Even Suckling's entitled to change his mind ...

[John M.]

Hey Peter--I respect your opinion, I just look at if differently--there's no shame or crow to eat. Let's enjoy and try to find that elusive 100.

[Eric Menchen]

Hey Peter--I respect your opinion, I just look at if differently--there's no shame or crow to eat. Let's enjoy and try to find that elusive 100.

And then try to find one that is even better!

[Roy Hersh]

Yes, 95+% of wines, Ports etc. that are using the 100 point scale wind up at least at 80 points. Retailers know the difference in one point between 89-90 points is worth huge number of sales. The same can be said for the difference between 99 and 100. But a 99 point wine will still sell a ton, while an 89 point may be overlooked by those deeming anything below the magical score of 90 to be unworthy of their purchasing dollars. Very ignorant way of looking at wine, as it is something to be enjoyed and not picked apart. I left one tasting group after nearly a decade of participation because they would FOCUS on the flaws in wines they tasted and not look for the good components and what they liked about the wine. The negativity nabob started to really bug me, but I put up with it for a long time because they were friends that I do like.

I have rated wines below 80 points about 10x for every one I've ever rated 100 points. Then again, I believe there have only been 3 x 100 point wines I've ever had (one Port, one Madeira and one table wine) in 31 years of "legal" wine consumption. Then again, I was not rating wines until the early 1990s.

[Peter W. Meek]

I see several people have responded while I was composing this, but here it is anyway:

Here is a better explanation of what I would expect from a good scale, and perhaps the better tasters actually come close to this. This diagram explains that the extremes (beyond 3SD) should occur only about 4 times in 1000 (two at the high end, and two at the low end). So, my guesstimate of one or two hundred-pointers in a thousand samples was pretty much on the mark. The actual number is roughly two.

If Roy has rated three wines at 100 in his entire rating history, then he is pretty doggone close to matching this expectation.

On a 5-star scale, I would expect about 1% to get 5 stars and about 1% to get 1 star, nearly 70% should get 3 Stars. 2 and 4 stars should get about 15% each.

Image and caption from the Wikipedia article on Percentile

The dark blue zone represents one standard deviation on either side of the mean, which accounts for about 68.2% of the population. Two standard deviations from the mean (medium and dark blue) account for about 95.4%, and three standard deviations (light, medium, and dark blue) account for about 99.6%. (From Wikipedia under the common license.)

[Peter W. Meek]

... I left one tasting group after nearly a decade of participation because they would FOCUS on the flaws in wines they tasted and not look for the good components and what they liked about the wine....

This reminds me of the story about a famous music critic (whose name I can't remember). Someone came up to him after a (very good) concert and asked if he had noticed some particular errors in the performance. The critic replied, "Anyone can hear a wrong note."

[Glenn E.]

Plus there is the skewing factor: experienced tasters seek out the better wines, and may experience them a higher percentage of the time than the average person. Plus an additional skewing factor: presumably wine makers attempt to make extremely good wines and succeed at times.

One more skewing factor, related to your second one: over time, wine makers simply get better at their jobs and so over time, wines of all types keep getting better and better.

That's not to say that the best wines keep getting better and better, but for sure the mid and mid-upper tiers of wines are better now than they were 20-40 years ago.

And FWIW, the 100-point scale is really a 50-point scale of which only the top 20 points are typically used. I've heard (but never seen explained) that the 100 point scale literally starts at 50 points and goes up from there.

[Eric Menchen]

Another skewing factor: Will a very bad wine even be released? If you made it and thought it was a 70-pointer, would you release it? You might try to figure out something to do with it (blend it, distill it, ...) so as to not be a total loss, buy you probably wouldn't put your premiere label on it.

[Peter W. Meek]

... I've heard (but never seen explained) that the 100 point scale literally starts at 50 points and goes up from there.

According to Wikipedia (always take these with a grain of salt) Parker based his 50-100 point system on the American academic grading system used in the 1970s where below 50 was a failing grade and the rest went up by decades to 100.

Maybe you rarely see wines rated below 80 points because, to paraphrase the late Col. Townsend Whelen*, "Only good wines are interesting."

--------------
*Whelen originally said "accurate rifles".

[Richard Henderson]

To me the most remarkable thing about it is such a score for a Dow. I have always liked Dow but I never had one that was close to 100. You guys in the know, Roy ,Andy et al, what caused this surge in quality?

[Andy Velebil]

You guys in the know, Roy ,Andy et al, what caused this surge in quality?

Your guess is as good as mine.

[Roy Hersh]

It must have been a great cigar that came before it.

[Eric Ifune]

Never underestimate the qualities of a good cigar!

[Mahmoud Ali]

Pete, you shouldn't have backed down. You're actually right.

After years of reading Spectator and Parker scores one comes to realize that they are effectively 20-point scales.In addition, both reviewers use qualitative terms to describe wines in the 96-100 range, the 90-95 range, the 85-89 range, etc. Whose kidding who, it ends up being akin to a 5-Star system.

I also believe that most people realize that a 5-Star wine is a top quality wine in the highest bracket, not necessarily the very best wine. We all know that 5-Star hotels and restaurants are not equal nor identical, just among the very best.

Don't apologize, there are many of us who agree with you.

Cheers...................Mahmoud.

[Moses Botbol]

I have rated wines below 80 points about 10x for every one I've ever rated 100 points. Then again, I believe there have only been 3 x 100 point wines I've ever had (one Port, one Madeira and one table wine) in 31 years of "legal" wine consumption. Then again, I was not rating wines until the early 1990s.

I have never rated a port 100 points, but a few were certainly up there and probably worthy of such a score. I just don't think there is a 100 point for me.

