crazy experiments with youtube

I would like to ask for your help in my second youtube experiment.  The question is:  when one searches for something on youtube, youtube shows the results in a certain order.  What factors into that order? 

I have read a lot of stuff over the last week about how some videos get 100,000 views and some get only 10 views.  There are so many factors that may or may not be.  And maybe none of it makes a lick of difference.  It seems wise to at least know.

For those that are interested in a bit of an experiment, I would like to ask the following of you.  I have four sepp holzer videos on youtube.  They are conveniently numbered 1 through 4.   The fourth is the least popular.  Therefore, if we do the things that people say will make a difference, then it should move #4 ahead of #1.

So if you search for "sepp holzer" on youtube,

video 1 is at position 9

video 2 is at position 11

video 3 is at position 14

video 4 is at position 15

So the experiment is:  go to video 4.  Give it five stars.  Add it to your "favorite" list (the little heart just below the stars).  And then add a comment.

As I write this, video 4 has 7 five star ratings, is on four favorite lists, has three comments and 964 views.

Video 1 has 16 ratings averaging 4.88 stars, is on 22 favorite lists, has three comments and 3831 views.

I suspect that if 20 people add stars, add a comment and add to the favorites, then #4 would end up ahead of #1 even though it has a lot fewer views.  Otherwise, all this stuff I've been reading could be a steaming pile of horse potatoes.

Here is video #4: 

Please help me with this experiment - I promise to share the results when I'm done.

Thanks!

FYI:  to rate something "five stars", you must click on the fifth star.

It looks like a lot of folks have helped out with this. 

Video 4 now has 27 five star ratings, is on 21 favorites lists, has 23 text comments and has been viewed 1046 times. 

The search results are the same - but I think it takes a few days for the youtube search stuff to catch up.

prolly off topic but i really liked the "greenhouse rocket mass heater construction" video on youtube.  Paul, you're using this web 2.0 stuff really well.  I continue to be impressed with your mad skills.

chris

Chris K e n d a l l wrote:
prolly off topic but i really liked the "greenhouse rocket mass heater construction" video on youtube.  Paul, you're using this web 2.0 stuff really well.  I continue to be impressed with your mad skills.

Well, here is the thing I came up with yesterday that goes with this: "When you like a youtube video, please encourage the author by:  clicking on the fifth star (ratings), the little heart (favorites) and writing a text comment.  Plus, if you use stumbleupon, please click on the thumbs up and then the talk bubble to leave a comment."

The experiment with this one video may prove this theory to be true.

I think that a lot of people go to youtube to learn something, but there is a mountain of drivel to wade through.  And the stuff for the permies/homesteaders is there, but it is really hard to find.  So it seems like I need to be savvy to what the drivelers are savvy to so that my stuff isn't washed away. 

This experiment will teach me a lot. 

Wow, almost three days have passed and it hasn't budged.  Interesting.

Still no change

still no change.

So, the lesson learned here might be:  the ratings, favorites and text comments all make a difference in the first 24 hours.  And some in the second 24 hours.  But zero difference after that.

We'll give the experiment a little more time to be sure.

day 6:  still no change.

day 7:  still no change.

 

Day 9 and no change. 

I think this is far enough along that we can call it. 

During the first 24 hours, getting lots of stars, hearts and comments can make a big difference in the video getting pushed to the top.  After that, the second 24 hours makes less difference.  And the following five days makes very little difference.  But after seven days, these things make no difference. 

After seven days, I think the only thing that makes a big difference is freaky numbers of views.

A month and a half later:  no change.