RE: Wall Street Journal Article: Wine Critics Are Not Consistent in Ratings

View related threads: (in this forum | in all forums)

Logged in as: Guest

Users viewing this topic: none

All Forums >> [Cellar Talk] >> General Discussion >> RE: Wall Street Journal Article: Wine Critics Are Not Consistent in Ratings

Page: << < prev 1 [2]

Message

<< Older Topic Newer Topic >>

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 10:39:41 AM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

TWG, in retrospective I am thankful, Theory of Probability helped me a lot in real life.

"What doesn't kill you..."
;)

_____________________________

tsk....

(in reply to TexasWineGeek)

Post #: 31

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 11:02:39 AM

TexasWineGeek

Posts: 1568
Joined: 8/18/2009
From: Mclean, Virginia and Cascais, Portugal
Status: offline

quote:

ORIGINAL: Serge Birbrair

TWG, in retrospective I am thankful, Theory of Probability helped me a lot in real life.

"What doesn't kill you..."
;)

Hmmmm, Serge, what was the probability you would have come to that decision....

_____________________________

TWG

Consuming wine in moderation daily will help people to die young as late as possible. ~ Dr Philip Norrie

Too Much Wine and Not Enough Time ~

(in reply to Serge Birbrair)

Post #: 32

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 11:16:55 AM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

quote:

ORIGINAL: TexasWineGeek

quote:

ORIGINAL: Serge Birbrair

TWG, in retrospective I am thankful, Theory of Probability helped me a lot in real life.

"What doesn't kill you..."
;)

Hmmmm, Serge, what was the probability you would have come to that decision....

100% +/- 3 points ;)

_____________________________

tsk....

(in reply to TexasWineGeek)

Post #: 33

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 11:54:35 AM

masi3v

Posts: 264
Joined: 8/1/2008
From: Philadelphia
Status: offline

A couple observations:

Not to put words onto Serge's fingers, but instead of random, could you mean arbitrary? Can any taster, RP included, really put as fine a point between an 88 and an 89 (or any other two numbers). Scoring it an 89 over an 88 is an arbitrary choice no matter how much you try to quantify it. Scoring something a 90 over an 89 is no more arbitrary, but the economic consequences for the winemaker can be huge. All reviewers claim to be impervious to such pressures, but I am not so sure. A winery that has been very accommodating and welcoming....

As for the average CT scores, I am not so sure that they are more 'honest' (in the sense of 'lack of bias'). I would maintain that by far, the vast majority of the wines scored on CT were purchased by the 'reviewer' and that s/he knew what the wine was when reviewing it ('i.e., not 'blind'). This almost has to have an effect on the score. If I plop down $120 on a bottle of Champagne (my weakness), I am going to expect (hope?) that it is better than the average $30 N.V. If it isn't, I just blew a cool $90 that I could have used to buy more of the $30 N.V.! So, even though I will claim on being objective, will I (subconsciously) score the $120 bottle even slightly higher to justify my indulgence? I would venture that I might.

_____________________________

“Je ne peux vivre sans Champagne, en cas de victoire, je le mérite ; en cas de défaite, j’en ai besoin.” -- Napoléon Bonaparte

Blog Link:The Drunken Cyclist

(in reply to Serge Birbrair)

Post #: 34

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 12:02:44 PM

TexasWineGeek

Posts: 1568
Joined: 8/18/2009
From: Mclean, Virginia and Cascais, Portugal
Status: offline

quote:

ORIGINAL: masi3v

A couple observations:

Not to put words onto Serge's fingers, but instead of random, could you mean arbitrary? Can any taster, RP included, really put as fine a point between an 88 and an 89 (or any other two numbers). Scoring it an 89 over an 88 is an arbitrary choice no matter how much you try to quantify it. Scoring something a 90 over an 89 is no more arbitrary, but the economic consequences for the winemaker can be huge. All reviewers claim to be impervious to such pressures, but I am not so sure. A winery that has been very accommodating and welcoming....