What are typically considered some of the best ports ever made are still 97-98 points in my book. At some point it comes down to the bottle along with the vintage. 70-80 years in the bottle and lack of multiple bottles to sample makes the score unique to the bottle at hand than the whole vintage for me.

[Andy Velebil]

I know Parker used to publish his low scores, but now I don't think he publicly releases any score that is less than about 80-83 points. Or at least he vary rarely does. I am almost positive he said somewhere he does that on purpose. If that's the case, then what's the point of being a wine reviewer if you're afraid to post low scores on wines that aren't so good? IMO it does a dis-service to the consumer who should know what is good AND not so good.

[Peter W. Meek]

I know Parker used to publish his low scores, but now I don't think he publicly releases any score that is less than about 80-83 points. Or at least he vary rarely does. I am almost positive he said somewhere he does that on purpose. If that's the case, then what's the point of being a wine reviewer if you're afraid to post low scores on wines that aren't so good? IMO it does a dis-service to the consumer who should know what is good AND not so good.

It my be a case of "If you can't say something nice, then don't say anything at all." Certainly that has always been an adage drilled into children.

I agree, that no review is not quite as useful as a poor review. However, posting bad reviews annoys producers who may be your source for the next great wine that you DO want an early sample of.

[Michael Hann]

The scoring systems are conventional, and it would be hard to make objective arguments supporting them. For professionals the resolution implied by a 0-100 point scale may make sense, but not for a non-professional like myself. For me, what is the difference between a 93/100 and a 94/100 Port? Precious little. I would be entirely prone, I suppose, to rank Port A 93/100 and Port B 94/100 on one day and on another day rank Port A 94/100 and Port B 93/100. I do not have the sensitivity to discriminate at this level of resolution. For me a 10 point scale would make more sense. I can resolve the difference between a 8/10 Port from a 9/10 Port and have some chance of repeating this difference in the same way on different days.

I'm not complaining, just attempting to exhibit something about the 0-100 scale. For me I accept the scale and don't question it. I try to learn what the norms are -- what is the rating for an average LBV? what is the rating for an above average LBV? -- and accommodate myself to these norms and interpret evaluations of Ports based on this standard accordingly. So long as the scoring scale is used reasonably consistently, it will convey valuable information and is hence useful. Is the scale perfect, rational, the best scale? I leave that to others to debate.

I like Peter's analysis. My mind finds the idea of placing wines on a quality distribution continuum very attractive, while not necessarily practical or implementable. What more do you need to know about a VP than that it is in the upper 95% quality distribution for VPs? or in the upper 80% quality distribution for VPs? I even accept the idea that this distribution would shift and evolve over time as the mid-tier and lower-tier shippers improved their game. On the other hand, if used well, scores given in the 0-100 point system can be mapped over to the quality distribution Peter alludes to by appreciating correctly the existing norms.

Intesting topic.

[Luc Gauthier]

I've always used a 20 pt scale .
Why ? I just decided to follow Fred .
Simple scale , simple answer ...

[Lamont Huxley]

I've been scoring wines on the 100 point scale for a few years now and have come to think of it more as a 30 point scale (70-100) as that's the range that all of my scores have fallen within thus far. It seems like everyone looks at these scales a little bit differently, but I think that's fine. We all get the general ideal that someone is trying to get across when they score a wine, but it's when we are personally scoring a wine that the small differences are really important because that's where we are honing our unique sense of what we like and basing our score on all of our previous experiences. The difference between my own 89 and 90 points is much more meaningful to me than someone else's.

We can never hope to perfectly understand another person's palate, but if you're just careful not to put too much weight into any one person's opinion, then it can certainly be very helpful to check out what other's think. Your own opinion should always be the most important, but if you aren't able to taste the wine and decide for yourself it's good to get as many different opinions as possible.

Edit: I think scoring the same wine differently on different occasions is only natural. I'd be more surprised to see someone rate the same wine the same score each time, although a major swing might make me scratch my head...

[Roy Hersh]

Bottle variation counts and when you have been scoring a wine (blind or otherwise) in a range of 2-4 points over the course of 4-20 bottles as an example only ... then when you find a score from that person for the same wine and the variation is significantly out of lock step with said individual's range, then more often than not it speaks more about that particular bottle.

Nonetheless, I do agree with the 30 point range mentioned by Mr. Huxley.

It does little service to even score below 70 points as that would be a defective or undrinkable wine and then the question arises why have the arbitrary 100 point range. I did not create it, so I won't defend its presence any further.

I do believe the most important thing when following a professional critic, is that person's consistency. So as you saw from the title of the thread, I have no issue with James flip flopping on the Dow preference over the Graham's as anybody, including a pro like JS has the right to do so and especially when re-evaluating between a cask sample done as early as May and a "finished" young bottle that is tasted 6 mos. or more down the road. Gentle teasing is one thing, but in reality, I respect James' move! (the 100 points on the Dow ... I'll leave that for others to debate).

I have been part of judging professional wine competitions where we rate wines based on organoleptic qualities and each one has a specific range of points it falls into, the total of which make up 100 points. It is a great way to learn to score for those looking to do so. I have used it for year.

That said, when I typically score in almost any other setting, my ratings of a wine (ANY wine) are subjective, and when I say that ... the difference between 1-2 points given a particular wine are based on norms that I have created myself within a particular category. For example, I can count on a few fingers of one hand how many LBVs have ever scored over 93 points. Some would say, if you love the 1994 Noval so much, why wouldn't that be a 96-98 point experience? My answer has always been that when I think of Ports that score in that range, they have qualities that the LBV's just never seem to reach for me, no matter how great they may be. This opens up another whole pandora's box, but why not?