The real limitation of the S&D model whose graph ripster likes to put up is that it assumes some conditions that ripster doesn't. He is committing a logical fallacy. You can't just ignore assumptions of a theorem, especially when some are wrong (see [1] , [2] below)
See, the S&D model assumes a ceteris paribus condition (all else being the same) that is almost never satisfied in real life (except in commodities markets, etc.) The lack of this condition being satisfied means that the "true" model has error: expressed in the deviance from the prior sales. It might still mean that demand is diminishing, but it could also mean that there is enough error in the model is now great enough that it can't be statistically valid conclusion.
Speaking of statistics brings me to a second point: size of the data set. See, the S&D model also assumes one is applying it to an ideal, large data set. It is really only useful for tracking trends, but the utility is significantly reduced when one is monitoring a small data set. We cannot make conclusions about the market with confidence until we have more data. If ripster is happy with his conclusion, I am saddened by the vastness of the confidence interval on it (probably something like 35%)
These two assumptions that the S&D model takes combine to reduce the value of his conclusion (or any other one really, for example ,I can't take this and say "the demand is increasing" either) quite a bit below any sort of useful or rational level.
I would also like to note that he makes the logical fallacy of "post hoc, ergo propter hoc" ("After this, therefore because of this") relatively often as well, and would do good to avoid it in the future.
Other fallacies to note (courtesy of the great wikipedia):
Ludic fallacy [1]
Fallacy of the single cause [2]
Regression Fallacy [3]
[1] and [2] correspond with the use of the S&D graph and principle. [3] refers to the usage in the second point above.
I would also note that ripster's arguments from repetition (
Ad Nauseum) are generally also annoying.
Finally: I would like to agree with "sam" and note that my first name [shortened] is "Sam".
