Is your wine really getting better?

WINE TALK
41

I find it funny that on this board you never see consensus on a post especially with more than 30 responses. Actually forget about consensus you never see a controversial topic not go to the gutter with 15 posts let alone more than 30. But on this post everyone in a different way believes wine improves with age.

I actually in the back of my mind always think of this issue myself and wonder. So here is some of my data points.

  1. I have wine aging myself so I am like many of you
  2. I’ve always wondered why if you look at cellar tracker data it almost never supports improvement with time. I understand everyone’s viewpoints and agree/want to agree however if you look at cellar fracker most often the scores are higher in the beginning not the end. Everyone can say what they want but I just don’t see it with the scores and to me if you were to speak from a data centric standpoint everyone’s qualitative opinions are unfounded.
  3. over the course of the last 5 years most of the aged wine I’ve drank has been cheaper. I’ve done a good job of staying away from my cellar so it can get older and most of the older wine I’ve drank has not been super expensive wine. I can tell you that in general with aged cheaper (sub $50) wine I would say you could make the argument wine doesn’t improve. Many major non wine drinkers have told me when they drink some of this wine they dislike the old flavor.
  4. probably the best wine I have had in the past 5 years is a 1964 Côte rotie at berns. I know this counteracts what I’ve said above but if I’m to look objectively I have to look at everything not just one wine moment.
  5. I truly believe Scot has validity in his point that if truly objective statisticians came in they would not only not see a correlation but could objectively state cellar fracker proves wine doesn’t improve. Obviously this is just an opinion.

So anyway I don’t have a definitional opinion myself on this subject. I’m just truly surprised how this topic is the only one I know of on winebezerkers that creates consensus.

42

Att. Davidlown.
Regarding aging of sub $50 wine, I have to disagree, at least concerning French red wine.
Many classic rustic wines are not approachable in their first years at all (some 10y++).
As type examples from Bordeaux: Lanessan, Sociando Mallet, Lafon Rochet.
Burgs from e.g. Faiveley (lower priced), are also a sharp experience the first many years.
IMO, these tough wines need to shed/integrate their tannins, before they show their qualities.

Kind regards,
Soren.

43

CT has a set rating scale that’s supposed to be used by every member yet there are plenty who deviate from it. I even questioned one CT member why he said a wine was merely good yet gave it a score of 94. Here’s the exchange:

Me: Help me out here - “good but not great” is a 94? I’m still relatively new so am confused comparing this to the scoring guidelines given when you click on “Wine Rating Assistant” which shows Good as 80 - 85 and Very Good as 86-89. A 94 would be the start of Outstanding so I’m thrown by your tasting note in combination with your score. Thanks!

CT Member: _Hello xxxxxxxxx, you are raising a very good point here, as a matter of fact an evening filling topic. Let me try to summarize as follows:

The 100 point scale was originally developed by Robert Parker and had the following nomenclature (still officially stated as such on the WA website):

<59 Appalling,
60-69 below average,
70-79 average,
80-89 above average to excellent,
90-95 outstanding,
96-100 extraordinary.

Putting the discussion aside that in English outstanding and extraordinary are synonyms and therefore mean the same thing, there is a bigger issue with this scale. Robert Parker has not adhered to it himself because a) he does not taste the really cheap, super market stuff that would typically score in the 60ies and 70ies and b) when he gave scores below 85 in the past for well known wines, he got in trouble with wine makers. So, today he does not use the full scale anymore and does not score below 85 but rather remain silent in such a case.

Most tasters who use the 100 point scale have also adopted this approach and therefore the 100 point scale today is essentially a 15 point scale, 85-100. It therefore accepted knowledge that the 100 point scale must be read as follows:
\

79: Appalling (almost never used in practice, but if you want to offend a wine maker, go for it)
80-84: Bad (rarely used in practice, still better be silent, I personally use this range for bad, super market kind of wines)
OK, so the scoring really starts here then:
85-87: Unsatisfactory, without having to offend anyone (hey, I still gave it 85/100…)
88-89: Decent wines but nothing exciting
90-94: Good, fully satisfying wines, but no greatness in the glass yet.
95-97: Very good wines. Great stuff by any measure
98-100: Simply the best of the best

As you can see, both the meaning of the scale and also where the bands are cut are different. Essentially good wines start at 90, great wines at 95, legends at 98. As a wine maker, if your wine scores below 90, you will have a commercial problem with it in the fine wine segment and have to go back to the drawing board._

So for some reason, people feel it necessary to recalibrate a perfectly good scale used on CT as a common reference point with their own scale that’s adjusted for commercial interests. Personally, I think if you’re rating a wine for CT, you should use CT’s scoring system.

