe that there's a widespread behavior amongst viewers to look at a recommendation for a video they want to watch and to have them go, "You know what, I'm gonna skip this one and wait for a reaction." Especially considering that they could just watch both.

As an aside, even if Necrit's assumptions about this behavior were correct, the analysis that he's doing still wouldn't be particularly fruitful. Comparing all reactions to his content to see changes would have best only showed changes to the amount of viewers being lost, not the total amount. In other words, if Necrit was losing a consistent 10% of views on each of his videos, the methodology he's using wouldn't even capture that.

The fourth level, or problem, whichever you prefer.

How he is testing for people skipping over the original video to watch the reaction doesn't make any sense.

This is because obviously the variables that impact each original video and reaction would not remain constants.

The originals and reactions are of different lengths and of course the different topics will play better or worse on different channels.

They release the different days, at different times, they have varying amounts of other videos released alongside them, and so on.

Hell, even the creators themselves and their channels can change between reactions.

Amongst all the noise of confounding variables, even if there was an increase in the amount of people skipping over the original video to wait for a reaction, you wouldn't really expect to see it. This especially when just comparing the view counts in a small sample size of seven pairs.

From the outset, you should be highly skeptical of the claim that all original videos and all reactions just so happen to maintain the exact same viewership ratio.

This leads finally to the fifth and final problem.

To reiterate, Necrit’s claim is that he knows his original videos are not losing views to reactions because the ratio of views that his original videos get compared to Asmongold's reactions remains consistent each time.

In other words, if Necrit gets 100,000 views, Asmongold will get 300,000 views. If Necrit gets 200,000 views, then Asmongold will get 600,000 views every single time.

He says that we'd have to see Necrit's views go down and the reactor’s go up for there to be any possibility of a loss.

But staggeringly, if you just look at the numbers comparing Necrit's videos to Asmongold's, the ratio of viewership isn't consistent at all. The ratio is in fact all over the place, ranging from Asmongold getting 1.8 times as many views in one case and four times as many views in another case.

I even did this same analysis for another creator that Asmongold has reacted to far more, Josh Strife Hayes. I of course also found a huge difference in the ratio of views between the original and reactions.

These range from Asmongold getting half as many views as the original to getting six times as many views.

If Necrit is an honest person, it's very strange for him to make this video and at no point seek to do this analysis with another creator as well. Like he obviously had every ability to compare other people's videos to the reactions that are done by Asmongold, but he just didn't do that and I wonder why?

On the surface, these results appear to entirely contradict Necrit's claims. But if you examine Necrit's video closely, what he actually seems to mean when he says the ratios are "the same" is that they exist in some undisclosed range of similarity that he has arbitrarily decided to call "the same".