Learn beverage judging by training the three skills separately: write two-column tasting notes that isolate observation from conclusion; identify faults by cue pattern rather than single smells; score through defined weighted criteria; calibrate scores with written descriptive anchors; judge typicity as deviation from a named style's range; declare conflicts before judging; and score panels independently before any discussion. Administrative details such as formats, eligibility, and scheduling are set by the relevant organizing body and are not covered here.
Where Observation Ends and Interpretation Begins in a Tasting Note
A judging note has two layers: observation (what your senses report — intensity, descriptors, structure) and interpretation (what you conclude — quality, fault, typicity). Keep them physically separate on paper so every conclusion can be traced back to evidence.
The core discipline is deductive tasting structure: work systematically through appearance, nose, and palate, recording only what you perceive at each stage, and only afterward forming quality judgments. Compare the note 'high viscosity, pronounced citrus and wet-stone aromas, tense acidity, long finish' with the note 'rich and excellent.' The first can be checked by another taster; the second is an unverifiable impression. Blurring the layers produces scores that look confident but cannot be defended, because nobody — including you — can tell which observation the score rested on.
In practice, require every interpretive statement to cite at least two supporting observations. 'This wine appears heat-damaged' is acceptable only if your note lists the cooked-fruit character and the flattened acidity that justify it. If you cannot cite the support, mark the statement as tentative and let it influence the score less. This habit pays off twice: in scoring it creates reasoning you can audit later, and in panel calibration it lets other judges check your logic rather than having to trust your palate sight unseen. A useful rewrite drill: take an old tasting note, split it into two columns, and draw an arrow from every conclusion to its supporting observations.
Telling TCA, Oxidation, and Volatile Acidity Apart
Faults are identified by pattern, not by a single smell. TCA mutes fruit behind damp-cardboard mustiness; oxidation changes the fruit character itself; volatile acidity adds a pungent vinegar-like sharpness layered above otherwise intact fruit.
Learn each fault as a relationship between a leading cue and the fruit's behavior. Cork taint (TCA and related compounds) suppresses: the fruit seems faint or absent, and a musty, damp-cardboard or damp-basement note dominates. Oxidation transforms: fresh fruit gives way to bruised apple, dried fruit, nuts, or sherry-like character. Volatile acidity protrudes: a sharp vinegar or nail-polish note sits on top while the fruit underneath may remain recognizably sound. Spoilage yeasts such as Brettanomyces add band-aid, barnyard, or clove-like phenolic notes, whose fault status depends on intensity and context. One wine can show several faults at once, and a trace of oxidative character in a deliberately mature or oxidative style can fall within typicity — the judgment is always deviation from the expected style, not the mere presence of a cue.
Worked scenario 1. A paper tasting note reads: 'fruit noticeably muted, dominant damp-cardboard mustiness, short finish, no sharpness, no bruised apple or nutty character.' Plausible mistake: calling it oxidation because the fruit has faded. Better decision: name cork taint as the leading suspicion. The defining evidence of oxidation — transformed fruit character such as bruised apple, dried fruit, or nuttiness — is explicitly absent, while the suppressive mustiness that defines TCA-type taint is present. Why it matters: a misnamed fault invalidates your justification even if your instinct about 'something is wrong' was right, and in a panel it sends the discussion toward the wrong reference standard.
| Fault | Leading cue | Fruit behavior | Judging implication |
|---|---|---|---|
| Cork taint (TCA-type) | Damp cardboard, basement mustiness | Suppressed or absent, not transformed | Fault at perceptible levels; note cue pattern in justification |
| Oxidation | Bruised apple, dried fruit, nuts, sherry-like tones | Changed into dried or cooked character | Fault in fresh styles; possibly within style for mature or oxidative styles |
| Volatile acidity | Vinegar or nail-polish sharpness | Fruit underneath may stay intact | Judge how far the sharpness masks the fruit; describe level, not just presence |
| Brettanomyces character | Band-aid, barnyard, clove-like phenolics | Fruit present but overlaid | Weigh intensity and context; some styles tolerate low levels |
Weighted Criteria versus Holistic Impression in Scoring
Two scoring logics compete: weighted criteria assign points per attribute (appearance, aroma, palate, finish, typicity) and total them, while holistic scoring assigns one overall mark guided by criteria. They reward different habits and fail differently.
Weighted scoring forces attention to every attribute and produces arithmetic you can explain, but it has a known weakness: averaging can quietly absorb a serious fault. A sound scheme therefore includes provisions for faults — for example, a rule that a confirmed fault caps the achievable total rather than merely subtracting a few points. Holistic scoring is faster and captures a wine's overall impression, but it is much harder to defend and to calibrate, because one number hides the reasoning. Know which logic your scheme uses, and where its fault provisions sit, before you score anything.
Worked scenario 2. A flight contains wine A — heavily oaked, yet balanced, structured, with a long finish — and wine B — technically clean but short, dilute, and unremarkable. Plausible mistake: marking A down out of personal dislike of oak. Better decision: apply the criteria. A meets or exceeds expectations for intensity, structure, and length; B underperforms on concentration and finish. Record your oak preference separately as a style comment, not as a deduction. Why it matters: preference leakage — letting 'I don't like this style' masquerade as 'this wine performs poorly against the criteria' — is exactly what calibration sessions exist to expose. The two statements lead to different scores, and only the second one is a judging judgment.
