View Single Post
  #9 (permalink)   Report Post  
Dale Williams
 
Posts: n/a
Default

This comes up periodically. I'm no fan of WS. I don't like panel tastings (no
one to "calibrate" to), I don't think that Suckling or Mansson (sp?) match my
tastes very well, etc. I don't subscribe, though I did in my more innocent
years, and have long list of
problems I could name with the "Speculator". But while I've often heard the
complaint that they trade points for ads, I've never seen anyone try and really
analyze it, except Jon Reuter.

Jon, a poster on WCWN who is apparently a statistician(he was a
frequent and respected poster who has no apparent connection to WS) posted this
a couple years ago:

"I've actually done a fairly technical (and therefore boring) review of WS
advertising and ratings (using WA ratings as a sort of control group) and found
only a slight bias at WS. For the majority of wines, the WS and WA ratings are
statistically indistiguishable. However, it does appear that WS is more likely
to retaste wines from advertisers and that these wines as a group benefit from
being retasted (to the tune of 2-3 points). To put that effect in context
though, less than 5 percent of wines are retasted so the overall average bias
is quite small.

Furthermore, conditional on price, production, and actual WS rating, there does
not appear to be any bias in who receives the various awards. So the earlier
post by a former WSer claiming that advertising and awards are unrelated
appears to be dead on."

There was a disclaimer that he only did the analysis for US wines for a 3 year
period I believe.

Of course, an extended analysis AFTER a winery scores high would probably show
a bigger correlation, because if WS scored one's wine a 94 wouldn't you think
of advertising there to remind readers once that issue has passed?

Now, the entry level restaurant awards are another thing. Pretty much any
restaurant that pays the $100 fee gets an award (602 out of 763 first time
entrants according to their own website).


Dale

Dale Williams
Drop "damnspam" to reply