I'd say there can be a massive element of luck in a game, but it's not mandatory. You raised the point some time back about certain analysts disregarding 1-0 wins I think, and I certainly see the validity in that.
Backing a higher team without form or lower with form?.. If you were a certain type then you'd say class was permanent and back that way regardless I suppose. I'm not a betting person of any degree, but I'd guess the bookies in your example will still shorten the odds of Chelsea over Swansea, after all other factors have been taken in consideration.
What happened to Leicester's matchday odds in their title winning season? Did the bookies follow the then recent evidence and make them favorites as the season progressed, or were they still classed as underdogs when playing the likes of Man Utd?
Whilst the bookies put out initial odds it's the punters that decide them by the money they bet. It's a misconception that bookies have a view on whether a team will win, draw or lose - they simply gauge what the punters think in order to make a market where they are guaranteed a win whatever the result.
The comparison with Leicester is interesting but there's one much closer to home. When the Cobblers were on their winning run in the promotion year you could always get way better odds than the more fancied teams such as Oxford and the like. Punters assumed the bubble would burst much as they did with Leicester and there must have been quite a few Cobblers fans than won silly money over that time.
I think people can be lazy when looking at lower league teams and simply go for the names. For example Portsmouth were always skinny odds no matter how badly they were doing.