Last year Classic Wine magazine conducted a blind tasting of 46 brandies, both blends and potstills. Brandies were scored between 12.7 and 19 points out of 20 by a panel of 5 tasters. Of course, the elephant in the drawing room at Classic Wine (and also at the controversial Platter sighted guide that last year waded back in to the water to rate brandy) is that the most important consumer variable, price, is ignored. Which is a totally ridiculous state of affairs for consumers in whose name the whole exercise takes place.

A further problem is the non-linear nature of the scoring scale. The difference between a brandy scored 14 (so-s0) out of 20 and one scored 15 (average) is much smaller than the difference between an 18 (excellent) and a 19 point orgasmic stunna. Even though in both cases, the numerical difference is 1. How to made the tasting process more useful to consumers?

Wotwine? have come up with a novel solution to the problem. Rather than a score, assign a cash value to a wine (or brandy, in this case) after tasting it. Then consumers can compare whether the retail price asked makes the product a gimme or a give it the flick. The ratio of what a tasting panel reckons the product is worth to what it costs, gives a notional value.

While ratings are not linear, neither are prices, as the graph above shows. The curve is simply recommended retail prices sorted from smallest to most expensive and smoothed to remove artefacts. Clearly there are two (or possibly three) linear sections here: a ramp from R80 to just under R400 and then the plutocratic steep hike to over R1000 for the most expensive brandy tasted, the Van Ryn 20 year old.

The value distribution of expected price based on score to RRP above, flags two clear bargains: the KWV 5 year old and Richelieu International. Likewise any brandy with a value less than 1 is not a good deal. There are 17 such duds while 23 are bargains. No RRPs were available for 6 brandies and so they were ignored.

The methodology opens up a fascinating and useful tool for rating just about anything. Of course the price function for Pinot Noir is unlikely to be the same as the one for Pinotage. And while the Pinotage Association may stamp its tiny foot, it is a commercial reality. Likewise the Chardonnay curve and the one for Chenin will likely be quite different.

Value is an interesting variable indeed for it is unitless, being the ratio of a price predicted from a tasting score divided by the RRP. So values can be compared between brandies, Bordeaux blends and Burgundies. The methodology has a great future as a tool for consumer decision making.