Why the Scoville Scale Ignores 90% of What It Measures

Written By Timothy Kavarnos

Quick Scope

The Scoville scale isn't wrong — it's incomplete. HPLC testing detects 20+ capsaicinoids but the Scoville formula counts only two or three — capsaicin, dihydrocapsaicin, and (in most labs) nordihydrocapsaicin. The other 17+ compounds — the ones that determine where heat hits your mouth, how fast it builds, how long it stays — get measured and then discarded.

Ohio State researchers discovered three natural pungency suppressors inside chili peppers that modulate perceived heat without changing capsaicin levels at all. Add capsaicin's documented 45% aroma amplification effect and the research conclusion becomes clear: a single concentration number cannot predict the experience it claims to represent.

Salamander designs for perception, not just concentration: habanero and jalapeño in ratios calibrated for front-to-mid mouth heat that enhances flavor instead of overwhelming it. Original tests at 7,300 SHU — nearly twice Tabasco — using fresh peppers and real bourbon, no extract, no powder. The number tells you concentration. The experience tells you everything else.

In This Guide

The science behind capsaicinoid perception — 20+ compounds, suppressor molecules, mouth-location mapping, and aroma amplification — explaining why identical Scoville numbers create completely different heat experiences, and why that gap matters for how sauce is made.

The Scoville scale is supposed to tell you how hot a pepper is. What it actually measures is the concentration of two or three capsaicinoids — nordihydrocapsaicin, capsaicin, and dihydrocapsaicin per the ASTA standard used by most labs — in a dried, powdered sample, expressed as how many times that extract needs to be diluted before heat becomes undetectable. Even that number only applies to the dry powder. Add water, add other ingredients, use fresh peppers instead of dried — and the number shifts before you've changed a single capsaicin molecule. It doesn't tell you where you'll feel the heat, how quickly it arrives, how long it stays, or why two peppers with the same score can feel completely different. That's not a flaw in the measurement — it's a limitation of reducing multidimensional chemistry to a single number.

By Timothy Kavarnos, Founder | Salamander Sauce Company

Key Takeaways

The Fact: The Scoville scale counts 2–3 of 20+ capsaicinoids — nordihydrocapsaicin, capsaicin, and dihydrocapsaicin per the ASTA standard used by most labs (including Southwest Bio-Labs). The other 17+ compounds get detected by HPLC and discarded from the final score.

The Data: Different capsaicinoids activate different mouth locations at different speeds. Nordihydrocapsaicin creates immediate front-mouth heat that fades in 1–3 minutes. Homodihydrocapsaicin produces delayed throat burn lasting 5–15 minutes. Same total capsaicin — wildly different experience.

The Insight: Heat perception isn't just concentration — it's capsaicinoid ratios, suppressor compounds, and capsaicin's 45% documented aroma amplification effect. A jalapeño and a habanero don't just differ in intensity. They differ in composition.

The Scoville scale isn't wrong. It's incomplete. And the gap between what it measures and what you experience tells you more about how the hot sauce industry chose to simplify heat than about how peppers actually work.

What the Scoville Scale Actually Measures

The original Scoville test (1912) measured how much sugar water it took to dilute a pepper extract until tasters couldn't detect the heat. Subjective, inconsistent, but directionally useful for a pharmacy problem Wilbur Scoville was trying to solve.

Modern Scoville uses HPLC (High-Performance Liquid Chromatography) to measure capsaicinoid concentration. The machine identifies 20+ different capsaicinoids. Then the formula ignores most of them.

The HPLC Scoville formula counts two or three compounds:

  • Capsaicin (weighted at 1.0x)
  • Dihydrocapsaicin (weighted at 0.82x)

The other 18+ capsaicinoids — nordihydrocapsaicin, homodihydrocapsaicin, homocapsaicin, and everything else the machine detects — get measured, recorded, and discarded. As Wikipedia's Scoville scale entry confirms: "spicy compounds other than the two most important capsaicinoids are ignored, despite the ability of HPLC to measure these other compounds at the same time." The number reflects concentration of two molecules. Not the composition of all of them.