44

Great post there, Dennis - it captures succinctly why trying to statistically infer anything from a bunch of random scores on CT is really a waste of time.

If you actually have tasters who think a 94 is a “good” wine, what could you possibly hope to glean by aggregating their scores.

Oh, and yes, to the OP, my wine is getting better. :wink:

45

two things I’m seeing a lot here which I should have referenced earlier.

  1. CT launched in 2003. Thirteen years of data on wine aging is just not enough time. Mature wine is typically 15+ years, 20+ for the top wines. Best case scenario, we’re not there yet.

  2. Scores, even from the critics do not increase as wine ages or “gets better” - we don’t even have a working definition for what that is, in any event. A Grand Cru burgundy can be 95 points at release, and 95 points at full maturity 30 years later.

46

That’s because they’ve always been grading for potential. #nailedit

47

This is precisely why scores are useless to anyone but the scorer. These types of scores can be incredibly misleading. For instance, a young Barolo that gets a 96 point score might be practically undrinkable after release and somebody who scores based on how the wine is showing at the time of consumption (the way I try to score for my personal tasting) might give that wine an 85 if drunk at release. Since scorers don’t have to qualify their scores when posting them, the notes are really the only thing that can be qualitatively assessed.

Therein lies the rub. A taster may give a score of let’s say 95 at release of a very young wine that is very high based on expectation even though the wine isn’t really enjoyable (which I find presumptuous and far too subjective to that tasters prognosticative abilities). That same taster may revisit the same wine 5-10 years later and think the wine is mature and incredibly enjoyable and score the wine a 94. Does this mean the wine has gotten worse over time? No, it means that taster is a bad prognosticator.

That said, I have not yet had the opportunity to taste an older wine (greater than 15 years or more) that I thought it has gotten to some magical stage. This is due to many factors but mostly to my age and financial situation (too young and poor to age my own or buy aged fine wines). I hope to someday have that watershed moment but for now will drink and enjoy my modest wines as soon as they come into a good drinking window. So to say wines can improve significantly over time, I can’t really say but my experience has been that wines improve from release until they are mature enough to enjoy and then there is a point of diminishing returns and eventually the wine gets tired and dies. This life cycle, I would imagine, varies wildly for each wine and is much longer in those classic wines but I have yet to experience the magic of a wine hat gets to a good drinking window and then is significantly better 10-15 years later in that drinking window. Obviously others have been to the mountain top and have professed the blessings of these fully mature and well-aged wines. Until I get there, I can only speculate.

48

scores plus tasting notes from a trusted source are immensely valuable - that’s the whole … point. focusing just on points is generally useless, but it’s a decent shorthand and a starting point.

Obviously others have been to the mountain top and have professed the blessings of these fully mature and well-aged wines. Until I get there, I can only speculate.

this is a strange way to approach this hobby. you don’t have to speculate at all. you can rely on generations of people with experience on the subject. in fact, because so many wines have been around for so long, you can actually predict very well how they will perform in the future.

49

I just checked my wine last night and its absolutely getting better.

50

I think we should just head straight to a binary system of rating wines.

51

It’s a great post Scott. And it’s really true for everyone. Moreover, if you have a lot of experience with old Barolo, that doesn’t mean you know anything at all about old CA Cabs. Back when Jay Miller was reviewing Spain, he would announce that a wine was 100 points with 100 years. In many cases he’d never had the wines before and he most assuredly hadn’t had them with some years. Those of us who had knew that the wines were going to crap out in a few years because of what seemed to be bacterial problems. At the bodega the wines were best, on release they were still magnificent, and after five years you were wondering what happened.

I don’t want to be a jerk here but that’s why I discount most people’s reviews. Unless I know that person has had a lot of wine over a long time, and appreciates the development, I get no real information from their scores. And unless I know the person and his/her palate, their experience isn’t necessarily relevant because we may not come out at the same place anyway.

So while the OPs question is in fact reasonable, it can’t be answered in the way he’s trying. OTOH, it may be useful for people to do that for themselves. In fact, I think it’s very useful. It was only by doing that - recording my own impressions of various wines at various points, that I learned what I liked and that I preferred wines from area A over wines from area B.

I can’t really say but my experience has been that wines improve from release until they are mature enough to enjoy and then there is a point of diminishing returns and eventually the wine gets tired and dies. This life cycle, I would imagine, varies wildly for each wine

I think that is exactly true. Yacov is right in that there are some wines from some areas that have track records and those are very useful.

