How to get the wine you really want

Twenty years ago I was a tea-totaller, but there was a group of people I wanted to hang out with who would get together to drink wine so I decided to try to develop a taste for it. I eventually got to be a pretty hard-core wine snob. My wife and I even ran a wine-tasting group for several years. Then one day we decided to start doing blind tastings, and we soon learned that there was very little correlation between what we liked and what the wine critics and the market liked. The real turning point came when we did a blind tasting of Cabernets, all from California. The wines ranged from a $5 bottle of Barefoot cab up to a $100 bottle of Silver Oak. There were twelve of us in the group and every single one rated the Silver Oak dead last, and the Barefoot either first or second. Since then we've repeated similar experiments many times, usually with similar results. We can still freak out wine snob friends on occasion by sneaking a bottle of Barefoot in to a high-end tasting.

On another occasion we were at a bed-and-breakfast that served afternoon wine and cheese. The whole room (including us) was raving about how complex and delicately nuanced the white wine was. It turned out to be Gallo Rhine Wine in a box.

(BTW, it's absolutely not the case that all cheap wines taste good. We tried the notorious two-buck chuck and thought it was absolutely horrible.)

The point is: wine snobbery is bullshit. Find what you like and drink that, and don't pay attention to anything that anyone else says. If Beringer white zinfandel is what turns you one, more power to you. (Actually, the worst possible outcome for a real wine fan is to discover that they really do enjoy Romanee Conti or Screaming Eagle. I really like Opus 1, but I wish I didn't :-)

So I think this roll-your-own-wine blender is a terrific idea. I can't wait to try it. (I wish I'd thought of it!)

Winemaker here. I would like to try those wines, but I don't think you will get good results out of it.

Taste of wine is mostly structural thing: it is created during fermentation by polymerisation of different compounds. Even mixing different wines don't give you good results, unless you fiddle with different ratios (I do as much as 10 for cuvee of two wines, trying to find sweet spots) and then letting it settle for some time.

Well, it might be fun activity, but I am sceptical that you will create wine that you will actually like.

I guess I'm spoiled. I get the wine I really want by living in Portugal.

There's so much good wine here that even the cheap stuff is good quality (we're talking $3 for a bottle). Unfortunately, I've found myself slightly allergic to reds (unless I only have a single glass) so a lot of the local market is off-limits thus I stick to white and 'green' (sweet wine).

I know of some local places here in Lisbon where a small glass of red is 75 cents. The best are Porto's wine 'caves', though, where one can try some really great varieties - five for 12 bucks. Great value for the quality and with prices that don't break the bank, it's of course possible to explore more of what the national market has to offer.

I have developed a new theory about wine and selecting bottles you/I like.

The theory is that pallets for wine develop over time as you drink more, similar to how beer will change. When I was 16, all beer tasted terrible. Then at 19 we 'upgraded' to bud light.

Now 15 years later I drink for taste and I like IPAs,or other non standard pilsner.

For me, I'm a beginner wine drinker so I do not want the complexity of fancy wine. I ask for 'beginner, fruit forward' and not expensive wine and have had amazing results. The problem is that pallets are different so blind taste tests won't solve the problem

In 15 years, I'll probably 'upgrade' again.

I have that Asian gene that limits how much alcohol I can drink, so I never drank more than a few sips of wine until I was well into my 30s. One day some friends and I went to a nice restaurant to celebrate a milestone, and we ordered a ~$120 bottle of syrah based on an app that showed critics' ratings.

It was a revelation. Whenever I had had wine before, I always picked wines that were considered "approachable." Sweeter, fruitier, and so on. But this wine was different, highly complex and paired beautifully with food (a first for me). Afterwards I chased down the wine online and found one of the few places that sold it, and had a few bottles shipped from across the country. Trying it again later with my wife, who also rarely drank, she agreed it was something special. Since then I've explored other wines and have found that my favorites are generally not the ones that are pushed as approachable for beginners.

I guess the lesson is, if you want to understand what the fuss is all about, don't be afraid to try ones that have more to offer than fruit and sugar. Tannins, full-bodied structure, and balance can be extraordinary.

I have found that there tends to be a sweet spot around $40-100/bottle for the best quality/price. Though to be fair, there are plenty of good wines for less, and plenty of duds in that range and above.

