None of those are 'toys', which was the point of what I was saying... if you are counting toys for a 'Transformers Count', polls on those sort of things (and most of the Articles) would have been unnecessary. It just over-complicated a process that could have been a lot simpler (and more objective) from the outset.
And personal bias doesn't boost collections, it prohibits others from counting *toys* they have in their collection. I have a room full of toys, but according to what other people have voted, a large number aren't toys. I support a universal method for statistical purposes (like this project), but that just bugs me.
Which was the point of the 'devils advocate' bit on how accurate is 'an average' if the sample polled is so different each time? (I'm not saying a poll shouldn't be taken - just that people participating or compiling shouldn't get too worked up over its accuracy or parameters) It's like surveying people about what their religion is, and each time it involves different people... statistically it would appear that the demographic has changed, but those people in the first survey may not have changed their religion at all...
As per my posting... if they are toys, and they are legit Transformers (r), they count. As per the existing UCM, you may have to take a poll to see what the 'collectors' think.