As for the average CT scores, I am not so sure that they are more 'honest' (in the sense of 'lack of bias'). I would maintain that by far, the vast majority of the wines scored on CT were purchased by the 'reviewer' and that s/he knew what the wine was when reviewing it ('i.e., not 'blind'). This almost has to have an effect on the score. If I plop down $120 on a bottle of Champagne (my weakness), I am going to expect (hope?) that it is better than the average $30 N.V. If it isn't, I just blew a cool $90 that I could have used to buy more of the $30 N.V.! So, even though I will claim on being objective, will I (subconsciously) score the $120 bottle even slightly higher to justify my indulgence? I would venture that I might.

Ahhh, grasshopper, you support my point exactly. Forget the term 'honest' if it bothers you. The point is that you get to read what those on CT sense and feel about their wine and over time it creates a 'profile'. When I read the TN's I can ultimately identify someone whose 'profile' mirrors mine and then I can virtually taste before purchasing. The 'probability'

that the wine will be to my liking will therefore increase and I should be a happy camper

. Forget the scores....WAY TOO ARBITRARY

_____________________________

TWG

Consuming wine in moderation daily will help people to die young as late as possible. ~ Dr Philip Norrie

Too Much Wine and Not Enough Time ~

(in reply to masi3v)

Post #: 35

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 12:11:19 PM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

masi3v, I think you used better word than me and random, should definitely be read as arbitrary

thank you.

_____________________________

tsk....

(in reply to TexasWineGeek)

Post #: 36

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 1:03:25 PM

fingers

Posts: 8261
Joined: 8/26/2006
From: Santa Ana, CA
Status: offline

quote:

ORIGINAL: masi3v
Not to put words onto Serge's fingers,

Somehow, I felt violated when I read that ;)

(in reply to masi3v)

Post #: 37

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 1:30:16 PM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

Here is a small experiment anybody can do.

Well, I'm sure the most of you are familiar with Bell Curve

tell your child of 10 years and older to rate wines in your collection based on the wine labels, your spread sheet records, CT name records, or any other means.

Give a child the following instructions:

1) Give most of the wines scores of 92 plus minus 3 points arbitrarily, and stay between 100 and 85

2) compare the child scores to Parker's scores. If child's score within 3 points from Parker, the child get ONE, if more than 3 points - the child gets ZERO

Question:
what is % of child scores correlate with Parker's ( plus / minus 3 points)

I have a good feel that majority of you are having wine geniuses for the kids.
Those who have no kids - you can borrow neighbor's kids, but alas, this won't give one bragging rights.

Care to prove me wrong?

_____________________________

tsk....

(in reply to fingers)

Post #: 38

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 1:39:01 PM

Blue Shorts

Posts: 2782
Joined: 2/5/2008
From: Santa Cruz
Status: offline

I believe that you are referring to a Normal distribution. The bell curve only applies to normally distributed data, and you haven't proven that the data is normal, so we don't know anything about the distribution.

If you want to compare the Averages of the scores, then the bell curve, or something similar, will apply.... by definition.

(in reply to Serge Birbrair)

Post #: 39

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 1:43:35 PM

TexasWineGeek

Posts: 1568
Joined: 8/18/2009
From: Mclean, Virginia and Cascais, Portugal
Status: offline

quote:

ORIGINAL: Serge Birbrair

Here is a small experiment anybody can do.

Well, I'm sure the most of you are familiar with Bell Curve

tell your child of 10 years and older to rate wines in your collection based on the wine labels, your spread sheet records, CT name records, or any other means.

Give a child the following instructions:

1) Give most of the wines scores of 92 plus minus 3 points arbitrarily, and stay between 100 and 85

2) compare the child scores to Parker's scores. If child's score within 3 points from Parker, the child get ONE, if more than 3 points - the child gets ZERO

Question:
what is % of child scores correlate with Parker's ( plus / minus 3 points)

I have a good feel that majority of you are having wine geniuses for the kids.
Those who have no kids - you can borrow neighbor's kids, but alas, this won't give one bragging rights.

Care to prove me wrong?

