Grape confusion pairs are not only structure walls, oak stance, Syrah pepper-smoke, Sauvignon citrus, or floral perfume. Studies 1a–1l ranked mutual planting weight, neighborhoods, one-way hubs, identity traps, planting-asymmetric mutuals, inbound magnets, color-frame stop rules, oak-led separators, dual-structure forks, pepper-smoke aroma tells, citrus-led white forks, and floral-perfume cues. Study 1m cuts the savory object: savory-led separators — first cues that name gamey savory depth, red-pepper savory tone, earthy savory edge, savory herbs, or Mediterranean herbal as the deciding fork.

Headline finding: under default filters (combined planting rank ≤75; identity, color-frame, oak-led, dual-structure, pepper-smoke, citrus-led, and floral-perfume excluded), the catalog yields 10 savory-led edges (2 mutual). Two magnets dominate: Carmenère (red-pepper savory) against Merlot, Malbec, Zinfandel, and Cabernet Franc; Mourvèdre (gamey savory depth) against Grenache, Zinfandel, and Tannat. A Mediterranean/Italian close adds Montepulciano’s earthy savory, Nero d’Avola’s Mediterranean herbal, and Sangiovese’s savory herbs. These are the edges where candidates invent “dark spicy red” or collapse Carmenère into Syrah pepper-smoke without logging the savory stance the catalog already printed.

This is still catalog-graph research, not population miss-rates (Study 2 / topic `137` when volume clears). Companions: 1a · 1b · 1c · 1d · 1e · 1f · 1g · 1h · 1i · 1j · 1k · 1l. Method: Carmenère vs Merlot · Mourvèdre vs Syrah.

Methodology (read this before citing)

FieldValue
SourceSensium `confusionPairs` — first `separators[0]` string
MatchFirst separator matches gamey savory / red-pepper savory / earthy savory / savory herbs / Mediterranean herbal cues
ExcludeIdentity-trap, color-frame, oak-led, dual-structure, pepper-smoke, citrus-led, and floral-perfume first separators (1d / 1g / 1h / 1i / 1j / 1k / 1l)
FilterCombined `globalPopularityRank` ≤ 75
Table below10 teaching pairs (full filtered export)
Export date2026-7 / 8 / 9
Catalog size1,534 dossiers · 10 savory-led edges under default filters (2 mutual)
Re-run`node scripts/data/export_savory_led_pairs.mjs`

What this is not: a claim candidates miss Carmenère most often in live exams, or a chemistry paper on pyrazines. It is the set of pair separators the editorial graph already writes with savory stance as the deciding aroma cue — and it deliberately excludes Syrah’s pepper-olive-smoke triad (Study 1j).

Why savory-led pairs deserve their own study

Study 1j named Syrah’s black pepper, olive, and smoky savoriness. Candidates who only know that triad still fail when Carmenère’s red-pepper savory (softer than Cabernet) or Mourvèdre’s gamey savory depth is the real fork — and they write “Syrah pepper” on both. Study 1m answers: which edges force a savory line that is not the Syrah pepper-smoke separator?

Practically: leafy Cabernet Franc (pyrazines) is also “green,” but it is not red-pepper savory. Zinfandel’s spicy-sweet is not gamey. Keep the three stop rules on one sheet: pepper-smoke · red-pepper savory · leafy lift.

Google’s 2026 non-commodity bar still applies: dated method, re-run command, clear “not telemetry” label.

The 10 teaching savory-led pairs

Separators trimmed to the catalog’s first cue. Open Compare for the full fingerprint.

Mourvèdre gamey savory depth

#PairMutualCatalog stop ruleDrill
1Grenache (#7) vs Mourvèdre (#17)Gamey savory depth + stronger tannin + darker fruit vs red-fruited warmth + spice + lower tanninMourvèdre vs Syrah · GSM literacy · Compare
2Mourvèdre (#17) vs Zinfandel (#22)MGamey savory depth + stronger tannin + darker fruit vs ripe brambly + high alcohol + spicy-sweetZinfandel vs Primitivo — spicy-sweet ≠ gamey
3Mourvèdre (#17) vs Tannat (#58)Dark fruit + very firm tannin + black-tea grip vs gamey savory depth + stronger tannin + darker fruitStructure wall vs gamey — both dark, different savory

Carmenère red-pepper savory

#PairMutualCatalog stop ruleDrill
4Merlot (#2) vs Carmenère (#32)Red-pepper savory + softer texture than Cabernet vs plum-led + softer tannin + rounder mid-palateCarmenère vs Merlot · Compare
5Malbec (#18) vs Carmenère (#32)Red-pepper savory + softer than Cabernet vs inky blackberry-plum + broad textureInky ≠ red-pepper savory
6Zinfandel (#22) vs Carmenère (#32)Red-pepper savory + softer than Cabernet vs ripe brambly + high alcohol + spicy-sweetSpicy-sweet ≠ red-pepper savory
7Cabernet Franc (#30) vs Carmenère (#32)Red-pepper savory + softer than Cabernet vs red-currant + leafy lift + medium bodyPyrazines · leafy ≠ red-pepper savory · 1e asymmetric

Mediterranean / Italian savory close

#PairMutualCatalog stop ruleDrill
8Montepulciano (#21) vs Nero d'Avola (#24)MDark plum + medium-plus tannin + earthy savory edge vs ripe dark fruit + warmth + Mediterranean herbal edgeNero vs Sangio — two savory maps
9Sangiovese (#11) vs Blaufränkisch (#49)Sour-cherry + black-pepper + bright acid vs sour-cherry + savory herbs + bright acid spineBlaufränkisch vs Syrah — pepper ≠ savory herbs
10Nero d'Avola (#24) vs Corvina (#41)Sour-cherry + almond + moderate tannin vs ripe dark fruit + warmth + Mediterranean herbal edgeItalian regions

(M) = mutual · (→) = one-way · Widen `--max-combined` for a longer export.

