WINETASTER ON 09/08/03 WITH 7 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2003 Richard E. Quandt
FLIGHT 1: Number of Judges = 7 Number of Wines = 8
Identification of the Wine: The judges' overall ranking:
Wine A is Ch. Latour 1983 ........ 5th place Wine B is Ch. Beychevelle 1983 tied for 7th place Wine C is Ch. Rausan Segla 1983 ........ 2nd place Wine D is Ch. Calons Segur 1983 ........ 3rd place Wine E is Ch. Lafite Rothschild 1983 ........ 6th place Wine F is Ch. Palmer 1983 ........ 4th place Wine G is Ch. Montrose 1985 tied for 7th place Wine H is Ch. Gruaud Larose 1983 ........ 1st place
The Judges's Rankings
Judge Wine -> A B C D E F G H Alexa 4. 3. 1. 5. 7. 6. 8. 2. Bob 2. 6. 5. 4. 7. 8. 3. 1. Orley 4. 3. 2. 5. 1. 7. 8. 6. Frank 2. 7. 5. 4. 8. 1. 3. 6. John 8. 4. 6. 2. 5. 1. 7. 3. Ed 3. 8. 1. 5. 4. 2. 7. 6. Dick 7. 8. 6. 2. 5. 4. 3. 1.
Table of Votes Against Wine -> A B C D E F G H
Group Ranking -> 5 7 2 3 6 4 7 1 Votes Against -> 30 39 26 27 37 29 39 25
( 7 is the best possible, 56 is the worst)
Here is a measure of the correlation in the preferences of the judges which ranges between 1.0 (perfect correlation) and 0.0 (no correlation):
W = 0.1186
The probability that random chance could be responsible for this correlation is rather large, 0.5622. Most analysts would say that unless this probability is less than 0.1, the judges' preferences are not strongly related. We now analyze how each taster's preferences are correlated with the group preference. A correlation of 1.0 means that the taster's preferences are a perfect predictor of the group's preferences. A 0.0 means no correlation, while a -1.0 means that the taster has the reverse ranking of the group. This is measured by the correlation R.
Correlation Between the Ranks of Each Person With the Average Ranking of Others
Name of Person Correlation R Alexa 0.2156 Ed 0.2036 Frank 0.0120 Dick -0.1198 John -0.1325 Bob -0.2036 Orley -0.4820
The wines were preferred by the judges in the following order. When the preferences of the judges are strong enough to permit meaningful differentiation among the wines, they are separated by -------------------- and are judged to be significantly different.
1. ........ 1st place Wine H is Ch. Gruaud Larose 1983 2. ........ 2nd place Wine C is Ch. Rausan Segal 1983 3. ........ 3rd place Wine D is Ch. Calons Segur 1983 4. ........ 4th place Wine F is Ch. Palmer 1983 5. ........ 5th place Wine A is Ch. Latour 1983 6. ........ 6th place Wine E is Ch. Lafite Rothschild 1983 7. tied for 7th place Wine G is Ch. Montrose 1983 8. tied for 7th place Wine B is Ch. Beychevelle 1983 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 5.8095. The probability that this could happen by chance is 0.5622 We now undertake a more detailed examination of the pair-wise rank correla- tions that exist between pairs of judges. First, we present a table in which you can find the correlation for any pair of judges, by finding one of the names in the left hand margin and the other name on top of a column. A second table arranges these correlations in descending order and marks which is significantly positive significantly negative, or not significant. This may allow you to find clusters of judges whose rankings were particularly similar or particularly dissimilar. Pairwise Rank Correlations Correlations must exceed in absolute value 0.74 for significance at the 0.05 level and must exceed 0.64 for significance at the 0.1 level Alexa Bob Orley Alexa 1.000 0.286 0.357 Bob 0.286 1.000 -0.310 Orley 0.357 -0.310 1.000 Frank -0.238 0.095 -0.667 John 0.024 -0.405 -0.238 Ed 0.190 -0.286 0.262 Dick -0.214 0.357 -0.595 Frank John Ed Alexa -0.238 0.024 0.190 Bob 0.095 -0.405 -0.286 Orley -0.667 -0.238 0.262 Frank 1.000 0.000 0.381 John 0.000 1.000 -0.024 Ed 0.381 -0.024 1.000 Dick 0.119 0.452 -0.143 Dick Alexa -0.214 Bob 0.357 Orley -0.595 Frank 0.119 John 0.452 Ed -0.143 Dick 1.000 Pairwise correlations in descending order 0.452 John and Dick Not significant 0.381 Frank and Ed Not significant 0.357 Bob and Dick Not significant 0.357 Alexa and Orley Not significant 0.286 Alexa and Bob Not significant 0.262 Orley and Ed Not significant 0.190 Alexa and Ed Not significant 0.119 Frank and Dick Not significant 0.095 Bob and Frank Not significant 0.024 Alexa and John Not significant 0.000 Frank and John Not significant -0.024 John and Ed Not significant -0.143 Ed and Dick Not significant -0.214 Alexa and Dick Not significant -0.238 Alexa and Frank Not significant -0.238 Orley and John Not significant -0.286 Bob and Ed Not significant -0.310 Bob and Orley Not significant -0.405 Bob and John Not significant -0.595 Orley and Dick Not significant -0.667 Orley and Frank Significantly negative
COMMENT: Appended are comments by Orley Ashenfelter, written the night before the tasting. This wine tasting created an extraordinary natural experiment. Our host, whose cellar contained all these wines since first purchased, mistakenly put a 1985 Montrose into a tasting of 1983s.Thus, unbenownst to any taster present, one of the wines was from the 1985 vintage, while all 7 others were 1983s. Although rated at the bottom, tied with 1983 Beychevelle,the wine was not signficantly lower quality than the others. Gruaud Larose continues its winning tradition in our tastings, particularly price adjusted. The majority of the group was surprised by the extent to which thse wines have reached the peak of their maturity, with the possible exception of Lafite, which remains tannic and closed. The group felt that the wines were not terribly distinctive from the standpoint of bouquet but were remarkably soft and easy to drink. This is borne out by the quantitative ratings, which were not significantly different.
