Originally Posted by jjbir:
“Thats alright, your system is still valid based on the first two predictions. It seems that tonights result suggests that there is a parameter that effects the result that is not covered by just the social media metric. My guess is that whatever that parameter is (and my guess is performance, she may have got a last minute boost because of it) very reactive and unpredictable. I would also say looking at your numbers she would have made it by the skin of her teeth, definitely bottom half.”
Maybe. I'm thinking its the power of rebound voting, performance maybe and possibly visibility depth. It was fighting today with the Niall thing and rate limited on the twitter api several times (which caused darkness for about an hour, maybe 2). Which was a bug, so a few things went wrong.
Well it got Hannah b2 1, Tamera and then Hannah b2 2 so that's 3/4 although the first wasn't published anywhere publically, just to people I know personally when I set out.
But immediately you get someone who will go "well its failed now" and that means its worthless trying to see.
Better luck next week I guess.
What surprises me most is that Hannah competed with such a low social engagement against two guys with buzz that made her look like an also ran. That is surprising.
What you say about performance is interesting as it updates them in sequence and a long process time can cause a delay in updating the next (as the API's are limited to x requests per 15 minute window on twitter). So maybe it was that, that she boosted during the result show and it didn't catch up, they can go up quickly. But that should even out latently.
There is another flaw that can't be addressed. Hannah has no identifiable Facebook account to act as a catalyst and the others do. Even if its a well used fan one, which means she isn't quite even but FB doesn't deliver the same % of information as twitter and YouTube do.