Wine Reviews: Almost Useless in Its Current Form

The image above is I believe, an example of a better way to extrapolate useful information about the quality of wine and I would be very pleased to see reviewers use it. Forget all of the pretty words that critics struggle with for the sake of keeping their journalistic prose fresh (because our language is finite, this is impossible and ends up being just embarrassing and a waste of time for serious readers). How useful is the current system of wine reviews? How much real information can we gather aboutwine being reviewed? Do we taste the same things reviewers taste? Will thetasting notes change over time, making a reviewer’s notes obsolete? Whatinformation would be best to use while assessing wine?
The only thing actually useful about a wine review by theexperts is their score. But the score itself does not tell us anything aboutthe wine – it isjust a grade, which may be useful if you have aligned your palate with aparticular reviewer’s taste. From there, one can adjust their expectations. Me,for example: I have learned that I can take a Robert Parker score and safelydowngrade it by 4 or 5 points and it will line up fairly close to my opinion. Ican take Stephen Tanzer scores and add 2 to 3 points. James Suckling atSpectator is nearly always right on for me while James Laube at Spectator is awild-card and is hit-and-miss for me. And for me Jay Miller has zerocredibility as a wine critic.
Obviously scoring wine is very subjective and personalpreferences are very much a part of the human process of forming an opinion. Ibelieve however that we may be able to get a bit more objective information bychanging the way wine is scored. Here are a few problems with the currentsystem:
First,there is very little information in a wine review that is actually somewhatobjective or useful to a wine aficionado. If a professional reviewer sat nextto you or me while reviewing a wine and we compared notes, I may pick up someof the flavor or aromatic notes I missed if he mentions them – just because ofsuggestion. Conversely, he may pick up notes I observed just because I mentionthem. The point is if you read a review before you taste the wine, chances arepretty good that you’ll pick out some of the elements mentioned in the review,which is due to subconscious power of suggestion and its pretty powerful withwine notes. Because our palates are not the same, it is very unlikely that we’dactually have the same wine experience without the power of suggestion and eventhen, we may not experience the same aromas and tastes.
Secondly,all of the things the reviewer tastes and smells during the moment of hisreview can change substantially with time – even within a short amount of time.Chances are very slim that even the same reviewer would pick up the same notesafter a few months, if the wine were blind-tasted. There are many possiblevariations that could change a wine’s descriptors such as storage conditions,the amount of time the bottle has been still (in other words it didn’t just getdelivered by FedEx an hour ago), serving temperature, how much wine thereviewer has already tasted that day, and many other significant variables.
Theend result is that the description of a wine becomes useless because there isno way to assess the condition of the wine or the reviewer. Therefore tastingnotes are generally just indulgent journalistic explorations to find words andphrases that have not been used as descriptors; I don’t believe there’s muchleft that hasn’t been written before and so we read the same things over andover again that are characteristics of the wine being reviewed. This finiteamount of text simply has little practical relevance or use to the reader.
Thirdis the wasted ink that describes the color of the wine; many reviewers beginwith a description of the color. Who cares? Red wines all have various hues ofred and purple, while whites may be described with straw or golden hues. Thiswill never change unless the wine is flawed. Now observing a flaw is useful information, for example:the wine was amber on its edges, or brown, or the color of red brick. These areall giveaways that the wine is probably well past it prime and therefore ofquestionable value and quality. But why not just say the wine has a flaw ratherthan describe its color? Besides it is very rare a reviewer writes aboutseriously flawed wine.