Serge,

Your approach is tantamount to predicting IQ scores from shoe size. It 'could' happen. Additionally, you 'restrict' their range which 'inflates' the probability that they will prove your point.

Probability theory suggests that by chance alone a given child will have chosen wines normally distributed. For your experiment to be meaningful 'statistically' the child(ren) would have to repeat the experiment a myriad of times and then assess the 'overall' distributions of the multiple trials. If the average of the trials results in a normal distribution you then 'may' have a point.

_____________________________

TWG

Consuming wine in moderation daily will help people to die young as late as possible. ~ Dr Philip Norrie

Too Much Wine and Not Enough Time ~

(in reply to Serge Birbrair)

Post #: 40

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 1:44:29 PM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

Blue, you don't have to use Bell curve, I MADE an ASSUMPTION based on it that your kid will do just fine.

All YOU have to do is explain your kid what is required, pull a 20 bottles and let the kid score the wines.

If the kid is within 3 points from the masters - kid gets 1. Simple, isn't it?

_____________________________

tsk....

(in reply to Blue Shorts)

Post #: 41

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 1:46:17 PM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

TWG, excuse me, but do the wine critics taste wines the numerous times blind and THEN publish the scores? No, they do not!
All I am saying is that the children will do as good as wine critics.

PROVE ME WRONG.

_____________________________

tsk....

(in reply to Serge Birbrair)

Post #: 42

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 1:51:56 PM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

In case you are wondering,
THIS has been done before, in different field but the similar methods:
http://www.automaticfinances.com/monkey-stock-picking/

I didn't invent the wheel
;)

_____________________________

tsk....

(in reply to Serge Birbrair)

Post #: 43

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 5:11:11 PM

TexasWineGeek

Posts: 1568
Joined: 8/18/2009
From: Mclean, Virginia and Cascais, Portugal
Status: offline

quote:

ORIGINAL: Serge Birbrair

TWG, excuse me, but do the wine critics taste wines the numerous times blind and THEN publish the scores? No, they do not!
All I am saying is that the children will do as good as wine critics.

PROVE ME WRONG.

Agreed and not arguing your point per sei, however, if you are to appeal to the normal distribution than your method is flawed. If on the other hand all you want, as you say, is to compare the children with the 'pro's' than fine. It is your application of the normal distribution I take issue with.

Again, while the article you appeal to has merits in its own right I take issue with any such approach based on a normal distribution theory for far too many statistical reasons than worthy of my taking up space for statistical theory on this forum.

You win

_____________________________

TWG

Consuming wine in moderation daily will help people to die young as late as possible. ~ Dr Philip Norrie

Too Much Wine and Not Enough Time ~

(in reply to Serge Birbrair)

Post #: 44

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/17/2009 7:00:21 PM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

TWG, the Bell curve use is what I base my theory and experiment on. All one has to do is to prove me and my theory wrong.

You think I'll be proven wrong by experiments?
;)

Honestly.

_____________________________

tsk....

(in reply to TexasWineGeek)

Post #: 45

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 5:05:14 AM

TexasWineGeek

Posts: 1568
Joined: 8/18/2009
From: Mclean, Virginia and Cascais, Portugal
Status: offline

quote:

ORIGINAL: Serge Birbrair

TWG, the Bell curve use is what I base my theory and experiment on. All one has to do is to prove me and my theory wrong.

You think I'll be proven wrong by experiments?
;)

Honestly.

The curse of my profession is that I DON'T think anything about outcomes until I see the data

I guess we can just agree to disagree...that's what is so great about this forum and friendship. I actually hope someone tries the experiment, I think it would be interesting to see the results.

_____________________________

TWG

Consuming wine in moderation daily will help people to die young as late as possible. ~ Dr Philip Norrie

Too Much Wine and Not Enough Time ~

(in reply to Serge Birbrair)

Post #: 46

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 5:23:38 AM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

quote:

ORIGINAL: TexasWineGeek
I actually hope someone tries the experiment, I think it would be interesting to see the results.

exactly! The theory is just a theory if not backed by the experiment data!
;)

I'd do it myself (I have 13 y/o in the household) but.....
I have no idea what Parker's or others scores are on my wine :)

_____________________________

tsk....

