WINETASTER ON 04/07/14 WITH 8 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2014 Richard E. Quandt, V. 1.65
FLIGHT 1: Number of Judges = 8 Number of Wines = 8
Identification of the Wine: The judges' overall ranking:
Wine A is Kistler Cuvée Catherine 2001 ........ 4th place Wine B is Kistler Cuvée Catherine 2007 ........ 5th place Wine C is Kistler Vineyard 2000 ........ 7th place Wine D is Kistler Cuvée Catherine 2006 ........ 2nd place Wine E is Kistler Vineyard 1999 ........ 8th place Wine F is Kistler Cuvée Catherine 2004 ........ 3rd place Wine G is Kistler Cuvée Catherine 2008 ........ 6th place Wine H is Kistler Cuvée Catherine 2005 ........ 1st place
The Judges's Rankings
Judge Wine -> A B C D E F G H Frank 7. 4. 6. 3. 8. 5. 2. 1. Orley 1. 6. 7. 4. 8. 5. 3. 2. Ed 3. 5. 4. 2. 6. 1. 8. 7. Mike 6. 5. 7. 1. 8. 4. 2. 3. Burt 5. 6. 8. 3. 7. 1. 4. 2. Bob 3. 5. 4. 7. 8. 2. 6. 1. Zaki 3. 4. 1. 2. 8. 6. 7. 5. Dick 5. 2. 3. 4. 8. 7. 6. 1.
Table of Votes Against Wine -> A B C D E F G H
Group Ranking -> 4 5 7 2 8 3 6 1 Votes Against -> 33 37 40 26 61 31 38 22
( 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 = 0.3631
The probability that random chance could be responsible for this correlation is quite small, 0.0049. 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 Burt 0.6429 Mike 0.5509 Frank 0.5030 Orley 0.4524 Bob 0.4192 Dick 0.3571 Zaki 0.1667 Ed 0.0732
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 Kistler Cuvée Catherine 2005 --------------------------------------------------- 2. ........ 2nd place Wine D is Kistler Cuvée Catherine 2006 3. ........ 3rd place Wine F is Kistler Cuvée Catherine 2004 4. ........ 4th place Wine A is Kistler Cuvée Catherine 2001 5. ........ 5th place Wine B is Kistler Cuvée Catherine 2007 6. ........ 6th place Wine G is Kistler Cuvée Catherine 2008 7. ........ 7th place Wine C is Kistler Vineyard 2000 --------------------------------------------------- 8. ........ 8th place Wine E is Kistler Vineyard 1999 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 20.3333. The probability that this could happen by chance is 0.0049 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 Orley Ed Frank 1.000 0.476 -0.357 Orley 0.476 1.000 -0.048 Ed -0.357 -0.048 1.000 Mike 0.857 0.548 0.000 Burt 0.595 0.571 0.238 Bob 0.262 0.500 0.167 Zaki 0.000 0.119 0.476 Dick 0.548 0.262 -0.167 Mike Burt Bob Frank 0.857 0.595 0.262 Orley 0.548 0.571 0.500 Ed 0.000 0.238 0.167 Mike 1.000 0.738 0.071 Burt 0.738 1.000 0.476 Bob 0.071 0.476 1.000 Zaki 0.048 -0.214 0.190 Dick 0.238 0.000 0.429 Zaki Dick Frank 0.000 0.548 Orley 0.119 0.262 Ed 0.476 -0.167 Mike 0.048 0.238 Burt -0.214 0.000 Bob 0.190 0.429 Zaki 1.000 0.595 Dick 0.595 1.000 Pairwise correlations in descending order 0.857 Frank and Mike Significantly positive 0.738 Mike and Burt Significantly positive 0.595 Frank and Burt Not significant 0.595 Zaki and Dick Not significant 0.571 Orley and Burt Not significant 0.548 Orley and Mike Not significant 0.548 Frank and Dick Not significant 0.500 Orley and Bob Not significant 0.476 Burt and Bob Not significant 0.476 Frank and Orley Not significant 0.476 Ed and Zaki Not significant 0.429 Bob and Dick Not significant 0.262 Frank and Bob Not significant 0.262 Orley and Dick Not significant 0.238 Ed and Burt Not significant 0.238 Mike and Dick Not significant 0.190 Bob and Zaki Not significant 0.167 Ed and Bob Not significant 0.119 Orley and Zaki Not significant 0.071 Mike and Bob Not significant 0.048 Mike and Zaki Not significant 0.000 Burt and Dick Not significant 0.000 Ed and Mike Not significant 0.000 Frank and Zaki Not significant -0.048 Orley and Ed Not significant -0.167 Ed and Dick Not significant -0.214 Burt and Zaki Not significant -0.357 Frank and Ed Not significant
COMMENT: This tasting represents the inclusion of the very rare cuvée Catherine in a tasting of Kistler Pinot Noirs. We wondered how this wine would taste among its peers of multiple vintages. Our conclusion was this is a great wine that is enjoyable back through 1999, 15 years. The hypothesis to be investigated, as stated by our host, was whether Kistler Pinot Noirs age as well as old Burgundies. Out tentative conclusion, having disregarded the data from a flawed bottle and confirmed by a tasting of the same vintage from another bottle was that these wines, while being fabulous, appeared to peak in the 7-9 year window. We had an incredibly generous host who, when faced with a 1999 which was flawed, replaced it with a fresh bottle. In addition, our generous host provided a fresh bottle of the 2009 vintage as well. Thus, we were able to taste the 2009 and the 1999, each from fresh bottles, to assess the effects of age on the wines. The 1999 Kistler replacement was a Sonoma coast versus Russian river, which may or may not be signifcant. Unlike many other tastings, the wines did not change very much in taste or bouquet after they were opened. This tasting was a truly memorable experience. It is worth mentioning that since April 2004, one or two Kistler Pinot Noirs (for a total of six) in four different tastings (not counting the present one), matched in each case by other American Pinot Noirs and Burgundy Pinot Noirs (in one tasting a New Zealand Pinot was also included). The six Kistlers ranked in their respective tastings first in three tastings (once tied for first place), second once, and fifth twice, a remarkable overall record. The Kendall W coefficient is 0.3631, and the probability that this could have occurred by chance is a mere 0.0049, a highly significant result. Since there was almost complete agreement that wine E was flawed, one may wonder to what extent overall result is driven by the agreement on wine E. To account for that, we reranked the results as if wine E had not existed at all. Note that we are not proposing to make fiduciary statements about the results of this reranking, since using (part of) the data for such an exercise would not be valid; accordingly we are solely interested inthe mechanical alteration of the results. The upshot of this exercise is that the W coefficient declines to 0.1440 and the probability that this could have occurred by chance increases to 0.3292, from which we would have inferred that no significant agreement existed.
Return to previous page