Parker's "Magical 20": In Search of the Objective Criteria

Robert Parker's announcement of a Bordeaux "Magical 20" chateaux list -- whose wines will be tasted at The Wine Academy of Spain's Wine Future Hong Kong event on November 8 -- is significant in that it has the potential to: (i) cut through the confusion of the existing Bordeaux classification schemes to create a new Bordeaux "super-scheme" which is positioned just after the First Growths;  (ii) creates winners and losers among the chateaux; (iii) exposes a set of wines below the First Growths to the Asian market -- with attendant benefits to the chosen chateaux and price increases to your favorite wines if you are a consumer in other markets; and (iv) kick-start efforts to modernize the outdated 1855 classification scheme.  Given the potential impact, it is important that we understand Parker's criteria for inclusion on this list.  While we await Mr. Parker's revelations, I will attempt to use available data to try to tease out some of these criteria.

In his announcement, Parker stated, "I have chosen estates that produce wines of first growth quality, although technically not first growths.  Consequently, they are under-valued and very smart acquisitions."  Parker will be accompanied at the tasting by representatives from each estate.  The estates, their current classification, and  their respective AOC are shown in the table below.

My first impression in looking at the list is that there are some names that I would have liked to see on here that are not mentioned.  But then again, I am @wineOrl and he is Robert Parker.  He wins.  The list contains 11 chateaux from the 1855 Classification (5 second growths, 3 third growths, and 3 fifth growths), two from the St Emilion classification, and three from the Graves classification (Pomerol properties have never been classified.).  Every meaningful Bordeaux commune is represented in the list with 70% of the properties being left-bank-based.  Of the individual communes, Pauillac, Marguax, and Pomerol, with four chateaux each, are the most represented.  Nothing in this table appears to be list-composition drivers.

Not finding the criteria above, I turned to the release price of each chateau's primary wine.  In that the 2009 release of the wines will be tasted at Wine Future Hong Kong, I wanted to include that year's price in the analysis but I also wanted to look at least five years back so that meaningful analysis and comparisons could be made.  I was able to obtain time-series data on release prices from  genesiswines.com and those data are presented in Table 2 for the years 2004-2009. 

The series includes two exceptional years (2005 and 2009) and two less-than-stellar years (2004 and 2007).  The prices are stated in British Pounds.  In that 2009 was a stellar year, I wanted to compare it against another stellar year and 2005 fit that bill.  The second comparison was the pricing differential between 2008 and 2004.  In the cases where a price range is given, the low end of the range is used.  The comparisons are shown in the last two columns and indicate no correlation between inclusion on this list and 2004-2008 or 2005-2009 price growth so Mr. Parker could not have used these as criteria for composition of the list.  Onward then.

I looked further into areas such as ownership (a mix of owners to include industrialists, families, insurance companies, etc.), consultants employed (Michel Rolland, Stephen Derenoncourt, among others, but not in a suffocating way), and recent quality-focused investments (Cos, Pontet, and Angelus stand out) but no clear commonalities exist that one could point to and say "these are the objective standards for inclusion in the list."

If we have exhausted the possible objective selection criteria, then inclusion on the list must be subjective.  Could Parker scores have been the basis for inclusion? Let us take a look at that.  Table 3 shows Parker scores for each of the chateaux between the years 2004 and 2009 and a calculated average in the rightmost column. In looking closely at

the list, Pontet-Canet, Pape-Clement, and Smith Haut-Lafite show excellent scores in the down 2007 vintage.  On the other hand, Lynch-Bages and Pichon Lalande scored poorly in 2005, a year when even street-corner vendors were making excellent wines.  This is not the first time that Pichon Lalande has received low scores in a good vintage year.  You might remember a similar occurrence in 1990, another stellar Bordeaux vintage. And this begs the question as to why Ducru-Beaucaillou is not on this list.  It had an average score of 94.8 over the highlighted period; better than some of the wines included herein.

The average scores earned by the chateaux range from the low 90s to 96, a non-uniform distribution.  However, according to The Wine Advocate Rating System, a wine scoring between 90 and 95 is "an outstanding wine of exceptional complexity and character"; and all of these wines, save one, fall within that orbit.  But so do a lot of other Bordeaux wines that have more stuffing than Le Gay, for example.  Did they create a list and then stop calling when they got 20 people willing to travel to Hong Kong to participate in the event?  I hope not because this is potentially a guide to the Chinese as to what they should be buying next; and in so doing will create winners and losers among the second-tier Bordeaux players.  It will also have pricing implications for Bordeaux buyers like me.

With all of the foregoing, my biggest concern is the potential for creation of a new "super-second" classification scheme which is both subjective and single-sourced.  All previous Bordeaux classification schemes have come about as a result of group action.  In the case of the 1855 classification of the Medoc, Napolean requested a classification schema and it was developed by the brokers based on chateaux reputation and prices.  In the case of the St. Emilion classification, the Syndicat Viticole had begun considering the idea in the 1930s but held off until 1954 when the Institute Nacional des Appellations d'Origine (INAO) agreed to manage the process.  The Graves classification was implemented in 1953 by a jury appointed by INAO and was approved by the Ministry of Agriculture.  The scheme was updated in 1959 and has survived in that state to the current day.  All of these systems need updating to account for the passage of time (the St. Emilion classification is supposed to be updated every 10 years but the 2006 update had been mired in legal wrangling which has finally been resolved) but, given the implications, I would expect a studied, broad-based effort(s) to be the vehicle for such change given the number of stakeholders.  I fervently hope that Parker's "Magical 20" does not become a de facto Bordeaux classification scheme.