That would be fine if all capsaicinoids behaved the same way in your mouth. They don't. A 2026 systematic review examining the relationship between quantitative capsaicin measurements and sensory effects found no direct relationship between capsaicin levels and reproducible sensory measurements. The concentration is measurable. The experience isn't predictable from the number alone.

The Capsaicinoid Profile: Why Ratios Matter

Different peppers produce different ratios of capsaicinoids. A jalapeño is capsaicin-dominant with meaningful nordihydrocapsaicin content. A habanero skews more heavily toward capsaicin with minimal minor capsaicinoids. Same Scoville score doesn't mean same composition — and composition determines the experience.

Krajewska and Powers mapped this in 1988. Their research, published in the Journal of Food Science, tested isolated capsaicinoids on human subjects and documented the specific sensory characteristics of each compound — where it activates in the mouth, how quickly, and how long it lasts.

The Mouth-Location Map (Krajewska & Powers, 1988)

Nordihydrocapsaicin: Front of mouth and palate. Immediate onset (0–5 seconds). Described as "least irritating" with a "mellow warming effect." Heat sensation develops immediately and recedes rapidly — 1–3 minutes total.

Capsaicin & Dihydrocapsaicin: Mid-mouth, mid-palate, throat, and back of tongue. Moderate onset (5–15 seconds). "Typical" heat sensation, more irritating than nordihydrocapsaicin. Medium duration (3–5 minutes).

Homodihydrocapsaicin: Throat, back of tongue, and palate. Delayed onset — does not develop immediately. Described as "very irritating, harsh and very sharp." Heat affects these areas for a prolonged period — 5–15 minutes.

A pepper with high nordihydrocapsaicin hits fast and fades fast. A pepper with high homodihydrocapsaicin creeps up on you and won't let go. The Scoville number doesn't distinguish between them — it treats them as invisible.

The data exists. The research has been published in peer-reviewed journals for decades. Different capsaicinoids have different binding affinities to TRPV1 receptors, different rates of attachment, different activation intensities, different dissociation rates. Each creates its own characteristic heat sensation. The question isn't whether food scientists know that heat perception is more complex than a single number. The question is why the supply chain is built to ignore it.

Mouth-Location Mapping: Where You Feel the Heat

You don't feel capsaicin uniformly across your mouth. Different capsaicinoids activate TRPV1 receptors in different locations at different intensities — and that architecture shapes everything about how you perceive a sauce.

Front-mouth heat (nordihydrocapsaicin): Immediate, accessible, approachable. This is the heat that makes jalapeños friendly — you feel it right away, and it doesn't overstay. It's heat that invites rather than demands.

Mid-mouth heat (capsaicin, dihydrocapsaicin): The standard burn. What most people think of as "hot sauce heat." Moderate onset, moderate duration. The baseline that Scoville was designed to capture.

Throat heat (homodihydrocapsaicin, homocapsaicin): Delayed, harsh, prolonged. This is the heat that builds. The kind that makes you reach for milk five minutes after the bite. The kind that can still be present when the meal is over.

Two peppers with identical Scoville scores can have completely different capsaicinoid profiles — meaning one gives you front-loaded heat that integrates with food while the other gives you delayed throat heat that sits on top of it. That information exists in the capsaicinoid composition. The number collapses it into a single value and discards the distinction.

The Suppressors: Why Same Capsaicin = Different Heat

Ohio State researchers discovered something unexpected in May 2025: chili peppers contain compounds that suppress perceived spiciness. The research originated from a consistent observation — some peppers with high measured capsaicinoid concentrations didn't taste as hot as their chemical profile suggested they should.

To test this systematically, Dr. Devin Peterson's team standardized ten different pepper varieties to contain exactly 800 SHU of capsaicinoids and presented them to a trained taste panel. Despite identical SHU ratings, tasters reported wide variability in perceived heat. Other metabolites — non-capsaicinoid compounds — were influencing the outcome.

