Report
WINETASTER ON 03/01/04 WITH 8 JUDGES AND 5 WINES BASED ON RANKS, IDENT=N
Copyright (c) 1995-2003 Richard E. Quandt, V. 1.65
FLIGHT 2:
Number of Judges = 8
Number of Wines = 5
Identification of the Wine: The judges' overall ranking:
Wine A is Ch. Leoville Las Cases 1964 ........ 1st place
Wine B is Ch. Latour 1961 ........ 5th place
Wine C is Ch. Ausone 1962 ........ 3rd place
Wine D is Ch. Latour 1962 ........ 4th place
Wine E is Ch. Palmer 1961 ........ 2nd place
The Judges's Rankings
Judge Wine -> A B C D E
Frank 3. 5. 1. 2. 4.
Grant 2. 4. 3. 5. 1.
John 2. 5. 1. 4. 3.
Mike 1. 5. 4. 3. 2.
Burt 1. 5. 4. 2. 3.
Bob 1. 5. 2. 3. 4.
Orley 1. 5. 4. 3. 2.
Dick 2. 5. 3. 4. 1.
Table of Votes Against
Wine -> A B C D E
Group Ranking -> 1 5 3 4 2
Votes Against -> 13 39 22 26 20
( 8 is the best possible, 40 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 = 0.5781
The probability that random chance could be responsible for this correlation
is quite small, 0.0010. 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
Orley 0.8208 0.1000
Mike 0.8208 0.1000
Dick 0.8208 -0.4000
Grant 0.7182 -0.3000
Burt 0.7000 0.3000
John 0.7000 -0.2000
Bob 0.7000 0.3000
Frank 0.2000 -0.2000
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 Ch. Leoville Las Cases 1964
---------------------------------------------------
2. ........ 2nd place Wine E is Ch. Palmer 1961
3. ........ 3rd place Wine C is Ch. Ausone 1962
4. ........ 4th place Wine D is Ch. Latour 1962
---------------------------------------------------
5. ........ 5th place Wine B is Ch. Latour 1961
We now test whether the ranksums AS A WHOLE provide a significant ordering.
The Friedman Chi-square value is 18.5000. The probability that this could
happen by chance is 0.0010
We now test whether the group ranking of wines is correlated with the
prices of the wines. The rank correlation between them is 0.0000. At the
10% level of significance this would have to exceed the critical value of
0.8000 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 1.00 for significance at the 0.05
level and must exceed 0.90 for significance at the 0.1 level
Frank Grant John
Frank 1.000 -0.200 0.700
Grant -0.200 1.000 0.500
John 0.700 0.500 1.000
Mike 0.100 0.600 0.400
Burt 0.300 0.200 0.300
Bob 0.700 0.200 0.800
Orley 0.100 0.600 0.400
Dick 0.100 0.900 0.600
Mike Burt Bob
Frank 0.100 0.300 0.700
Grant 0.600 0.200 0.200
John 0.400 0.300 0.800
Mike 1.000 0.900 0.600
Burt 0.900 1.000 0.700
Bob 0.600 0.700 1.000
Orley 1.000 0.900 0.600
Dick 0.800 0.500 0.400
Orley Dick
Frank 0.100 0.100
Grant 0.600 0.900
John 0.400 0.600
Mike 1.000 0.800
Burt 0.900 0.500
Bob 0.600 0.400
Orley 1.000 0.800
Dick 0.800 1.000
Pairwise correlations in descending order
1.000 Mike and Orley Significantly positive
0.900 Mike and Burt Significantly positive
0.900 Burt and Orley Significantly positive
0.900 Grant and Dick Significantly positive
0.800 Mike and Dick Not significant
0.800 John and Bob Not significant
0.800 Orley and Dick Not significant
0.700 Frank and John Not significant
0.700 Burt and Bob Not significant
0.700 Frank and Bob Not significant
0.600 John and Dick Not significant
0.600 Grant and Mike Not significant
0.600 Mike and Bob Not significant
0.600 Bob and Orley Not significant
0.600 Grant and Orley Not significant
0.500 Grant and John Not significant
0.500 Burt and Dick Not significant
0.400 John and Mike Not significant
0.400 John and Orley Not significant
0.400 Bob and Dick Not significant
0.300 John and Burt Not significant
0.300 Frank and Burt Not significant
0.200 Grant and Bob Not significant
0.200 Grant and Burt Not significant
0.100 Frank and Mike Not significant
0.100 Frank and Dick Not significant
0.100 Frank and Orley Not significant
-0.200 Frank and Grant Not significant
COMMENT:
Despite the claim that these are ancient wines, it is a tragedy that
people do not keep wines as long as our host has kept these. One of the
most pleasant tastings we have ever had, apart from the 1961 Latour, which
was clearly a flawed (but neither oxidized nor corked) wine, the wines
are terrific drinking right now and were worth every penny they cost. Our rank-
ing of the wines was probably the most highly agreed upon in our history.
If you own some 1964 Leoville Las Cases, our top rated wine, you should
thank god and your good fortune. Several tasters agree that the Leoville Las Cases
outranks several first growths. Several of us believe that Leoville Las Cases
is the leading candidate for promotion to first growth status. Since
Ch. Gruaud Larose has been another contender for exalted status in our
tastings, we plan to organize a taste-off between Leoville Las Cases
and Gruaud Larose.
Return to previous page