It is, but I don't buy it at all. If you selectively extract segments of footage, you could support any theory at all. For example, you could propose a theory that if someone touches their nose twice and then their ear, they are lying. You then just trawl the hours of footage until you find...