Three Natural Heat Blockers

Using high-resolution mass spectrometry and nuclear magnetic resonance (NMR), the team isolated three compounds that effectively dull capsaicin's fiery sensation without altering capsaicinoid concentration:

Capsianoside I (Glycoside): Naturally occurring in Capsicum fruits. Reduces perceived intensity by altering TRPV1 receptor signaling.

Roseoside (Glycoside): Found in various botanical sources. Non-flavor-altering heat reduction.

Gingerglycolipid A (Glycolipid): Found in peppers and ginger. Anti-inflammatory properties. Works via competitive antagonism at the TRPV1 receptor site.

These compounds don't impart discernible flavor or aroma when dissolved in water. They work by targeting the neurochemical pathways of heat perception — reducing capsaicin's binding affinity to TRPV1 receptors or downregulating receptor expression.

Two peppers with identical capsaicinoid levels can feel completely different if one has higher suppressor content. The Scoville scale doesn't measure suppressors. It can't account for them. It was designed before anyone knew they existed.

This is the most significant challenge to the Scoville scale's predictive value since its inception — demonstrating that a pepper's experienced heat is a balance between pungent alkaloids and internal suppressors. An industry paying attention to the full chemistry would be asking different questions about how sauce gets made.

How Capsaicin Amplifies Flavor

Capsaicin doesn't just create heat. It amplifies aroma perception by 45%. Not metaphorically — measurably. Yang et al. (2021) used real-time mass spectrometry with 15 participants, testing aqueous solutions containing 3-methylbutanal (a nutty aroma compound) with and without capsaicin.

The finding: Capsaicin didn't change the actual concentration of aroma compounds released from the solution. It changed how intensely people perceived them. Aroma perception ratings were 45% higher in capsaicin-containing solutions (p < 0.0001).

The Physiological Mechanism

Two drivers explain the amplification:

  • Capsaicin enhanced average saliva flow by 92% (p < 0.0001). Saliva carries flavor molecules and facilitates their interaction with taste and olfactory receptors. Participants with higher stimulated saliva flow reported significantly higher aroma ratings.
  • Trigeminal nerve stimulation heightens sensitivity in the orbitofrontal cortex to odor and taste signals. The integration of trigeminal sensations with taste and aroma perception in the brain enhances other sensory attributes — including the ones that have nothing to do with heat.

This is why hot sauce makes food taste more like itself. The capsaicin enhances your perception of the flavors already present — it doesn't add them, it amplifies them. Good sauce leverages this. Bad sauce fights it. A sauce built only for maximum Scoville — minimum everything else — misses the entire point of what capsaicin actually does to a dish. That distinction is what separates good sauce from loud sauce.

Heat Designed for Perception, Not Just Concentration

Salamander sauces use habanero and jalapeño in ratios designed for front-to-mid mouth heat that enhances flavor instead of overwhelming it. No Scoville trophy chasing. Just vegetables, real bourbon, and the right capsaicinoid composition.

Shop All Three Sauces

Jalapeño vs Habanero: A Case Study

A jalapeño averages 2,500–8,000 SHU. A habanero averages 100,000–350,000 SHU. But the difference isn't just "more heat." It's different heat — and that comes from capsaicinoid composition, not just concentration.

Capsaicinoid Profile Differences

Jalapeño (C. annuum): ~64% capsaicin, ~30% dihydrocapsaicin, ~6% nordihydrocapsaicin. Ratio: approximately 2:1 capsaicin to dihydrocapsaicin. Higher minor capsaicinoid content — more front-mouth character.

Habanero (C. chinense): ~73% capsaicin, ~25% dihydrocapsaicin, ~2% nordihydrocapsaicin. Ratio: approximately 3:1 capsaicin to dihydrocapsaicin. More heavily capsaicin-dominant with minimal minor capsaicinoids.

