Open Bottle Storage Experiment:

Discovering Optimal Method to Preserve an Open Bottle of Whisky

Introduction

Many whisky lovers have opened a bottle of whisky only to enjoy a few drams, then put the rest of bottle away for another time. But how should you store the rest of that whisky? While there are many schools of thought on the subject, they come from the research done in wine or a chemistry class from ages ago. But no one seemed to know definitively, the best way to preserve the taste of a freshly opened bottle of whisky.

This is the easy to read version with more pictures and graphs. But if you would like to view the scientific journal we wrote and based this off of you can download it here. The raw data is also available for you to download it here.


Thanks!

I would like to thank Tatsu Oiye for all of his help with this project. My tasters: Michael Ries, Josh Peters, Bino Gopal, and Marshall Naiman. Also thanks to Linh Do, Dan Long, Ian McLaren, Michael Nemcik and Nettie Rose Sher.


The Setup

If we are looking for the “best” way to store an open bottle of whisky, we need to define what it means to be the “best” method. For this study “best” is the storage method that has the least impact on the taste of the whisky regardless if that impact makes the whisky taste better or worse. Ideally whisky that has been stored would be indistinguishable from a freshly opened bottle. This is because in a case that may make a particular whisky taste better, it may make a different whisky taste worse; so we are looking for zero impact.

Choosing which methods to test for was difficult since there are hundreds of thousands of permutations we could test. After we polled over a dozen prominent whisky experts and industry professionals we gathered a list of the most common storage methods and why they thought it would be the most effective method. After we analyzed each method suggested we picked the ones we thought held the most promise and tried to cover the major points of each method suggested.

    • No Action: alcohol is very stable
    • Cool Dark Place: to prevent light and temperature reactions
    • Upright: to prevent cork degradation
    • Argon Gas: to prevent oxidation
    • Decanting: to prevent oxidation by removing headspace
    • Paraffin Wax Seal: to create a barrier that would prevent oxidation and leaks
    • Doesn’t Matter Until the Bottle is ½ Empty: alcohol is pretty stable, but can oxidize given enough air and time

Off to the store to buy some whisky to test. We bought six bottles of a single barrel bourbon all from the same barrel, from the same store and from a sealed case. This would be the closest we could get to guaranteed that all 6 bottles would taste the same. There is no such thing as single barrel blended scotch so we settled on 6 bottles of a blended scotch all from the same store, from a sealed case. We chose bourbon and blended scotch to test with since they are the most widely sold whiskies in the world. One bottle of scotch and one bottle of bourbon were set aside as controls. Both were kept sealed in a temperature controlled environment with no exposure light. The other 10 bottles were opened and stored in 27 different methods.

We chose a year because we thought about how people drank and chose the one that had the most extreme yet reasonable interval of time between pours from an open bottle. We knew of people who had a special bottle of whisky that they would drink from only once a year, such as an anniversary of a wedding. See you in a year!


A Year Later

After a year we had a panel of 4 tasters for bourbon and 4 tasters for scotch do a blind tasting. They rated each specimen by filling out a scorecard. That scorecard had 3 parts following the general whisky rating format: nose, palate and finish. However instead of an overall letter grade, the taster marked where on the scale each part of a specimen would be categorized (see Table 1). This is because letter grades were too closely related to quality and we are not concerned with quality. We are concerned with differences in taste compared to the control and felt this was the best way to get tasters to concentrate on differences not preference.

Table 1. Example of a filled out scorecard.
#H (specimen label) Nose Palate Finish
Indistinguishable      
Vague hint of difference      
Barely noticeable difference only if looking hard for one   X X
Slight difference, but not much      
Noticeable difference, but pretty close X    
Noticeable difference, but passable      
Very noticeable difference      
Totally different      
I hate you making me try this      

When tabulating results each mark is assigned a point value based on the weighted conversion scale (see Table 2). Points are then aggregated and averaged together so we can analyze the data cohesively. To identify the standard deviation, we had two specimens that were stored identically and tasted blind mixed in with the rest of the other specimens while tasting. We compared the results of each identical specimens and found the difference to be 0.39 points which established our standard deviation.

When we speak about the differences in taste or score, it is the difference between a specimen and the control, and another specimen and the control. Not the direct difference between the two specimens.

Table 2. Point scale.
Mark to Points Conversion Point Score
Indistinguishable 0
Vague hint of difference 1
Barely noticeable difference only if looking hard for one 2
Slight difference, but not much 3
Noticeable difference, but pretty close 4
Noticeable difference, but passable 5
Very noticeable difference 7
Totally different 10
I hate you making me try this 14

Open Bottle Storage Experiment Chapters:

Updated: 2023.07.30