To be clear, Alexa isn't something that "VideoSift uses." Rather, it's a broken statistical collection system that the Internet industry (advertising, primarily) still strongly relies on. It's not a system VS has installed or anything, the data collection is performed via web browsers that have the toolbar installed.

I look forward to the day a much more reliable method of web usage quantification becomes the most accepted.

James, I could give you a list of videos and argue from an anecdotal standpoint, but that's not really conclusive (and besides, wouldn't it be easier to just take a poll and ask the other sifters if they've noticed this pattern?). If our debate is of a factual nature, i.e. do videos sit in the middle of the queue as much as I think they do - and I think the type of videos one sifts has a lot to do with it - then the matter is simple to deal with. How's about we set up a data collection (if there isn't one already) that will record when votes are made in terms of a video's life in the queue? I'd be willing to bet that the two kinds of voting patterns will arise, and that there will be a dead zone of voting in the late part of day 1 all the way to the beginning of day 4.

Heck, such data would be invaluable anyways, because voting trends and timing would be great in figuring out all types of adjusting to the queue and such.

By the way if anyone is interested in Videosift.com traffic numbers they are available via Alexa. However these results are highly debatable due to the nature of data collection that Alexa does, so do take them with a grain of salt.

http://www.alexa.com/data/details/traffic_details?q=&url=videosift.com