Before reading the article I thought it was going to be about taking people's DNA samples and using that to select the wine (https://www.thedrinksbusiness.com/2016/10/us-startup-claims-...) or using the phase of the moon (http://archive.fortune.com/magazines/fortune/fortune_archive...) but it actually sounds quite sensible.

I occasionally drink wine (in the restricted sense, i.e. made from grapes), but I prefer beer. There's a lot more variation in flavour (because there are at least four ingredients in beer, and umpteen different recipes), and even the very best beers - things like Trappist ales - aren't obscenely expensive.

Quelle horreur!

They found that most customers are stick-in-the-muds. Instead of sampling different regions, grape varieties and vintages, they tend to order the same plonk every time they go out.

And the solution isn't the Vinfusion, its getting rid of the snobbery and not making people feel like fools for not knowing their wine.

I was hoping this would be a guide to tweaking the linux wine. This does sound neat though, kind of like the paint mixing machines at department stores with a personality test on top. I wonder how far this could go, maybe something could suggest different ingredients based on your mood, and what mood you want to be in. Or maybe even using various biometrics to determine your overall "state" and the best way to alter it.... Just Friday afternoon musings.

> Traditionalists may be appalled by all this, but they should not be. In Mr Wimalaratne’s mind, the function of the Vinfusion system is in principle little different from the blending of grape varieties that goes on in many vineyards, to produce wines more interesting than those based on a single variety.

After spending a couple weeks in central coast and napa recently, tasting probably ~100 glasses, this kind of sounds insulting. Especially after seeing how much effort goes into cultivating the grapes, crafting these wines, and how they're stored and aged.

I get it though and perhaps this will work for some bar/restaurant. Whatever happened to house red or house white for those who don't care or know?

There's an excellent opportunity to apply machine learning here to automatically let you discover what you like. Starting from a set of varietals and/or random blends, it would generate new blends based on your rating feedback, eventually converging towards your preferred tastes. In addition to arriving at your optimal wine(s), a further analysis of the results (plus qualitative labels from experts) could also generate knowledge useful to you elsewhere, like "You like full-bodied left-bank Bordeaux, and also California Merlots. You dislike light-bodied wines in general."

The article mentions the customer configuring 3 sliders of wine characteristics, that might be too much decision making for me at a restaurant.

I'd still ask a sommelier standing nearby to pick something for me.

The trouble I have with wine is that I pretty much like all wine. I don't like extremely sweet wines, or very light in color wines because they're usually too sweet. But in my range, I can't really tell the difference.

So the $3.00 WalMart Oak Leaf is just fine with me.

I've read wine snob reviews of it, and they tend to be "hey, not bad."

There are wines that suit your taste and wines that don't.

I'm acetic that all wine recommendations are good.

The "70% disappointed" figure has nothing to do with anything but the fact that 95% of UK restaurants and pubs sell appalling wine. A merlot is a merlot, but if you can only order shit merlot you'll be disappointed regardless.

In South Africa, our family troll past-time at parties was to re-label wine we brought. We used the same wine everytime (our personal favourite) and disguised it with some fake fancy label.

> Instead of sampling different regions, grape varieties and vintages, they tend to order the same plonk every time they go out.

Of course they do! The problem with wine is that the user experience of selecting one is shitty. If you study how users work, to get them to explore, you need a couple of things:

1. They need signals and affordances to give them hints as to which new actions are most likely to be what they want.

2. After they have made a choice, they need to be able to quickly get feedback to determine if it was what they wanted.

3. They need to be confident that if they try an action and it isn't what they want, that they can retract their attempt easily and without negative consequences.

Wine fails at pretty much all of these.

1. The packaging and terminology around wine is inscrutable. The labels usually mention brand (sometimes), location, and variety. As far as I know, most of those have little internal correlation. Two wines from the same winery can taste totally different. Likewise, two wines from the same area, and using the same variety. If I like, say, Chateau LeBlah Merlot from Bordeaux, that doesn't give me much confidence that I can predict what Chateau LeBlah's Cabernet is like, or a Bordeaux with a different variety, or whatever.

The label also tells you the year and the price. Again, those numbers often do not correlate with anything I can easily discern in the taste of the wine.

