Red · Toscana · Italien
Brancaia Tre
Scored from 7,357 calibrated Vivino reviewers, each re-centered on their own rating baseline before the bottle is ranked. Cohort: Red · Italien (289 wines).
Find this wine: view on Vivino · compare prices on Wine-Searcher
What reviewers say
“Brancaia Tre 2016 opened in 2023. Lovely Toscana blend of Sangiovese (80.0%), Merlot (10.0%) & Cabernet Sauvignon (10.0%) aged in oak casks (2/3) and stainless steel tanks (1/3). Deep crimson color with viscous legs.”
From a top-rated Vivino review for this bottle.
Summary
Brancaia Tre is a blend of 80% Sangiovese and 20% Merlot and Cabernet Sauvignon. It has fresh cranberry-cherry floral fruit nose flecked with liquorice and tobacco.
Brancaia Tre is an Italian red from Toscana. The cheapest 750 ml listing found for it is $18.00, which puts the bottle in the $15–30 band. The grape is Sangiovese.
The cohort it is ranked inside holds 289 Italian reds. The calibrated figure is built from 7,357 reviewers whose rating history varies enough to place their own average. A broader pool of 7,534 reviewers with two or more ratings feeds the raw-average track instead.
How the score was built
We compare the reviewers of this bottle against every other wine those same people reviewed. Each review is scored relative to the reviewer's own baseline — four stars from someone who hands out fives means something very different from four stars from someone who rarely goes above three. Every reviewer then counts the same, so the result is a calibrated consensus rather than a star average.
- Global percentile
- Where Brancaia Tre lands against every wine we rank.
- In-cohort percentile
- The same comparison narrowed to its cohort — wines of the same type from the same country. A Chilean red is judged against Chilean reds, not against Champagne. Ranked against Red · Italien (289 wines).
- AI-adjusted percentile
- The same reviewer-normalized score, corrected for how much confidence the sample supports: thin samples are pulled toward the corpus average. Each wine keeps the share n / (n + 53) of its measured distance from the average, where n is its calibrated-reviewer count, and the result is ranked on the same scale as the global percentile.53 is the median sample size across everything we rank. Here n is 7,357.