The habanero's higher capsaicin dominance (73% vs 64%) correlates with delayed onset and prolonged duration. The jalapeño's higher nordihydrocapsaicin content (6% vs 2%) contributes to its more immediate, front-of-mouth heat that fades cleanly.

Jalapeño Heat Profile

  • Dominant compounds: Capsaicin (64%) + higher nordihydrocapsaicin (6%)
  • Onset: Immediate (0–5 seconds)
  • Location: Front to mid-mouth
  • Duration: 1–3 minutes
  • Character: Clean, approachable, fades cleanly — heat that integrates

Habanero Heat Profile

  • Dominant compound: Capsaicin (73%) + minimal minor capsaicinoids (2%)
  • Onset: Delayed (15–45 seconds)
  • Location: Mid-mouth to throat
  • Duration: 5–15 minutes
  • Character: Fruity, builds gradually, intense, prolonged — heat that demands attention

The Scoville number tells you the habanero has higher total capsaicinoid concentration. It doesn't tell you that the heat will creep up on you (delayed onset from high capsaicin dominance), hit different receptors (mid-mouth/throat), and stay longer. That information lives in the capsaicinoid profile. The scale reduces it to a single number and calls it done.

The Wet Weight Reality: Why Lab Numbers Don't Match Experience

I lab-tested my own sauces to understand the gap between Scoville measurements and actual heat perception. The results confirmed what the research predicts: the scale measures concentration in a specific matrix. It doesn't measure experience.

The Fresh Pepper Problem

Most Scoville charts online rank habaneros at 100,000–350,000 SHU. Those numbers refer to dried, powdered peppers. Fresh peppers are 85–90% water. When you test a sauce made with fresh peppers, you're testing the wet weight — the capsaicinoids dissolved in all that water. Southwest Bio-Labs' published methodology shows fresh pepper pods test at approximately 10% of their dry-powder SHU value because water dilutes the per-gram concentration.

A fresh Carolina Reaper pod tests around 130,000 SHU. The same pepper dried and powdered tests at 1.3 million SHU — 10x higher despite identical capsaicinoid content. Same pods, same chemistry. Different matrix, different number.

Does Wall Thickness Change the Gap?

A reasonable follow-up question: if a thick-walled pepper like a jalapeño leaves more dry mass after drying than a thin-walled pepper like a habanero, does that extra non-capsaicin dry mass dilute the per-gram concentration further — making the fresh-to-dry gap bigger for thick-walled peppers?

The answer is yes, but the effect is small. Running the actual geometry: a jalapeño's walls represent approximately 32% of its total volume. A habanero's thinner walls represent approximately 18%. That 14-point difference in wall proportion reduces the dry mass denominator by roughly 13% if you gave a jalapeño habanero-thin walls. The resulting SHU increase: approximately 1.15x — from ~5,000 SHU to ~5,800 SHU. Real, but not meaningful against the actual 40x gap between the two peppers.

The reason wall thickness barely moves the needle is that capsaicin is synthesized in the placenta — the white tissue anchoring the seeds — not in the walls. Peer-reviewed research confirms the placenta accounts for 65–95% of a pepper's total pungency, while the fleshy walls contribute only 2–25%. The walls are mostly structural mass. Adding more of them dilutes concentration slightly, but they weren't contributing much capsaicin to begin with.

The overwhelming driver of SHU differences between pepper varieties is capsaicin production — how much the placenta actually synthesizes — which is genetically determined and environmentally influenced by water stress, soil, temperature, and ripeness. A habanero's placenta produces roughly 40x more capsaicin than a jalapeño's. Wall structure accounts for perhaps 15% of the concentration effect at the extremes. Water content — the same 85–90% across virtually all fresh pepper varieties — explains the fresh-to-dry gap universally. Those are the variables that matter.

