Hi @a-data-scientist, welcome to Steem and we're glad your first post is an Utopian analysis contribution. Great start! :)

Thanks :)

I'm surprised to see steemhunt that low in your engagement metric - how did you calculate the numbers? I think only the number of children to root posts should go in this metric, otherwise this would be biased by voted comments with no further replies.

I counted all the comments that were made on the posts upvoted by Steemhunt and then divided it to the total number of posts.

Steemhunt is also special in a way that it ranks all posts/comments from a period of 24h and then starts a voting round. This is why you found most of their votes at around the same time of the day.

Now I got it!

The daily/monthly upvote graphs per app are a bit hard to compare since the bins are ordered by value instead of all by day/month.

Yes, this is what I realized too after posting this analysis, will take care of it next time.

Looking forward to more analyses from you! :D

Thanks a lot, @crokkon :)