Stated vs Tested Alcohol Levels - '05 Two Hands Bella's

WINE TALK
1

Some context to start…

Here in the province of Ontario, all wine import and sales are controlled by our government monopoly - the LCBO (Liquor Control Board of Ontario). The huge, massive downside of this system is that the selection and price is dictated by government employed bureaucrats (selection poor, cost high). And unlike Pennsylvania, we have customs agents at the border who apply a 107% tax for imports above and beyond 2 bottles per trip :wink:

There are only two positive attributes of our system. 1) Any bottle can be returned for virtually any reason at any time. 2) Virtually every label (from every vintage) is tested in the lab before it is offered for sale. One attribute that is tested is the actual alcohol level in the bottled wine. If the label suggests a different value (I’m not sure if there is a threshold tolerance), the wine must have another label affixed showing this lab-tested value.

So on to tonight… I opened up a 2005 Two Hands Bella’s Garden Shiraz. It’s a very nice wine, intense, concentrated and pretty lively given the gobs of fruit that are the center of attention. There’s an LCBO alcohol label applied over a portion of the original label and it says 16.4%. The wine isn’t hot but I’m not surprised as the final lingering impression is certainly a little warm. I carefully peeled off the LCBO label and the original listed level is a more modest 14.8%.

My intention isn’t to complain about the wine - it was delicious, big and boisterous. But that’s quite a swing from 14.8 to 16.4 and goes to show that it is hard to trust what is listed on a label.

2

That is awesome Ken, I would love to see more tested wines from the LCBO up there, especially Aussies.

3

Always wonder what the truth is on some bottles, labeled vs. actual…

4

i’m sure this happens quite a bit for CA wines. Was told about one particularly popular pinot had a label of 14.8 but when tested was closer to 18%. Most of the time I can’t believe the labels since stuff tastes like rubbing alcohol

5

turley zins may say 15-16% but have to be pushing 18%+ easy

6

Outpost Zin is stated at 16.1% on the label and it is in fact very, very close if not exact. Thomas isn’t afraid to put the actual alcohol content on a bottle of wine. Ditto for Michael Browne of KB and Ed Kurtzman.

There are others that don’t.

Pegau every year on the REservee says (depending on the importer) 11-14% dessert wine (Martine’s in Novato, CA) and 13.5 on the bottles that Dan K imports, IIRC. The 2003 Pegau, tested by Mike Officer (IIRC), was closer to 15.5%-15.7% - I don’t remember the exact number other than it being in the mid 15% range.

More winemakers than you think routinely send samples of other people’s wines in for testing.

7

I think there would be a LOT of people left with a different view of some sacred wines if every winery from all countries put the real alcohol levels on their labels. Would be real fun to vintage variations in places like Bordeaux for instance.

8

Pavie 19%? [oops.gif]

9

Ken – Nice work! I hope you continue to peel labels and post the results here. Lots of us would be curious.

10

I was always told the regulation in the US was +/- 1.5% of what the bottle states. Having said that, I do not believe this is enforced all that rigorously.

Perhaps the guys and gals in Cellar Rats could weigh in on this.

11

This is an honest question . . .

Are singular tests 100% conclusive? If you get 16.1% on a test and then you (or someone else) tests again later will they get the exact same number? If there is any sort of fluctuation, how much generally (.01%, 1.0%)? Does that account for the government’s range on alcohol levels?

Reminds me a bit of scouts’ radar guns at baseball games . . .

12

Honestly, I can only recall seeing 6 or 7 wines with updated labels applied in the 4 years that I’ve been back living here in Toronto. In some cases, larger-distribution wine destined for this market will have Ontario-specifc labels (with the importing agent, etc.) and might reflect a different alcohol level than a label destined for a different market.

I’ll keep my eyes open and might pop in to the cellar tonight and check on some Pavie and Pegau [wink.gif]

13

Definitely not 100% conclusive… repeatability is usually good for one laboratory using one piece of equipment, testing the same wine at multiple points in time. Repeatability is worse and variation much higher for different people using different equipment at different times. Also keep in mind Ebulliometers are going to be less precise and have lesser repeatability than FTIR scanners, and presumably be less accurate, though there are no guarantees on those assertions :slight_smile:

Worth noting is that there are many producers who for various reasons – convenience, label registration ease, cost, etc, maintain the same alc % on their label from one year to the next, so long as they are fairly close.

I have no idea if there is any data on average margin of error for ABV tests, but off the top of my head I’d suggest that if you’re missing by .2% or even .5%, that is pretty understandable, when you’re missing by 1-2% or more, something is out of whack.

14

Interesting thread for me this week. I was brough a Syrah from a new Paso producer a couple of days ago. The label said 15.6%, but it tasted so hot I tested it and got 17.2%. And supposedly, a VERY well respected, cough cough Zin producer here in town had an '06 Zin out that stated about 16% and was actually over 19%.

Yes, testing methods vary considerably. Originally, the allowable variance was granted because the only testing method at the time, the ebulliometer, was so notoriously unreliable. Unfortunately for a lot of wineries, this is still the only testing method available, unless you send a sample to a lab that uses some sort of spectral method.

15

Many people will not buy wines with listed alcohol over 15%. In CA, wines have a plus or minus 1% leeway on labeling. Thus many cheat. If a wine is 15.8%, they can legally call it 14.8%. A big difference in marketing.

Kudos to the LCBO for enforcing truth in advertising.

16

It’s not necessarily that these wineries are cheating. As I stated above, there are some wineries that still use old methods of analysis like the ebulliometer, which are unreliable. There are also issues of purchasing labels and getting label approval. Many times labels will have to be submitted before a final blend of a particular wine has been assembled.

17

I know it’s common to not get a real alcohol level for any number of reasons, but I would love if there were someplace – label, winery web site, anywhere – where you could get honest alcohol figures. I don’t want to be part of a marketing game or a philosophical war. I just want to know how much booze is in what I’m drinking and keep trying to answer the question of whether I prefer higher or lower alcohol wines with the most objective data possible.

18

Wineries spend significant amounts of money yearly for all kinds of new equipment. What foolishness to say they can’t afford to buy modern equipment to test alcohol levels. This is not state of the art high tech equipment.

I see SVD’s that are of highly questionable markings also. Are you going to say that blending of a 600 case SVD is so difficult they may be off > 1%?

For wines with discrepancies of over 0.25% in alcohol levels. I’ll give the real issues 20% of the blame, with the other 80% going to winemaker’s not wanting to admit real levels.

19

The least expensive NIR option for testing alcohol levels that I know of is the Anton Parr, which I believe is about $15K. An EnoFoss or it’s larger and more expensive cousin, the Winescan, which uses FTIR technology, cost anywhere from $50-80K. Not a lot of wineries can afford either option. I’m not defending discrepancies of >1%, but not all winery operations are as neat/cut and dried as some of you might imagine. And I doubt that there is a large percent of these that are due to winemakers not wanting to admit the real levels. Many many wineries proudly display >16% ABV levels.