Not sure I agree with your title Melmoth. On a sample of several hundred games that I have bought over the years at every price bracket from bargain bucket up to day of release full price I can find no correlation whatsoever between the price I paid for a game and my estimation of it.
I believe that there’s a difference in picking up Beyond Good & Evil in the bargain bin because you’ve heard good things about it, and seeing L4D on Steam at half price and making the purchase because that price is almost too good to be true.
I was going to put an additional line in the post mentioning that it would be interesting to see how many of those copies of Left 4 Dead were still being played after the first week or so, and how any drop off in numbers compared to the drop in numbers after the initial retail release. They’re bound to have picked up some new converts because it’s an excellent game, but I wonder what proportion of ‘offer’ purchasers stuck with it compared to those who bought it at full price at release.
I doubt you’d get much but very broad strokes out of it. Maybe some time-based trends, like how the price/evaluation graph changes with economic stresses. (And how do you gauge “evaluation” or “esteem”?) Just looking at the numbers of sales doesn’t clearly represent outliers like early adopters or those who esteem the game highly but estimate the cost to be too high at release.
Of course I’m speaking of experience there; I still value Kingdom Hearts and KH2 as my most valuable games, but I got each of them for a third of the price I paid for a copy of Street Fighter 2 Turbo years ago. “Esteem” doesn’t always correlate well with price points, in other words.
Ok, so to qualify my statement: I think it would make for an interesting study, but I’ve no idea how you’d actually perform it in a rigorous scientific manner. Too many variables, and too hard to quantify.
And at the end of the day, one may simply end-up with a conclusion that is just as bland as ‘people buy more stuff when it’s cheaper’.
At the risk of massively overanalysing a pithy quote, there’s a couple of interesting articles over at Joel on Software: Camels and Rubber Duckies about setting software prices, and Price as Signal. Another interesting (but impossible to carry out) study would be comparative sales if Left 4 Dead had started out at the lower price point.