Calibrating with Descriptive Anchors Instead of Raw Impressions
Calibration works by attaching scores to written descriptive anchors — for instance, what a mid-level aroma intensity actually looks like in words — and testing your scores against reference standards and other judges' anchored notes.
A score without an anchor is an opinion; a score tied to a description can be compared across judges and across days. Build anchors as short definitions: 'length of 5/10 means the finish persists briefly but without a clear after-aroma'; 'aroma intensity of 8/10 means the glass announces itself before it reaches the nose.' Panels compile anchor sheets precisely so that a '7' from one judge and a '7' from another refer to the same territory. Individually, you can use the same technique solo: never record a number without the one-line description that earned it.
Practical exercise. Taste three similar commercial beverages blind (three unoaked white wines, or three apple juices). Score each on a five-criteria mini-sheet, then write a one-line anchor for every score you gave. Expected observations: your first-pass scores will scatter; when you re-read your anchors, you will find at least one score you cannot justify from your own observations — that gap is your calibration target. Self-check rubric: (1) every score carries at least two supporting observations; (2) each anchor line matches the observations it cites; (3) your rank order of the three samples survives a second blind pass a day later; (4) you can name the specific cue that changed your mind between passes. Treat rubric scores as learning milestones, not as predictions of any official result.
Typicity: Judging Deviation from a Named Style's Range, Not a Checklist
Typicity asks whether a sample plausibly expresses its named region, grape, or style. The skill is knowing the benchmark's range and judging deviation from it — not ticking varietal characters off a fixed list.
Benchmarks are ranges, not recipes. A warm-site expression of an aromatic variety can show riper fruit and still be within typicity, while a missing structural hallmark — say, the expected acid tension in a cool-climate white — can push a wine outside it even when the wine is clean and pleasant. This is why sound schemes treat typicity as its own criterion rather than folding it into overall quality. Keep the two judgments distinct: 'atypical but well-made' and 'typical but mediocre' are different findings that lead to different scores and different panel discussions.
Worked scenario 3 (brief). A note on a sample declared as a cool-climate aromatic white reads: low aroma intensity, flat acidity, noticeable alcohol warmth, sound fruit. A checklist approach might award partial typicity credit because 'some varietal fruit is present.' The range approach asks instead whether the structural signature fits the declared style — here it clearly does not, so the typicity score drops while the cleanliness and balance scores stand on their own merits. Why it matters: when typicity and quality are conflated, one judge's score becomes incoherent — a wine loses points for a fault it does not have, or gains credit for a character it does display — and the panel loses the ability to compare notes meaningfully.
Ethics: Silent Conflicts of Interest and Uniform Application
Ethical judging rests on declared conflicts, uniform application of criteria to every sample, and confidentiality of results. The real exposure usually comes from silent conflicts — relationships you never mention — rather than open favoritism.
A conflict of interest exists whether or not bias actually occurs. Judging a flight that contains a wine from your employer, your business partner, a student you teach, or a region you consult in is a conflict; the ethical duty is disclosure and, where appropriate, recusal. Other standing duties include keeping scores confidential until the organizer releases them, and never soliciting or accepting anything of value connected to an outcome. Because silent conflicts are invisible from the outside, the discipline lives in what you do before tasting, not in what you resist during tasting.
A practical rule set you can adapt: before each event, write down every relevant relationship you hold — trade, education, investment, family — and check them against the entry list where the organizer allows it; during judging, apply the identical criteria sheet to every flight, including wines you recognize; after judging, keep all results confidential until publication. If a known wine appears and recusal is impossible, judge it with the same sheet and report the relationship to the panel chair rather than quietly adjusting your score in either direction — compensating in a wine's favor distorts the result just as much as biasing against it.
Panel Dynamics: Independent Scores First, Discussion Second
Sound panels collect every judge's scores independently before any discussion, then reconcile outlying results through re-tasting or structured talk. This preserves each judge's independence while still using the panel's collective precision.
Early discussion anchors the group: once a vocal judge names a score, the consensus drifts toward it and nobody notices the shift. Practical countermeasures include silent scoring with sealed sheets, screening scores statistically for outliers before discussion, and re-tasting any sample whose scores diverge sharply. Understand what discussion is for: not convergence by persuasion, but resolving discrepancies that trace to different anchors, missed cues, or transcription errors. A panel that debates before recording has surrendered the very independence that makes its combined score more reliable than any single judge's.
Practice and readiness checks. Run a mock panel with two or three peers on the same three samples: score silently, compare, discuss, then re-score. Expect the rank order to survive discussion while absolute scores compress; large disagreements usually trace back to undefined anchors rather than genuinely different palates, which is your signal to repeat the anchor exercise from the calibration section. Readiness checks: you can score a silent flight with consistent anchors across two separate sessions; you can name the full cue pattern for four major faults; you can rewrite a preference statement as a criteria-based statement on paper; you can state a recusal rule from memory without hesitation.
- Adaptable preparation sequence: start with two-column tasting notes and sensory vocabulary until every note separates observation from conclusion.
- Build fault pattern cards next — one card per fault with leading cue, fruit behavior, and context caveats.
- Run the three-sample anchor-writing exercise, then repeat it blind a day later to test rank-order stability.
- Study typicity as ranges: take one named style and write down its structural signature before tasting against it.
- Finish with a mock panel: silent scores, outlier discussion, re-score; repeat the cycle rather than extending any single session.