- Blind Tasting Tea: A Method That Actually Teaches
- Why one cup teaches nothing
- The comparison has to be controlled
- Can you actually tell them apart
- What each test actually answers
- Set it up in ten minutes
- The five-minute version
- What to record, and how long to wait
- What blind tasting does not do
- Building the comparison habit
- What to do with the result
- What is blind tasting?
- How many times should I repeat a blind test?
- What is a triangle test?
- Does blind tasting tell me which tea is better?
- Why weigh the leaf?
Blind Tasting Tea: A Method That Actually Teaches
Why one cup teaches nothing
Tasting a tea and deciding you like it is not blind tasting. It is a preference, and preferences are cheap. Blind tasting is a comparison: two teas, no labels, one question, can you tell them apart, and if so, how.
The distinction matters because human perception is unreliable in a specific and predictable way. Expectation shapes what you taste, and a label is an expectation. A tea poured from a tin marked “competition grade” tastes better than the same tea poured from an unmarked tin, and the difference is real to the drinker even though nothing in the cup changed. Removing the label does not make you objective. It removes one source of bias so that the others become visible.
The comparison has to be controlled
A blind comparison is only informative if the two teas differ in one variable. Two green teas from different provinces, picked in different years, brewed differently, tell you nothing about province, year or brewing, because all three moved at once. The practical way to control it is to buy pairs that are as alike as possible except for the thing you want to test. Same seller, same year, same grade, different origin. Or same origin, same year, different roast. Or the same tea brewed two ways. One variable, two cups.
The professional version of this is a standardised brew: 3 g of leaf in 150 ml of water for five minutes, served from a white porcelain set so the leaf and the liquor can both be seen. That is not how anyone drinks tea, and it is not meant to be; it is a fixed reference so that two tastings months apart can be compared.
For your own comparison, brew both at the same time, in identical vessels, with the same water and the same leaf weight. This is not fussiness. A 1 g difference in leaf weight between two 100 ml vessels is a ten per cent difference in strength, which is larger than most of the differences you are trying to detect.
Can you actually tell them apart
This is the question blind tasting is designed to answer, and it is a different question from “which do I prefer”. Sensory science separates the two: analytical tests ask whether a difference is detectable, and affective tests ask which one people like. Confusing them is why most home tasting notes are useless.
The detectable-difference question has a statistical basis. The starting assumption is that there is no difference, and the test asks whether the results are unlikely enough under that assumption to reject it. In practice that means one correct identification proves nothing, because a guess has a real chance of being right. A pattern of correct identifications across several attempts is what carries information.
The simplest version is a triangle: three cups, two of one tea and one of another, and you name the odd one out. Guessing gives you a one in three chance each time, so a single success is weak evidence and a run of successes is strong. It is also a humbling test, because most people who are confident about their palate fail it repeatedly.
What each test actually answers
| Test | Question | Guess rate |
|---|---|---|
| Single cup | Do I like this? | Not applicable |
| Paired comparison | Which is stronger, darker, sweeter? | One in two |
| Triangle | Which one is different? | One in three |
| Repeated pair | Can I tell them apart reliably? | Falls with repetitions |
Set it up in ten minutes
Weigh the leaf. Two vessels, same weight, same water temperature, same steep. If the point of the test is the tea, everything else must be identical; if the point is the brewing, then only the brewing varies and the tea stays the same. The brewing variables worth testing are few: water at 80 to 85 °C against 90 to 95, a steep of 20 to 30 seconds against two to three minutes, and 3 g against 5 g in the same vessel. Label the cups underneath, not on the rim, and have someone else pour if you can. Pouring your own cups means you know which is which, and knowing is enough to bias the result even when you are trying not to.
Taste in the same order both ways. Take a sip of A, then B, then A again. The second pass matters because the first impression of a tea is dominated by its strongest feature, and the differences that matter are usually in the finish rather than the opening. Write the answer down before you look. A judgement recorded after the reveal is not a judgement, it is a rationalisation, and it will feel just as convincing.
The five-minute version
- Two teas, one variable, same leaf weight, same water.
- Cups labelled underneath, poured by someone else if possible.
- A, then B, then A again: the second pass is where the difference shows.
- Write the answer before revealing.
- Repeat on another day before believing the result.
What to record, and how long to wait
The differences that a blind test finds are usually in the finish rather than the opening, and the finish has a time scale. Sweetness returning to the mouth after the swallow, which Chinese tasters call hui gan, typically arrives within five to fifteen seconds and fades over a minute or more. If you judge a cup only at the moment of swallowing, you will miss the part where most teas actually differ.
A practical note-taking habit follows from that. Write three things for each cup: the first impression, the impression after ten seconds, and whether the finish lasted past half a minute. Those three columns separate teas that a single sip would have made look identical.
Keeping the notes across several sessions is what turns tasting into knowledge. A page of notes with a date, a leaf weight and a ratio on it is worth more than a memory of having liked something, because it can be compared with the next page.
What blind tasting does not do
It does not tell you which tea is better. Detectability and quality are different properties, and a tea can be distinctive in a way that is unpleasant. A blind test that ranks two teas by preference is an affective test, which is legitimate but answers a different question and is far more variable between people.
It also does not work well on teas you have never tasted. If both cups are unfamiliar, you are comparing two unknowns and the vocabulary you reach for will be borrowed from whatever you have read. The method works best on teas you know, where the question is whether a claimed difference is real. And it does not scale cheaply. Three cups is a ten-minute exercise; a meaningful result across several repetitions is an afternoon. That is why professional panels exist and why home tasting usually settles for a pair and a guess.
Building the comparison habit
The reason blind tasting is worth the trouble is that it builds a private reference library faster than any amount of reading. Two cups a week for a season is around twenty-five comparisons, and each one leaves a fact you actually tested rather than a claim you accepted.
Start with pairs you expect to differ. A cheap and an expensive Longjing, a roasted and an unroasted oolong, a fresh and a three-year-old white tea. If you cannot tell those apart, the problem is the method or the brewing, not your palate, and it is worth fixing before moving on to subtler pairs. Then move to pairs you suspect are the same tea sold at two prices. That is where blind tasting earns its keep commercially, because it answers a question that no amount of reading reviews can settle.
What to do with the result
If you cannot tell two teas apart, the honest conclusion is that the difference does not matter to you, and the cheaper one is the right purchase. That is a useful outcome and it saves money. It is also the outcome people resist most, because it contradicts the story they were told about the expensive one.
If you can tell them apart reliably, the next question is which feature you are detecting. Naming it is what turns a successful test into knowledge: the older tea has a rounder finish, the roasted one has a longer aftertaste, the higher grade is sweeter at the second steep. A result you cannot describe is a result you cannot use. How to build the vocabulary for describing what you find is covered in how to judge tea quality. The brewing variables that must be held constant during a test are set out in water temperature.
What is blind tasting?
A comparison of two or more teas without labels, asking whether a difference is detectable.
How many times should I repeat a blind test?
More than once. A single correct answer is within the range of guessing.
What is a triangle test?
Three cups, two of one tea and one of another; you name the odd one out.
Does blind tasting tell me which tea is better?
No. Detectability and preference are different questions, tested differently.
Why weigh the leaf?
A 1 g difference between two 100 ml vessels is a ten per cent strength difference.
