WINETASTER ON 02/02/04 WITH 8 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2004 Richard E. Quandt, V. 1.65 Number of Judges = 8 Number of Wines = 8
Identification of the Wine: The judges' overall ranking:
Wine A is Ch. Gruaud Larose '97 ........ 2nd place Wine B is Ch. Margaux '97 ........ 1st place Wine C is Ch. Lynch Bages '97 ........ 4th place Wine D is Ch. La-Fleur Petrus '97 ........ 7th place Wine E is Carruades de Ch. Lafite '97 ........ 6th place Wine F is Ch. Ducru Beaucaillou '97 ........ 3rd place Wine G is Ch. Pichon Lalande '97 ........ 5th place Wine H is Ch. Prieure Lichine '97 ........ 8th place
The Judges's Rankings
Judge Wine -> A B C D E F G H Frank 1. 4. 5. 3. 6. 8. 2. 7. Jim 7. 1. 6. 8. 5. 2. 4. 3. Orley 3. 1. 2. 8. 4. 5. 7. 6. Burt 7. 2. 3. 6. 4. 5. 1. 8. Alan 6. 3. 5. 8. 2. 1. 4. 7. Norton 1. 8. 3. 2. 4. 7. 6. 5. Bob 4. 3. 6. 2. 7. 1. 5. 8. Dick 1. 4. 3. 8. 7. 2. 5. 6.
Table of Votes Against Wine -> A B C D E F G H
Group Ranking -> 2 1 4 7 6 3 5 8 Votes Against -> 30 26 33 45 39 31 34 50
( 8 is the best possible, 64 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 = 0.1711
The probability that random chance could be responsible for this correlation is rather large, 0.2134. 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 Correlation Price Dick 0.6587 0.1190 Orley 0.5270 0.0952 Alan 0.2771 -0.2857 Burt 0.2619 0.3095 Bob 0.1429 -0.6190 Jim 0.0714 -0.0476 Frank 0.0000 0.3810 Norton -0.5952 0.0714
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 B is Ch. Margaux '97 2. ........ 2nd place Wine A is Ch. Gruaud Larose '97 3. ........ 3rd place Wine F is Ch. Ducru Beaucaillou '97 4. ........ 4th place Wine C is Ch. Lynch Bages '97 5. ........ 5th place Wine G is Ch. Pichon Lalande '97 6. ........ 6th place Wine E is Carruades de Ch. Lafite '97 7. ........ 7th place Wine D is Ch. La-Fleur Petrus '97 --------------------------------------------------- 8. ........ 8th place Wine H is Ch. Prieure Lichine '97 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 9.5833. The probability that this could happen by chance is 0.2134
We now test whether the group ranking of wines is correlated with the prices of the wines. The rank correlation between them is -0.0714. At the 10% level of significance this would have to exceed the critical value of 0.5240 to be significant.
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 Frank Jim Orley Frank 1.000 -0.524 -0.024 Jim -0.524 1.000 0.286 Orley -0.024 0.286 1.000 Burt 0.190 0.310 0.262 Alan -0.429 0.619 0.381 Norton 0.452 -0.952 -0.143 Bob 0.143 0.048 -0.071 Dick 0.095 0.190 0.571 Burt Alan Norton Frank 0.190 -0.429 0.452 Jim 0.310 0.619 -0.952 Orley 0.262 0.381 -0.143 Burt 1.000 0.524 -0.500 Alan 0.524 1.000 -0.643 Norton -0.500 -0.643 1.000 Bob 0.095 0.190 -0.167 Dick 0.024 0.310 -0.048 Bob Dick Frank 0.143 0.095 Jim 0.048 0.190 Orley -0.071 0.571 Burt 0.095 0.024 Alan 0.190 0.310 Norton -0.167 -0.048 Bob 1.000 0.286 Dick 0.286 1.000 Pairwise correlations in descending order 0.619 Jim and Alan Not significant 0.571 Orley and Dick Not significant 0.524 Burt and Alan Not significant 0.452 Frank and Norton Not significant 0.381 Orley and Alan Not significant 0.310 Jim and Burt Not significant 0.310 Alan and Dick Not significant 0.286 Bob and Dick Not significant 0.286 Jim and Orley Not significant 0.262 Orley and Burt Not significant 0.190 Jim and Dick Not significant 0.190 Frank and Burt Not significant 0.190 Alan and Bob Not significant 0.143 Frank and Bob Not significant 0.095 Frank and Dick Not significant 0.095 Burt and Bob Not significant 0.048 Jim and Bob Not significant 0.024 Burt and Dick Not significant -0.024 Frank and Orley Not significant -0.048 Norton and Dick Not significant -0.071 Orley and Bob Not significant -0.143 Orley and Norton Not significant -0.167 Norton and Bob Not significant -0.429 Frank and Alan Not significant -0.500 Burt and Norton Not significant -0.524 Frank and Jim Not significant -0.643 Alan and Norton Not significant -0.952 Jim and Norton Significantly negative
Comments:
The relatively poor showing of the Ch. Prieure Lichine confirmed the
view we have had of this wine in previous tastings. But all the wines
were very drinkable, in fact
rather extraordinary wines. All were very nice drinking now but many, Margaux and Ducru for sure,
for considerable keeping. They may be like the '53s, drinking well from start to finish.
Overpriced at the outset, these wines can now be real bargains. Prieure, the '97 vintage from before Sasha Lichine lost it finally,
was as bad as many of us remembered the wines of this chateau from vintages of the past. There was also
a feeling that the Gruaud Larose tasted well and was a
reasonable bargain (under $40) and that all the wines would probably
improve.