Order and fatigue.

Does the order of the glasses change what tasters report?

Status: Open
Protocol: Version 1.0, 2026-10-18
Consent: S.order.v1 (earlier data: S.order.v1.past)
Investigator: John Schuster, Drunken Unicorn Productions
Preregistration: Being posted to the Open Science Framework; the link will appear here
Ethics review: Not yet reviewed; we intend independent review before any results are published
Target: At least 252 oak-grip ratings at position 1 and 252 at position 3 or later
Runs until: The target is reached, or 2029-10-18

Why this study

Carry-over and fatigue are well known in sensory panels, which is why panels randomize serving order. Home tasters don’t. Our water research turned up a specific, untested claim: that oak grip (the drying pull of tannin) is rated higher the later a pour comes in a sitting. The same data can test whether blind accuracy drops across a night.

What taking part means

Nothing extra. Taste the way you already do. When each of your tasting sessions starts (which gives each pour its place in a night), your structure ratings for those pours (such as oak grip and heat), and the order and results of your triangle rounds on each night.

To join: in the app, turn on research contribution in Settings → Research, then open Order and fatigue under Research studies, read it, and agree. You can leave with one tap.

Hypotheses

H1 (primary). Oak-grip ratings are higher for pours in position 3 or later than for pours in position 1.
H2 (secondary). Triangle accuracy is lower for the fourth and later rounds of a night than for the first.

Design

Observational, within-taster, repeated measures across sessions. Savorant files each pour as its own session, so a pour’s place in a night is derived: a night is a run of one taster’s sessions each starting within three hours of the one before, and position is the order within that run. Triangle rounds are grouped into nights the same way, by when they were filed.

Who can take part. Adults 21 or older in the United States who use Savorant, have research contribution turned on, and have joined this study in the app.

Outcomes

Primary. Difference in mean oak-grip rating (15-point scale) between position 1 and positions 3 and later.

Secondary and exploratory.

Linear mixed model: rating ~ position (continuous), random intercepts for taster and whiskey, for each structure measure (exploratory beyond oak grip).
Triangle accuracy, first round of a night versus fourth and later rounds (mixed-effects logistic, random intercept for taster).

Sample size

At least 252 oak-grip ratings at position 1 and 252 at position 3 or later (d = 0.25, a half-point difference against an assumed standard deviation of 2.0; α = 0.05 two-sided; power 0.80). For H2, at least 392 first rounds and 392 fourth-or-later rounds (55% versus 45% correct, α = 0.05 two-sided, power 0.80).

Analysis

Primary: linear mixed model, oak grip ~ position group, random intercepts for taster and for whiskey, two-sided α = 0.05; effect reported in scale points with a 95% CI.

Data filed before a taster joined is used only if they also agree to the separate earlier-data question. It is labeled as earlier data. The primary analysis uses only data filed after joining; the same analysis including earlier data is reported as a sensitivity analysis.

When a study reports more than one secondary test, p-values are adjusted with the Holm method.

Limitations we know about

Tasters choose the order of their pours, so stronger whiskeys may tend to come later. The whiskey random effect and comparing the same whiskey tasted at different positions address this in part.

Standards every Savorant study follows

Preregistration. This page is the protocol of record, versioned and dated, and is posted to the Open Science Framework (OSF) before any study data is analyzed. Hypotheses, outcomes, sample size and analysis are fixed here in advance.
Primary and secondary outcomes. Each study names one primary outcome, tested at the stated alpha. Everything else is labeled secondary or exploratory and reported as such, with effect sizes and 95% confidence intervals rather than bare p-values.
No peeking. No interim analysis of the outcomes. The only checks before the sample size is reached are for data quality (duplicates, impossible values), and they never look at results.
Repeated measures. One taster contributes many rounds, so analyses use mixed-effects models with the taster as a random effect, or per-taster summaries, never pooled rounds treated as independent.
Exact methods for small counts. Triangle results are analyzed against the exact binomial with chance at one in three, the way ISO 4120 sets out the triangle test, adapted here for repeated measures.
Exclusions. Rounds are excluded only for the reasons listed in advance: test accounts, the founder’s accounts, rounds filed in a preview build, and duplicates. Every exclusion is counted and reported.
Missing data. Tasters who stop early are kept for every analysis their data supports. The number who stop, and when, is reported.
Deviations. Any change after data collection starts is logged on this page as a dated amendment with its reason, and reported in the paper.
Reporting. Results are reported following the STROBE guideline for observational studies, with the analysis code published alongside.
Conflict of interest. The study is run by the company that sells the app. Preregistration, publishing regardless of outcome, and public analysis code are how that interest is kept out of the result.

The consent you agree to

This is the exact text shown in the app before you join.

Amendments

None. This is the first version of the protocol.

All open studies · The evidence review · Privacy policy · palate@savorant.com