(in reply to TexasWineGeek)

Post #: 47

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 5:31:02 AM

nwinther

Posts: 929
Joined: 7/28/2006
From: Denmark
Status: offline

I don't have a lot of high-scoring Parker wines, so would it be alright, if I go by CT scores? If all I have is 90-point wines, then I'll by definition prove you wrong (as I tell the kid to rate it according to lable, the kid will not stack them all at 87-93, but surely will pour a lot way above 93).

If I prove you wrong, will it change anything for you, or will it essentially be a waste of time?

I'm much more intrigued by you original proposal, for me to taste and score wine (as the experiment requires that I drink wine!).

_____________________________

What I lack in size I make up for in obnoxiousness.

(in reply to TexasWineGeek)

Post #: 48

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 5:40:40 AM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

CT scores as we all suspect, otherwise we wouldn't be here are less arbitrary than critics scores.

Number two: the kid must be given instructions that 75% of the scores should be in the +/- from 90 range
(just like the critics scores, MY assumption, based on the Bell curve TWG tells me not to use, but I dare anyway :) )

Your kid's scores and results together with other, at least 10-20 results will show us all if the little monkeys are as good as crtics or considerable worse.

I say the little monkeys in our bedrooms are as good as WS, WA and everything in between
;)

_____________________________

tsk....

(in reply to nwinther)

Post #: 49

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 6:25:17 AM

nwinther

Posts: 929
Joined: 7/28/2006
From: Denmark
Status: offline

I don't have a kid to experiment on (geeh, that sounds eerie) - but I have several wine-oblivious friends - and I mean oblivious. The sort that can barely distinguish red from white - and really doesn't care, as they'd rather drink beer. Can I try with one of them?

_____________________________

What I lack in size I make up for in obnoxiousness.

(in reply to Serge Birbrair)

Post #: 50

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 6:37:14 AM

TexasWineGeek

Posts: 1568
Joined: 8/18/2009
From: Mclean, Virginia and Cascais, Portugal
Status: offline

quote:

ORIGINAL: nwinther

I don't have a kid to experiment on (geeh, that sounds eerie) - but I have several wine-oblivious friends - and I mean oblivious. The sort that can barely distinguish red from white - and really doesn't care, as they'd rather drink beer. Can I try with one of them?

Wait a minute...you live in Denmark...do people REALLY drink wine in Denmark or is this a rouge...hmmmm...maybe the experiement will work there...sure try one of them

_____________________________

TWG

Consuming wine in moderation daily will help people to die young as late as possible. ~ Dr Philip Norrie

Too Much Wine and Not Enough Time ~

(in reply to nwinther)

Post #: 51

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 7:34:18 AM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

quote:

ORIGINAL: nwinther

I don't have a kid to experiment on (geeh, that sounds eerie) - but I have several wine-oblivious friends - and I mean oblivious. The sort that can barely distinguish red from white - and really doesn't care, as they'd rather drink beer. Can I try with one of them?

of course! They are PERFECT!!!!!! You say they are oblivious to wines, you'll be shocked how GOOD they are in assigning points to your juice they never tasted!
;)

_____________________________

tsk....

(in reply to nwinther)

Post #: 52

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 7:35:14 AM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

quote:

ORIGINAL: TexasWineGeek

quote:

ORIGINAL: nwinther

I don't have a kid to experiment on (geeh, that sounds eerie) - but I have several wine-oblivious friends - and I mean oblivious. The sort that can barely distinguish red from white - and really doesn't care, as they'd rather drink beer. Can I try with one of them?

Wait a minute...you live in Denmark...do people REALLY drink wine in Denmark or is this a rouge...hmmmm...maybe the experiement will work there...sure try one of them

Denmark is #5 country in the world in alcohol PER CAPITA consumption!

_____________________________

tsk....