2. This is half there. I can take a sip and tell if I like it. Well, except that doesn't work out as well as you'd expect. Food and environment has a huge impact on taste. There is also the social experience of drinking. If a friend says, "Oh my God, you have to try this, it's amazing!" then it probably will taste better to me.

If I like a glass, did I really like it, or did I just like the experience? Or did I feel pressured to like it? Or did the high price tag convince me I should have liked it and if I didn't I just don't have a refined enough palate? Should I stick with it to see if I acquire a taste and understand what all the fuss is about?

3. In theory, most restaurants let you try a taste before you commit. But that in itself is a high pressure social act. The rest of my companions will watch my taste it and pass judgement. It's stressful being on the spot.

Usually, I just pick something and stick with it. If I'm not enthused about it, I'll slowly work my way through it sip by sip. I would feel weird ordering something else with a full glass sitting there. And I don't want to just gulp it down to move on to another glass because I don't want to drink too much.

If I'm buying wine at a grocery store, there's less social pressure, but I'm even more committed. I have to buy a whole bottle, and I can't return it.

Of course people stick with the one or two wines they know. The whole system points directly to that as rational behavior.

I worked at a very high end restaurant for a while. To me it appears that wine tasting is an attempt to justify going out and drinking for most people. Wine snob sounds better than functioning alcoholic.

The awesome thing about wine is that you don't know what you want until you accidentally find another great wine.

I'm looking at a set of areas in which information acquisition is difficult, and in which choices or decisions have very strongly social implications. These are quite frequently areas in which choice is very strongly ruled by fads or fashions -- strong and often rapid shifts in preferences, but without a long-term persistence -- a winner in any one period will be likely to revert to mean in another (though other influences may sway this).

The areas I've identified:

1. Clothing and fashion -- the practical reasons for distinctions are few, and often counter the popular choices. But these have a very high signalling rate for social status and alignment.

2. Management practices. Here, the problem is that determining actual quality is quite expensive, most especially in time (results may take years, and there's a high noise element to measurement). At the same time, adoption signals group allegiance and has strong carry-on effects with in a market (consulting services, training, education, organisational systems, etc.).

3. Programming languages, tools, operating systems, editors. As with management tools, measurement is hard and takes a long time. At the same time, there are network and lock-in effects which can develop, as well as organisational and market-signalling effects.

4. Food and diets. "Fad diet" is a well-developed concept, going back to the 1960s. The concept of food as a social identifier is much, much older, and can be an element in keeping social groups separate or allowing for their integration. The segregation of Chinese and Muslim populations in Indonesia, for example, derives in part from the affinity for pork by the first and absolute aversion by the second group. That's contrasted to the situation in Thailand, where the Buddhist indigenous population doesn't have a pork aversion and mixes far more freely with the Chinese immigrant elite (from Amy Chua, Harvard scholar and author).

5. Pop music. Beyond a certain base musicality, hit tunes in the 1940s - 2000s were largely a result of both highly focused marketing and some degree of organic development. Experiments with "artificial" hits (within selected group populations) has seen different choices evolve within different groups, from the same base set of songs. Again, music is very highly associated with social identity.

6. Wine. Fads emerge here because it seems so difficult to objectively identify and communicate distinguishing characteristics among different wines. The market offerings are huge, and quite confusing, with information aquisition by individuals difficult and expensive. Choices are very strongly influenced by advertising, marketing, and chance exposure.

There are numerous other goods or behaviours which could be treated similarly. Sports choices, physical fitness training modalities, books (like wine: expensive to judge), etc., etc.

Not a fan of wine in general, it's too sweet. Even the dry ones.

I've had a lot of fancy bottles in a lot of fancy places, but all things equal, I've never tasted a good wine in poor company.

Most wine I have in a restaurant is good. If they do a good job picking wines there shouldn't be a problem and you should trust their staff. If they don't have good wine selection I guess you could get a glass of wine from this thing, but it's probably a better idea to get something they do a good job of selecting. After all not much sense in trying to find the best wine in a beer bar.

If they don't do a good job selecting things then why are you there?!

I want a WINE that runs all the Windows programs in Windows without a hiccup (Adobe Suite included) but then again as a HACKER WINE means Wine Is Not an Emulator.

https://www.winehq.org