Three takeaways

1. Write the savory stance — not “spicy” and not “Syrah pepper”

If the separator names gamey savory depth, red-pepper savory, or savory herbs, log that line before “dark spicy red.” Syrah’s pepper-olive-smoke is Study 1j — a different object (pepper-smoke).

2. Carmenère and Mourvèdre are the savory magnets — drill the other grape

Like Syrah in 1j and Sauvignon in 1k, these two cultivars appear because the catalog uses their savory lines as contrast poles. Open Merlot, Malbec, Grenache, or Zinfandel first — then confirm on Compare.

3. Savory ≠ leafy pyrazine ≠ pepper-smoke

Cabernet Franc’s leafy lift is green/pyrazine-adjacent (pyrazines). Carmenère’s red-pepper savory is softer-than-Cabernet savory. Syrah’s triad is pepper-olive-smoke. Three stop rules — do not collapse them into “green” or “pepper.”

How to drill in two weeks

  1. Week 1 — magnets (rows 1–7). Mourvèdre vs Grenache/Zinfandel/Tannat; Carmenère vs Merlot/Malbec/Zinfandel/Franc. Write savory before fruit.
  2. Week 2 — Mediterranean close (rows 8–10). Montepulciano/Nero, Sangiovese/Blaufränkisch, Nero/Corvina. Revisit 1j so pepper-smoke and savory stay on separate lines.

A useful study habit: add a tenth stop-rule question after Studies 1d/1g/1h/1i/1j/1k/1l — (10) Is this a savory-led separator (and not pepper-smoke)? Candidates who skip it invent Syrah pepper on Carmenère/Mourvèdre — catalog-preventable.

Service note: savory attaches best to Chilean Carmenère, Bandol/Monastrell Mourvèdre, and warm Mediterranean reds (home flights; exam shopping). Structure after the savory line (structure tasting).

How the thirteen Study 1 cuts fit together

StudyCutQuestion it answers
**1a**Mutual edges × plantingWhich pairs does the product elevate globally?
**1b**Top-15 planted hubsWhat is the local four-neighbor map around Cabernet / Chardonnay / …?
**1c**One-way exam hubsWhich syllabus grapes does planting-mutual rank under-weight?
**1d**Identity separatorsWhich edges are stop rules (name / parent / synonym) before aroma?
**1e**Planting-asymmetric mutualsWhich mutual edges pair a top-20 hub with a mid-catalog twin?
**1f**Confusion magnetsWhich grapes do other dossiers point at most (inbound degree)?
**1g**Color-frame stop rulesWhich related names must stop at white / red / rosé frame?
**1h**Oak-led separatorsWhich edges are decided by oak compatibility / dill-leather / without-heavy-oak?
**1i**Dual-structure separatorsWhich edges name tannin and acid in the first separator?
**1j**Pepper-smoke separatorsWhich edges are decided by black pepper / olive / smoky savoriness?
**1k**Citrus-led separatorsWhich edges are decided by green-citrus / saline citrus / citrus-herbal?
**1l**Floral-perfume separatorsWhich edges are decided by rose / lychee / violet / perfume / rose-tar?
**1m**Savory-led separatorsWhich edges are decided by gamey / red-pepper savory / earthy savory / savory herbs?

Use them as a stack. A month that only drills 1j still fails Carmenère’s red-pepper savory vs Merlot; a month that only drills 1m still fails Cabernet vs Merlot under time. When Study 2 ships, savory-led misses are a natural telemetry slice — until then, cite this page as catalog savory-led separators.

What we will publish next (Study 2)

When anonymized Train/Blind wrong-answer volume clears a documented n threshold, Study 2 will publish miss-rates beside Studies 1a–1m. Until then, do not call these “most-missed Carmenère mistakes.” They are catalog savory-led pair separators.

Frequently asked questions

How is this different from pepper-smoke (Study 1j)?

Study 1j matches the Syrah-lane triad (black pepper / olive / smoky savoriness). Study 1m matches gamey / red-pepper savory / earthy savory / savory herbs — and excludes the 1j pattern. Carmenère and Mourvèdre can appear near Syrah in flights; the separator wording is the fork.

How is this different from Carmenère vs Merlot method post?

Carmenère vs Merlot teaches one pair. Study 1m publishes which pairs already encode savory stance in the first separator — including Mourvèdre and Mediterranean herbal edges that post does not own.

Why only ten pairs?

Prior Study 1 aroma cuts removed many edges; savory is a tighter residual set. Widen `--max-combined` for a longer export.

Can I cite this in a class?

Yes — attribute “Sensium savory-led pairs, 2026-07-09” and link this URL. Keep the “not telemetry” label.

How do I reproduce the list?

`node scripts/data/export_savory_led_pairs.mjs`


Open Carmenère vs Merlot and Mourvèdre vs Syrah this week, then force a savory line in Blind before any generic spice paragraph — and check it is not Syrah pepper-smoke. Savory first — then fruit.

← All guides