(in reply to TexasWineGeek)

Post #: 53

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 7:51:44 AM

TexasWineGeek

Posts: 1568
Joined: 8/18/2009
From: Mclean, Virginia and Cascais, Portugal
Status: offline

quote:

ORIGINAL: Serge Birbrair

quote:

ORIGINAL: TexasWineGeek

quote:

ORIGINAL: nwinther

I don't have a kid to experiment on (geeh, that sounds eerie) - but I have several wine-oblivious friends - and I mean oblivious. The sort that can barely distinguish red from white - and really doesn't care, as they'd rather drink beer. Can I try with one of them?

Wait a minute...you live in Denmark...do people REALLY drink wine in Denmark or is this a rouge...hmmmm...maybe the experiement will work there...sure try one of them

Denmark is #5 country in the world in alcohol PER CAPITA consumption!

Alcohol does NOT necessarily mean wine only. What is their ratio of wine consumption PER CAPITA with the world. I would think beer to be their primary choice but what do I know

_____________________________

TWG

Consuming wine in moderation daily will help people to die young as late as possible. ~ Dr Philip Norrie

Too Much Wine and Not Enough Time ~

(in reply to Serge Birbrair)

Post #: 54

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 8:37:16 AM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

http://www.winebiz.com.au/statistics/world.asp
bad formatting

< Message edited by Serge Birbrair -- 11/18/2009 8:38:53 AM >

_____________________________

tsk....

(in reply to TexasWineGeek)

Post #: 55

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 8:38:09 AM

Serge Birbrair

Posts: 4057
Joined: 4/23/2006
From: Boca Raton, Florida
Status: offline

http://www.winebiz.com.au/statistics/world.asp

Table 35. Top wine consuming nations per capita, 2005 Country 2003 2004 2005 % change (04-05)
France 56.8 55.1 55.4 0.5
Luxembourg 55.8 57.3 54.6 -4.7
Portugal 51.2 47.1 46.7 -0.8
Italy 50.6 48.8 46.5 -4.7
Slovenia 44.7 44.7 44.7 0.0
Croatia 38.7 40.9 40.8 -0.2
Switzerland 41.1 40.5 39.3 -3.0
Hungary 30.7 30.4 34.7 14.1
Greece 27.7 29.7 32.2 8.4
Spain 32.7 32.6 31.8 -2.5
Austria 29.2 29.4 29.3 -0.3
Denmark 31.6 29.8 28.7 -3.7

#12 in the world as of 2005

< Message edited by Serge Birbrair -- 11/18/2009 8:39:17 AM >

_____________________________

tsk....

(in reply to Serge Birbrair)

Post #: 56

RE: Wall Street Journal Article: Wine Critics Are Not C... - 11/18/2009 10:02:28 AM

TexasWineGeek

Posts: 1568
Joined: 8/18/2009
From: Mclean, Virginia and Cascais, Portugal
Status: offline

quote:

ORIGINAL: Serge Birbrair

http://www.winebiz.com.au/statistics/world.asp

Table 35. Top wine consuming nations per capita, 2005 Country 2003 2004 2005 % change (04-05)
France 56.8 55.1 55.4 0.5
Luxembourg 55.8 57.3 54.6 -4.7
Portugal 51.2 47.1 46.7 -0.8
Italy 50.6 48.8 46.5 -4.7
Slovenia 44.7 44.7 44.7 0.0
Croatia 38.7 40.9 40.8 -0.2
Switzerland 41.1 40.5 39.3 -3.0
Hungary 30.7 30.4 34.7 14.1
Greece 27.7 29.7 32.2 8.4
Spain 32.7 32.6 31.8 -2.5
Austria 29.2 29.4 29.3 -0.3
Denmark 31.6 29.8 28.7 -3.7

#12 in the world as of 2005

Thanks Serge,

Statistical analysis suggests that the countries differ in rank based upon percent per capita as follows:

First                 Luxembourg, France
Second            Italy, Portugal
Third               Slovenia
Fourth             Switzerland, Croatia
Fifth                Spain, Hungary, Denmark, Greece, Austria

(in reply to Serge Birbrair)

Post #: 57