In the end though, it comes down to your preference. I’ve mentioned before I had a few friends who knew much more about wine than I ever will, and we often drank the same types of wine. The difference is that one called the other a baby-killer, while the other guy referred to the first as a necrophiliac. One would drink his wines within the first ten years and the other wouldn’t touch them unless they were 30 plus, preferably much older. I came out somewhere in the middle but not that strongly - I enjoy both sides.

With each wine there’s a kind of curve. Sometimes the slope up is long and slow, sometimes fast. Sometimes there’s a really long plateau that only gradually slopes down, maybe even longer than your lifetime, and sometimes it falls off pretty fast. “Mature” is very wine-specific. And vintage matters too.

As I get older I’m finding that I prefer some grapes and blends fairly young, while others are so much better with age that it’s a shame to drink them fast. But those are only personal preferences, and not necessarily widely shared.

However, with time you also can develop more confidence in your own assessments, in which case consensus just doesn’t matter at all.

52

But sometimes critics do go back and rerate a wine. Parker has done it several times with wines that I know of. For instance, the 2002 Lokoya Diamond Mtn was around a 94 initial rating, but back in 2012-ish he upped his score to a 99.

Yes! 0 = don’t buy again; 1 = buy again.

53

To the OP, the short answer is yes, my wines are really getting better. How do I do I know this? Well, I have been buying and cellaring wines for 30 years and I know what I like and prefer (which is of course purely subjective). I don’t need CT scores to tell me this. I like older more mature wines, quite simple really.

Lots of good points made by Yaacov, Greg T and many others.

The flawed premise from the OP was to treat the CT scores as it they were data rather than opinions dressed up as data.

All scores are SUBJECTIVE and are based on many factors including the experience, preference, context of the taster. That they are presented as numeric scores does not make them objective.
Assigning a numeric value to the subjective sensory experience of putting a wine on your mouth is not particularly scientific. But plenty of intelligent people do it. In the wine business, points are used for many purposes, but we should never mistake them for anything objective or scientific (such as treating them to a simple bivariate statistical analysis as the OP did). We use points as a short hand for what we (subjectively) think about the wine.

An analogy might be a person’s height. My wife is from Malaysia, the average height of adult male in Malaysia is 5ft 4in. So when a Malaysian says that that guy over there is tall, he may only be 5ft 8in tall. The concept of tallness has an obvious cultural (subjective) context. In all cultures the sense of “tallness” is an opinion not a value.

To get around this cultural subjectivity we measure people’s actual height. So now we use a NUMERIC value such as 5ft 8in to describe the person’s height. This is a universal, repeatable measurement. No matter who you are and where you come from if you are 5ft 8in tall this is a measurable and repeatable observation. This concept that the observations are repeatable measurements underlies all statistics.
Wine scores are just opinions and some of the professional who score wines are very experienced and hence they are extremely valuable opinions. BUT THEY ARE NOT DATA.

Using my height analogy, a professional wine critic is using points to signal relative height amongst similar wines he is tasting that day or week. E.g. when Steve Tanzer scores a young barrel sample red Bordeaux 93-95 points, he is saying this will be a very tall wine when it grows up. But because he uses a numeric score to communicate this subjective view, it can create the impression of greater accuracy and reliability than actually exists. Basically wine writers and critics use points as a short hand to summarise their subjective impressions.

Some wines are clearly proven themselves in the market place to be “taller” than others. Such wines tend to get consistently higher scores over time. My comments about scores not being scientific should not be taken to imply that the relative quality of truly world class wines is not distinguishable to most experience tasters. For example, someone like Jeremy Holmes describes the quality of the sensory experience of the wines he drinks very effectively without ever using points.

Sorry for my long winded meandering around the topic of why scores are not scientific measurements (despite the misleading appearance that they are). A pet peeve of my mine you might say!

Cheers Brodie

54

In response to the substantive question raised by the OP rather than to the CT measurement problem:

I have been cellaring wine for about a decade now. There are several reasons why I do that. I like to buy wines when I have the proper opportunity to get those I really want at the right price, I like to have a rather wide variety of wines to choose from, and in some cases, but only some, I’d like them to age a bit before I drink them.

Those I cellar in the hope that they will get better with time accounts for less than half of my stock and largely consists of Nebbiolo (mostly Barolo and Barbaresco, which I often find most enjoyable at an age of somewhere between 10 and 15 years) plus some higher-caliber Rieslings (but only if I have bought them really young and not for a longer time than a few years). And yes, the ones I hope will improve with age usually do (up to a point), although exceptions exist and the number of completely spoiled bottles, usually due to cork problems of one kind or another, regrettably increases the longer you wait.