1983 Bordeaux viewed in 2003 Orley Ashenfelter The occasion of this tasting through a number of the 1983 red Bordeaux wines seems a good time to reminisce. After all, the wines are now 20 years old and many of us bought them 15 or more years ago, and some of us still have them in the cellar. And having tasted through a few of these wines in the last month with at least one member of this group, made me think it was worth sharing what we found out. The Weather and the Wines: As some people perhaps know, the quality of a Bordeaux vintage is a result of the quality and character of the grapes, and these are a result of the weather during the growing season. 1983, by the standards of the times, was a very warm growing season, much like 1982. Oddly, in the context of the subsequent two decades, 1983 would be considered warm, but not especially warm—in the context of the previous two decades it would have been considered extremely warm, the warmest since 1959. Is this global warming? I don't know, but if so, it is surely a good thing for red Bordeaux wines—at least until we hit something catastrophic. Rainfall at harvest was also much like 1982 and, in short, the wines should not be much different from the 82s. I think many people who taste the wines might agree with this, although, as always the wines vary somewhat from chateaux to chateaux. (If you do not remember these facts, dig out your old issues of Liquid Assets!) The market certainly does not agree with this assessment, but more about that below. 1983s then, are good wines, like the 1988s, but riper, perhaps less concentrated. They are much riper than the 1985s, but somewhat less concentrated. It is interesting to compare a 1983 and 1985 from the same chateaux, which I have not done as often as I wish—I usually always prefer the ripe (some might say California!) character of the 1983s, but I know many people who I admire greatly (like the Indiana wine collector William Dick, and the famous auctioneer Michael Broadbent) who would always take the 1985. To each his own! The 1983s are not up to the standard of 1989 or 1990, but none of the vintages prior to 1989 (apart from 1961) are either. Tasting the Wines Today: I think most of these wines are now ready to drink; a few may be too old, depending on your preferences. Having tasted Haut Brion several times in the last month, I can say, for me, this wine is now terrific—just what you wish you consumed in a 3-star restaurant, and our German wine friend, Karl Storchmann, thought it just about all you could ask for in a red wine, that is, nearly perfect. Lesser wines, like Grand Puy Lacoste are lighter, but great drinking. Wines like Lanessan, a simple wine meant to last, are also excellent. A few wines, to my mind, have odd characters—La Lagune seemed, recently, to have an odd nose, not what I expected. And Le Gay is tough and lacks the soft character that many would prefer in mature red Bordeaux—though these wines certainly have their places, often near a grill! The Market: It has always been annoying for younger wine buyers to hear stories about what buyers from older cohorts paid for their wines (I was myself one of these younger buyers, so I know how they feel!), but I think this is worth discussing because of current Bordeaux pricing. Frankly, the prices of red Bordeaux wines are so high that I fear that the vast majority of younger buyers can no longer regularly afford to taste through their wines and learn about them the old fashioned way. Most of us purchased wines like the 1983 Haut Brion for about $35 per bottle—Grand Puy Lacoste was about $10, and wines like Meyney were about $8. These prices no doubt seem shocking by today's standards, especially when young wines from the 2002 vintage, certainly not destined to go down in any record books in view of the rain at harvest, cost perhaps 4 or 5 times more. Of course, the 1983s were not good investments compared to the stock market—making the 2002s seem even more absurd to an investor today. The interesting thing about the 1983s is that they are really wines for drinking—at current auction prices they are terrific buys. The problem? Hardly any are for sale! You can buy all the 1982s you want at 3 times the prices of the 1983s, and as many of the California cults as you want, at even higher prices—but 1983s, they are hard to find! I think most of us who own and drink these wines know what is happening—most of us bought 1982s and 1983s. We sell our 1982s, indeed virtually all the mature wine on offer at the auctions is from this vintage, and we drink our 1983s! My Advice: The only vintage like 1983 that I know of today for current buyers to consider is 1997: fine wines, the 1997s are being drunk too soon, and they are selling at ridiculously low prices, although they seem to be becoming increasingly difficult to find. The 1997s need to be kept—let's just hope we are around to taste them in 2017!
Return to previous page