About Rankquant
The rating-inflation problem
Two reviewers can use the same rating scale very differently. One may give almost everything a high score; another reserves those scores for a few favorites. A raw average combines those different habits. Rankquant measures how each rating compares with that reviewer's own history before combining the evidence.
Our fix: a normalized percentile, not another average
Rankquant uses individual ratings from crowd-review platforms, including Vivino, Amazon and Goodreads. No professional critic publication is currently ingested. The pipeline has four parts:
- Per-reviewer z-score normalization. For every reviewer we compute their personal mean μu and standard deviation σu, then convert each of their ratings to zu,i = (ru,i − μu) / σu. Personal scale washes out; what's left is each reviewer's opinion of relative quality.
- One unweighted aggregate per product. We average those z-scores across every qualifying reviewer. Every reviewer counts equally — no reviewer and no source is weighted more heavily than another, there is no credibility multiplier and no per-source weight table anywhere in the pipeline.
- Empirical-CDF percentile.Every product's mean z-score is ranked against other items in its vertical and expressed as a 0–100 global percentile. A 90 is around the top 10% of that reference set. The in-cohort percentile re-ranks the same z-score among closer peers. Cohort definitions vary: wines use type and country; products use their product category. Each detail page names its comparison.
- A confidence correction alongside it. The AI-adjusted percentile takes statistical uncertainty into account: a product keeps the share n / (n + 53) of its measured distance from the corpus average, so a 4-reviewer +2.1 mean is pulled almost all the way back to the average while an 80-reviewer +1.6 mean keeps most of its distance. This reduces the influence of small samples; it does not eliminate uncertainty. The adjustment lives in the AI-adjusted figure alone: the global and in-cohort percentiles report the measurement as made, on every vertical.
What the percentiles tell you
Consider a hypothetical wine at global percentile 62 and in-cohort percentile 91. It sits above much of the wine catalog, and ranks especially highly among wines of its type from its country. These numbers illustrate the distinction; they are not ratings of a named bottle. Neither figure predicts whether you will enjoy it.
A percentile describes a position in the data we have. It is not a star rating or a guarantee of quality. Review coverage, small samples and incomplete reviewer histories remain limitations. Read the sample count and the AI-adjusted figure alongside the global and peer percentiles.
Why you should trust our logic
Rankquant is led by Ryan Siegal, whose fifteen-year quantitative-finance career is the source of the methodology. Ryan founded a multi-strategy hedge fund (14% average annual return, $12M AUM), traded event-driven and merger-arbitrage strategies at Alpine Global with 25–35% annual returns, and now applies Python and AI/ML to research and systematic process improvement at Relentless Upside Consulting. Washington University in St. Louis alum. Consulting statisticians Ben, Josh, and Yang contribute STEM-trained input on distribution analysis, normalization design, and outlier detection.
The methodology behind Rankquant — z-scoring, confidence intervals, shrinkage estimation — is the same statistical toolkit that runs trading systems. None of the math is novel. What's different is publishing it openly, applying it rigorously, and committing to constants up front so the output is reproducible. We don't claim novel math. We claim applied math — taking methods that work in finance and psychometrics every day, and using them on review data where the field has been making do with raw averages for thirty years.
Everything we do is documented at /methodology. Detail pages show the available normalized scores, reviewer counts and peer comparisons. The normalization is specified in full — admission rules, the shrinkage constant and the percentile mapping — at /methodology, so any percentile we publish has an explicit definition that can be examined and contested.
Independence and disclosure
Product pages, book pages, the film/TV catalog panel and the wine pages whose exact bottle Amazon sells carry Amazon Associates tracked links, which may earn us a commission at no additional cost to you. CruiseDirect links on ship pages are tracked through Commission Junction and may also earn a commission. Every other wine link, and the hotel and Cruise Critic links, carry no tracking parameter and earn us nothing. Each category has one fixed outbound destination, built from the record's own identifiers after the score is computed — there is no price-and-commission routing. The full description is at /methodology#affiliate. No outbound link affects the normalized score. Brands cannot buy a higher ranking.
How to use this site
If you're buying something right now:
- Choose wines, films and TV, books, hotels, cruises, or products from the navigation.
- Look at the top 5 by normalized score.
- Read the methodology-transparent breakdown on the product page.
- Use the source or retailer link to check current details and availability.
If you're a brand or researcher: category datasets are available at /data. The normalization procedure is specified in full at /methodology.
If you have feedback: use the contact page to report a data or methodology error. Include the page address and the evidence behind the correction.
Our editorial policy
- Coverage and evidence thresholds differ by category. Product listings with insufficient calibrated reviewers remain unscored or excluded from search indexing.
- Sponsored content is never re-ranked. Brand partnerships are labeled clearly.
- Categories are defined by our editorial team; we do not split or merge categories to benefit any specific product.
- Our top-pick logic is reproducible: publish the same raw inputs on the same date, get the same output.
- Scores describe the published data snapshot and can change as coverage changes.
Based in New York.