However, most of the bottles in my cellar don’t improve with age. They don’t get noticeably worse either as long as I keep them around for just a few years. I don’t expect them to change much and as a rule they don’t. They are simply there waiting for me to find the proper time to drink them.

Finally, there is a third category that will not stand much in the way of aging, at least based on my taste and my “cellar” (actually a well-insulated storage room above ground with an average temperature of about 18 C and a seasonal variation from about 15 C to about 21 C). The main examples of wines that won’t hold up for very long, at least in my cellar, are certain whites with prominent and delicate scents, such as Sauvignon Blanc and Muscat. After having learnt that the hard way in a couple of instances, I got a “wine fridge” for this third category (it’s just an ordinary fridge with good energy efficiency) which I run at a temperature of 8 C (the highest to which it can be set).

The long and short of it is that I think you simply have to try for yourself and see how much aging of this or that kind of wine you really like. Personally, I tend to like my wines relatively fresh and energetic, which means that cellaring for aging is something I do only for a select set of wines and then not for all that long a time.

55

Yaacov is absolutely right, and the data is garbage. You can’t aggregate something as trite as a single number, from a large group of unknown tasters, knowing nothing about their experience and judgement, and expect to interpolate the aging curve of a wine.

CT scores, once the group of tasters is large enough, largely reflects the price of a wine. That could be because quality correlates with price, but to an any wine tasted non-blind, the experience often correlates to expectation, which is driven from price paid.

If you start talking about aging potential (scores of a young MB you’d expect are based on potential) then tasters rely on track record. If you get a $9 Cab from Costco, and it’s oaky, tannic, and shut down, you might just say it’s crap, because you’re just not going to buy it by the case and cellar it for 10 years to see if it comes around. You taste a Monte Bello the same way, but because the label has a track record, you raise a toast to its aging potential, and stock up.

CT gets even weirder than that. I recently looked up notes on a wine that had a very specific flaw (ethyl acetate) that AFAIK is a winemaking flaw and not something that develops in bottle (not subject to bottle variation.) It was a relatively expensive bottle (>$30). Reviews on this wine were mixed - some calling flaw, some calling “off bottle” and yet scoring in the upper 80s (!) Note “flaw” does not figure into the average, which ultimately came to 91.1 points. Take from that what you will, I call it garbage.

So, you’ve aggregated the average expectation for the label and the price. You could follow a pro, who might have extended experience with the wine over multiple vintages and single vintages over time, but he or she will always score potential of a young wine, not as a snapshot in time.

What’s the answer then? Taste wine yourself, taste blind sometimes, and read tasting notes. Give up on scores. Seriously.

56

Notes yes, not scores though.

57

Thanks to everyone for your input. Sincerely, as I know how much time some of these notes take to thoughtfully compose. Good dialogue.

I agree that just using scores from CT to come to any hard conflusion is flawed. It’s all personal taste (or speculation, if scoring potential). But I do think there is some value in the numbers. I was surprised at how many posters dismiss the scores as completely useless. I would imagine that the average CT account holder has at least more than a passing interest in wine and some experience. Certainly more than the typical Yelper scoring your local Denny’s. So I’d at least give the scores some credibility in terms of a first-pass direction on the wine’s enjoyablity.

Bottom line is that nothing surpasses personal taste and experience.
champagne.gif

58

Perhaps Eric should “normalize” scores. I.e., each CT member has his/her own scale. Presumably each person is relatively consistent within their own scale, whether 94 is “good” “great” or “outstanding” I’d hope most of us would consider 1 wine we scored 94 comparable to another. Thus it’s possible, at least after more than a few ratings, to figure out what each CT member’s average score and deviation is. One could then normalize that score to a standardized scale (for sake of argument, let’s say “average” is 85). For the CT member who scored a 94 wine as “good” presumably that person scores lots of wines at or above 94, and maybe 93 is his average score. Solution - knock 8 points off of his score (93-85) in computing the CT normalized average. While this wouldn’t eliminate all statistical problems, it would help resolve a significant one.

59

So we need to handicap CellarTracker, eh? I like it. The particular CT member we’re discussing could have a “+8” next to or under his screen name so everyone knows that he’s always 8 points too high compared to the CT scale. [cheers.gif]

60

Scott, I agree with your bottom line. And I agree with your contention that many CT users have significant experience. Their numbers do have meaning… to them. But even though the number of tasters is large, because everyone is using a different internal scale and different subgroups rank different bottles, I find it difficult to ascribe any value to a group mean when comparing different wines.