The exception is engineered superhots like Pepper X and the Carolina Reaper, where capsaicin synthesis extends into the pericarp walls themselves — a genetic adaptation that eliminates the dilution effect entirely and turns structural mass into an additional production site. In those varieties, more wall mass doesn't dilute concentration. It contributes to it. That's part of why the Reaper's morphology — extreme ridges, convoluted surface, maximum placental area — isn't just visual. It's functional capsaicin architecture.

The Salamander Test: Expected vs Actual

Salamander Original contains 20% habanero and 14.7% jalapeño by weight. Using dry-powder math (100K minimum for habanero), you'd expect around 21,000 SHU minimum. The lab result: 7,300 SHU. Using fresh-pod math, the expected range is 2,000–7,100 SHU. The sauce tested at the peak of that range — exactly where it should be given the fresh-pepper formulation.

Lab Results vs Fresh-Pepper Expectations

Original (20% habanero / 14.7% jalapeño):
Expected: 2,000–7,100 SHU (fresh-pepper math)
Lab result: 7,300 SHU
Context: Nearly 2x Tabasco (4,000 SHU)

Whiskey (8.68% habanero / 8.68% jalapeño):
Expected: 890–3,100 SHU
Lab result: 2,890 SHU
Context: Middle-to-high end of expected range

Tropical (5.57% habanero / 5.18% jalapeño + 8 fruits):
Expected: 570–1,990 SHU
Lab result: 762 SHU
Context: Lower end due to fruit matrix and pectin entrapment during extraction

Why Sauce SHU and Raw Pepper SHU Are Different Numbers

The gap between expected and actual isn't a testing anomaly. It's how fresh-pepper sauce testing works — and the lab that ran these tests published an explainer on exactly this in 2018. Southwest Bio-Labs wrote it for hot sauce makers getting back results that seemed low: "The standard way people test hot and super-hot peppers is as a pure dry powder. This gives the highest Scoville Heat Unit value since most of the water (about 85%) has been evaporated through drying. However, many hot sauce makers don't use powders for hot sauces."

His Carolina Reaper example makes it concrete. A customer sent in fresh red Reaper pods. The result came back at 130,000 SHU. The customer called — should it have been 1.3 million? Johnson dried the same pods and tested again. Dried: 1.3 million SHU. Same peppers. Same capsaicin. Ten times the difference, entirely explained by the 90% water content of the fresh fruit. The capsaicin didn't change. The matrix it's measured in did.

Johnson's conclusion: "This lower heat value does not make fresh peppers or sauces seem less hot to our senses when we eat them." The detector found less because fresh peppers are mostly water. Your TRPV1 receptors don't care about the water content. They respond to the capsaicin molecules that reach them — and the delivery mechanisms that get those molecules there.

Most commercial hot sauce is tested as a finished sauce — mash, vinegar, and other ingredients combined — not as reconstituted dry powder. Every SHU number on a sauce bottle is a wet-matrix number. Salamander goes further: fresh whole peppers, not mash, which means even more water in the matrix and an even larger gap from the dry-powder baseline. Johnson's reference point: "Tabasco has about 4,000 SHU." That's the right comparison for a fresh-pepper sauce — a finished sauce number, measured in the same context, from the same lab methodology. Not the raw pepper chart.

Why Online Estimation Tools Overshoot Fresh-Pepper Sauces

To understand how far the gap can reach, run the Whiskey sauce through an online heat estimation calculator — the kind built for home sauce makers. Enter the actual ingredient weights: the habanero, the jalapeño, and the full dilution base. The tool returns approximately 30,000 SHU and labels it "Very Hot."

The actual Southwest Bio-Labs result: 2,890 SHU. A 10.7x overestimate. The tool isn't broken — it's doing exactly what it was designed to do. The problem is what it was designed with. Three compounding errors, each one layered on top of the last:

The Three-Layer Overestimate

Layer 1 — Peak vs. average (1.75x inflation): The calculator uses 350,000 SHU for habanero — the absolute ceiling of the published range. The Chili Pepper Institute at New Mexico State University puts the average at approximately 200,000 SHU. Wikipedia notes that pungency values within a species can vary by a factor of 10 or more depending on seed lineage, climate, humidity, and soil. Using the peak instead of the average inflates the estimate before anything else is wrong.

