Honey Colour ↔ Antioxidant Predictor

How well does darkness predict antioxidant capacity? Data: Gheldof & Engeseth (2002) & Bertoncelj et al. (2007)

🍯
r = 0.82 — strong correlation
Colour explains ~67% of the variance in honey antioxidant capacity across 16 tested varieties. But two varieties break the pattern in opposite directions.
Pfund Colour Score (mm) vs. ORAC Antioxidant Capacity (μmol TE / 100g)
0200400600800020406080100120Pfund Colour Score (mm) →ORAC (μmol TE/100g) →r = 0.82AcaciaWildflowerBlueberryManukaBuckwheatOn/near regressionAbove (higher than colour predicts)Below — Manuka anomaly
Tap/click a dot for details. Dashed line = OLS regression. Sources: Gheldof & Engeseth 2002; Bertoncelj et al. 2007.
🎨 Colour → Antioxidant Predictor
Light Amber
65 mm Pfund
50–85 mm range
270
μmol TE/100g
predicted ORAC
Typical range: 170–370 μmol TE/100g (±1 SD from regression)
Nearest varieties by colour: Wildflower (65 mm, 290 ORAC), Blueberry (68 mm, 315 ORAC), Manuka (72 mm, 215 ORAC)
0 mm (Water White)135 mm (Dark Amber)
ORAC ≈ −143 + 6.35 × Pfund(mm)
  r = 0.82  (Gheldof & Engeseth 2002)
  prediction ± 100 μmol TE/100g (1 SD)
📏 The Colour Rule

Darker honey → more antioxidants. This is not folklore — it is data. Gheldof & Engeseth (2002) measured optical absorbance at 450 nm and ORAC across 19 honey varieties and found r = 0.82. Every 10 mm increase in Pfund score corresponds to roughly +63 μmol TE/100g additional antioxidant capacity on average. The physical mechanism: darker colour correlates with higher total phenolic content (Maillard products, flavonoids, phenolic acids), which are the primary ORAC contributors.

⚠️ The Manuka Anomaly — 99 μmol below the line

Manuka honey sits at ~72 mm Pfund (Light Amber). The regression predicts ~314 μmol TE/100g for that colour — but measured ORAC is ~215 μmol TE/100g, roughly 99 units below. Manuka is a colour-grade outlier: it looks darker than a clover (38 mm) but delivers less antioxidant uplift than its colour implies.

Why? Manuka's commercial value is entirely from methylglyoxal (MGO), which forms via DHA → MGO conversion in Leptospermum scoparium nectar. MGO provides antimicrobial activity (UMF / MGO ratings) through a mechanism entirely orthogonal to phenolic antioxidant chemistry. A UMF 20+ Manuka is not delivering 20× the antioxidants of clover — it is delivering 1.5× the ORAC of clover at 3–5× the price.

✅ The Buckwheat Exception — 177 μmol above the line

At 120 mm Pfund (Dark Amber), the regression predicts ~619 μmol TE/100g. Buckwheat measures 796 μmol TE/100g — 177 units above the line, the largest positive outlier.Fagopyrum esculentum blossom nectar carries an exceptionally high chlorogenic acid fraction (~100–200 mg/kg of ~285 mg/kg total phenolics) that delivers more ORAC than colour alone would suggest. Buckwheat is the correct choice when maximising antioxidant density per serving; its dark colour sets appropriate expectations, but the reality exceeds them.

Frequently Asked Questions
Is darker honey always more nutritious?

For antioxidant capacity (ORAC), colour is a strong predictor — r = 0.82 in Gheldof & Engeseth (2002). However, there are important exceptions. Manuka honey is amber-coloured yet has only mid-tier ORAC (~215 μmol TE/100g) because its premium is from methylglyoxal (MGO) antibacterial activity, not phenolic antioxidants. Blueberry and wildflower honeys sit above the regression line for their colour. Colour is a useful heuristic but not a guarantee.

How is the USDA Pfund colour scale measured?

The Pfund scale measures optical density (light transmission) through a honey sample at a specific depth, reported in millimetres (mm) from 0 (water white) to 114+ (dark amber). Commercially, a Lovibond Pfund grader or spectrophotometer at 450 nm is used. The USDA established 7 official grades: Water White (0–8 mm), Extra White (8–17 mm), White (17–34 mm), Extra Light Amber (34–50 mm), Light Amber (50–85 mm), Amber (85–114 mm), and Dark Amber (≥114 mm).

Why does Manuka honey score below the colour–antioxidant regression line?

Manuka honey's premium properties come from methylglyoxal (MGO), which forms from dihydroxyacetone (DHA) in Leptospermum scoparium nectar. This MGO pathway is biochemically independent of phenolic antioxidants measured by ORAC. Manuka has ORAC ~215 μmol TE/100g — typical for its Light-Amber colour (~72 mm Pfund) but far below the marketing implication. UMF or MGO certification measures the antibacterial pathway; if you want antioxidants, buckwheat (~796 μmol TE/100g) outperforms Manuka 3.7×.

What does ORAC actually measure, and is it still used?

ORAC (Oxygen Radical Absorbance Capacity) measures a food's ability to neutralise free radicals in a controlled in-vitro assay, reported in μmol Trolox equivalents per 100g. USDA removed ORAC from its consumer nutrition database in 2012 due to concerns about therapeutic overclaiming — in-vitro antioxidant capacity does not directly translate to in-vivo clinical benefit. However, ORAC remains the dominant comparative assay in peer-reviewed honey research for ranking varieties. Use it for relative comparisons between varieties, not as a clinical health claim.

Embed on your site
Free to embed. No tracking. Attribution link included.
<iframe
  src="https://rawhoneyguide.com/tools/honey-color-antioxidant"
  width="100%"
  height="1020"
  frameborder="0"
  loading="lazy"
  title="Honey Colour vs. Antioxidant Predictor"
></iframe>
Dark theme: append ?theme=dark to the src URL.
Data sources: ORAC values from Gheldof & Engeseth (2002) J. Agric. Food Chem. 50:5870–77 and Bertoncelj et al. (2007) Food Chemistry 105:822–28. Pfund colour scores are variety-range midpoints compiled from USDA Honey Colour Standards (7 grades, 0–114+ mm) cross-referenced with published spectrophotometric measurements. Regression (OLS, n = 16): ORAC ≈ −143 + 6.35 × Pfund mm. ORAC was removed from the USDA consumer nutrition database in 2012; use for relative variety comparisons only, not clinical claims. See rawhoneyguide.com/learn/methodology.