To understand how little real information there is in atypical wine review just analyze the following reviews and scores from theforemost and highest regarded wine critics. There is a Piedmont (Italy) andBordeaux - also note the differences and similarities between scores:
KEY:
Useful terminology
Similar or common observations between reviewers
Substantial difference between reviewers
Barolo,Prapò 1996 Ceretto - Bricco Rocche
"...rich,with lots of extract,tannin, and body...moderatelyintense nose of scorchedearth, dried herbs, and sweet black fruits...anexciting level of fruitextract, superb purity..." (92-94 points) Robert Parker's Wine Advocate#124, August 1999 (Robert Parker, 08/01/1999)
Rating: 92
"Complex aromasof tar, blackberryand sage. Full-bodied, with fine tannins and a long, long finish. Balanced and beautiful.--1996 Piedmont retrospective. Best after2007." Wine Spectator (Wine Spectator, 01/01/2007)
Rating: 91
"Mediumred, with an orange rim. Ripebut vibrant aromas of cherry,raspberry, chocolate liqueur, mocha and menthol. Ripe and vinous, with perfumed inner-mouth flavors of tobacco anddried rose. Already quite expressive,especially in light of the vintage. This will be the first to give pleasureamong the Barolo crus in '96 but has the tannic structure to develop in bottle for eight toten years." (90 points) Stephen Tanzer's International Wine Cellar,November/December 2000 (Steven Tanzer, 11/01/2000)
Rating: 90
ChâteauLéoville-Las-Cases 1996
"...aspectacular nose of cassis,cherry liqueur, pain grillé, and minerals...powerful and rich on the attack...a remarkable,seamless, palate-staining, and extraordinarily elegant wine..." (98points) Robert Parker's Wine Advocate #122, April 1999 (Robert Parker,04/01/1999)
Rating:98
"...vibrant,very youthful aromas ofcassis, violet and bitter chocolate. Dense and powerful, with great clarity of flavor thanks to aterrific spine of acidity...Finishes very longand gripping, with a note of bitter chocolate..." (96(+?) points)Stephen Tanzer's International Wine Cellar, July/August 2002 (Steven Tanzer,07/01/2002)
Rating: 96
"Incredible nose of blackberry, mineral,cedar and currant [cassis and currant are the same DB].Full-bodied, withsilky and refined tanninsand a mediumcaressing finish.It's a beautiful wine that begs to be drunk now but will age and improve for along time.--'95/'96 Bordeaux retrospective. Drink now." Wine Spectator(Wine Spectator, 01/01/2007)
Rating:92
Notsurprisingly, there’s a lot of journalistic fluff but not a lot of reallyuseful information. The pain grillé and cassis type descriptors do not meananything to me as these words are used very often in red wine but what I reallywant to know about is the wine’s balance, intensity of flavor and aroma, itscomplexity, how it feels in the mouth (the body of the wine) and the wine’sattack, midpalate and finish. If I have this information, I can gauge with muchgreater accuracy, to what extent I will like a wine or not.
Iwould be pleased to email you a working version of this wine scoring algorithmto experiment with or use as you like. I am also open to your feedback andinput about ways to improve the usefulness of it. Please email me at david@classof1855.com with your requestor comments.
NOTE: Jancis Robinson is a very highlyregarded British wine critic and one of the few women to hold the highlyesteemed Master of Wine designation. She has her own 20 point scale, with ascore of 20 being the very best. Because of the difference in descriptors, Idid not do an in-depth comparison with her and the other reviewers.
David Boyer





I employ a complementary approach, what I'd term as a structural description:
http://thecabfrancofiles.blogspot.com/2009/12/graphical-representation.html
I went with the radar plot since it avoids the whole numerical precision issue. The goal is to characterize with broad categories instead of to quantify or list decriptors. I do think categories such as fruit or earth can be identified, whereas cassis vs. blackberry or bell pepper vs. capsicum is a senseless distinction. There is a definite weakness, though, in that purely structural descriptions/graphs don't broach issues like balance, complexity, attack, mid-palate and finish. Complexity and balance, however, are suggested by the shape of the contour.
There are drawbacks to quantifying finish, complexity and balance as well. These are incredibly personal impressions, so other than for one's own interest, I don't think they translate well. As I understand it, your system is designed to assign a number to quality, thus the subjectivity is not a major problem.
Anyway, it seems there is no perfect approach! Indeed the standard tasting note is less perfect than most.
Reply to this
Hi Greg,
Reply to this
David,
As you have said to Greg, "Indeed there is no perfect way to score or assess the quality of wine but I am delighted to know that you and others are considering a better solution".In South
Africa I have designed a system www.sawineindex.com which have just been launched because of similar problems with subjectiveness. Consistency in tasting results of various different wine panels made more sence to me. Then you must also realise that my own take is simple and personal. Either like it and ask yourself if it makes you think, otherwise don't drink it.
Izak Smit
Reply to this
Dear Izak,
Reply to this