WINETASTER ON 11/08/10 WITH 8 JUDGES AND 9 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2010 Richard E. Quandt, V. 1.65

FLIGHT 1: Number of Judges = 8 Number of Wines = 9

Identification of the Wine: The judges' overall ranking:

Wine A is Chateau Musar 1979 ........ 1st place Wine B is 1987 ........ 3rd place Wine C is 1980 ........ 5th place Wine D is 1986 ........ 4th place Wine E is 1981 ........ 8th place Wine F is 1993 tied for 6th place Wine G is 1991 tied for 6th place Wine H is 1998 ........ 2nd place Wine I is 1996 ........ 9th place

The Judges's Rankings

Judge Wine -> A B C D E F G H I Bob 6. 7. 5. 3. 9. 1. 2. 4. 8. Ed 6. 2. 3. 8. 5. 9. 7. 1. 4. Mike 1. 8. 7. 5. 4. 9. 2. 3. 6. Burt 6. 3. 7. 4. 5. 9. 8. 1. 2. Zaki 3. 5. 1. 9. 4. 2. 6. 7. 8. John 3. 6. 7. 4. 8. 1. 5. 2. 9. Orley 1. 3. 4. 5. 2. 7. 6. 8. 9. Dick 3. 2. 6. 1. 8. 5. 7. 4. 9.

Table of Votes Against Wine -> A B C D E F G H I

Group Ranking -> 1 3 5 4 8 6 6 2 9 Votes Against -> 29 36 40 39 45 43 43 30 55

( 8 is the best possible, 72 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 = 0.1318

The probability that random chance could be responsible for this correlation is rather large, 0.3923. 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 0.3798 0.5439 John 0.3333 -0.0084 Orley 0.2259 0.7029 Mike 0.0760 -0.1757 Burt -0.0168 -0.2929 Ed -0.0753 -0.1004 Bob -0.1674 0.0084 Zaki -0.2008 0.4017

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 A is Chateau Musar 1979 2. ........ 2nd place Wine H is 1998 3. ........ 3rd place Wine B is 1987 4. ........ 4th place Wine D is 1986 5. ........ 5th place Wine C is 1980 6. tied for 6th place Wine G is 1991 7. tied for 6th place Wine F is 1993 8. ........ 8th place Wine E is 1981 --------------------------------------------------- 9. ........ 9th place Wine I is 1996 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 8.4333. The probability that this could happen by chance is 0.3923

We now test whether the group ranking of wines is correlated with the prices of the wines. The rank correlation between them is 0.3950. At the 10% level of significance this would have to exceed the critical value of 0.4830 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 0.70 for significance at the 0.05 level and must exceed 0.60 for significance at the 0.1 level Bob Ed Mike Bob 1.000 -0.533 -0.067 Ed -0.533 1.000 0.000 Mike -0.067 0.000 1.000 Burt -0.517 0.683 0.100 Zaki 0.033 -0.050 -0.250 John 0.750 -0.333 0.100 Orley -0.367 -0.033 0.233 Dick 0.317 -0.050 -0.033 Burt Zaki John Bob -0.517 0.033 0.750 Ed 0.683 -0.050 -0.333 Mike 0.100 -0.250 0.100 Burt 1.000 -0.667 -0.250 Zaki -0.667 1.000 0.117 John -0.250 0.117 1.000 Orley -0.250 0.467 -0.100 Dick 0.133 -0.117 0.583 Orley Dick Bob -0.367 0.317 Ed -0.033 -0.050 Mike 0.233 -0.033 Burt -0.250 0.133 Zaki 0.467 -0.117 John -0.100 0.583 Orley 1.000 0.317 Dick 0.317 1.000 Pairwise correlations in descending order 0.750 Bob and John Significantly positive 0.683 Ed and Burt Significantly positive 0.583 John and Dick Not significant 0.467 Zaki and Orley Not significant 0.317 Bob and Dick Not significant 0.317 Orley and Dick Not significant 0.233 Mike and Orley Not significant 0.133 Burt and Dick Not significant 0.117 Zaki and John Not significant 0.100 Mike and John Not significant 0.100 Mike and Burt Not significant 0.033 Bob and Zaki Not significant 0.000 Ed and Mike Not significant -0.033 Mike and Dick Not significant -0.033 Ed and Orley Not significant -0.050 Ed and Dick Not significant -0.050 Ed and Zaki Not significant -0.067 Bob and Mike Not significant -0.100 John and Orley Not significant -0.117 Zaki and Dick Not significant -0.250 Mike and Zaki Not significant -0.250 Burt and Orley Not significant -0.250 Burt and John Not significant -0.333 Ed and John Not significant -0.367 Bob and Orley Not significant -0.517 Bob and Burt Not significant -0.533 Bob and Ed Not significant -0.667 Burt and Zaki Significantly negative

COMMENT: It is a pleasure to drink a vertical from a much less covered region such as Lebanon. (See http://en.wikipedia.org/wiki/Chateau_Musar for a detailed description.) And the wines did not disappoint and confirmed many of the groups intermittent exposure to the wine. The style overall is characterized by a rich sweet style that is pleasurable to drink. The wines do obviously very well with bottle age, the 1979 being the oldest and the highest rated and the 1998 is the second highest. The balance and acidity was good in all the wines. The Internet literature suggests that the 1987 (our rank 3) is flawed which we did not find to be so. One taster suggested that the wines are not "sweet," but "rich" and portlike. One taster opined that the 1979 Musar reminded him of a Ducru Beaucaillou. What was very interesting is that the 1993 generated a bipolar distribu- tion, with three taster ranked it No. 1 or 2 and three ranked it No. 9. The overall agreement in the group was not overwhelming, but even so, the 1996 showed up as significantly "bad."

Return to previous page