Layer 2 — Dry powder vs. fresh pepper (5.0x inflation): Every SHU value in the calculator's database was measured on dried powder. Fresh habaneros test at approximately 10% of their dry powder value due to water content — per Southwest Bio-Labs' own published methodology. A calculator applying dry-powder SHU values to fresh pepper weights will overshoot by 5x on this layer alone, regardless of everything else.

Layer 3 — Formula vs. reality (1.22x): Even after correcting for fresh-pepper dilution using SBL's methodology, the estimate still runs about 22% above the actual lab result. That residual gap is the fruit matrix, ingredient interactions, and the variables no formula can predict. The SBL method narrows the gap from 10.7x to 1.22x. The remaining 22% is what the lab actually measures.

Multiply the three layers: 1.75 × 5.0 × 1.22 = 10.7x. Exactly the observed gap between the calculator's estimate and the lab result. Every number is independently sourced. The math closes perfectly because it's describing the same measurement problem from three different angles simultaneously.

Johnson named this industry problem directly in his 2018 explainer: "Some sauce makers put on their bottles or websites that their sauces are made from 1.3 million SHU peppers, but don't list the actual SHU value — further adding confusion to the actual SHU value of the sauce." The calculator inherits that same confusion. It treats the raw pepper number as if it transfers directly to the sauce. It doesn't.

Why 762 SHU Eats Hotter Than Tabasco

Tropical tests at 762 SHU. Tabasco tests at approximately 4,000 SHU — Johnson's figure, from the same lab, same methodology, same wet-matrix measurement context. Both are finished sauces tested as finished products. The comparison is valid. By the number, Tropical should be roughly one-fifth of Tabasco's heat. It doesn't eat that way.

Part of the gap is measurement. Tropical contains eight different fruits — a pectin-heavy, fiber-rich matrix that can trap capsaicinoid molecules during HPLC extraction. HPLC uses aggressive solvents to pull capsaicin from the sample matrix. High-pectin fruits physically bind capsaicinoid molecules in their fiber structure, and if the solvent doesn't fully break through that matrix, some capsaicin never reaches the detector. The machine isn't wrong — it's accurately measuring what it could extract. The physical matrix determines how much that is. Pineapple adds bromelain, a proteolytic enzyme that survives into the finished sauce. Bromelain doesn't interfere with the HPLC detector — the extraction process uses heat and solvents that would denature it well before that point. Its role is sensory: bromelain breaks down proteins in the mouth, which can affect how capsaicin binds to tissue and how intensely you perceive the heat. Lower measured SHU, potentially higher perceived intensity.

But the larger part is perception. Tabasco leads with vinegar. Distilled vinegar is the first ingredient — it front-loads the palate and competes with heat perception rather than amplifying it. Tropical leads with vegetables and fruit. Eight fruits create a layered matrix that does the opposite: it amplifies. The capsaicin from jalapeño hits the front of the mouth first (nordihydrocapsaicin content, immediate onset, 1–3 minutes). The habanero arrives at the throat 15–45 seconds later and stays for 5–15 minutes. Sequential activation from two different compounds at two different locations reads as more intense than a single uniform burn — even at lower total concentration.

Add the 45% aroma amplification effect: capsaicin enhances perception of the tropical fruit aromatics through increased saliva flow and trigeminal sensitization, while the fruit aromatics make the heat feel more integrated rather than isolated. The citric acid from the fruits increases TRPV1 receptor sensitivity, making the same capsaicinoid dose register more intensely. The bourbon enhances the solubility and rapid release of capsaicinoids on the palate — capsaicin is alcohol-soluble, and even at finished-sauce concentrations that effect on flavor release is real.

762 SHU is what the HPLC detector found in the matrix. It's an accurate measurement of capsaicin concentration in the final product. It is not a prediction of what happens when you eat it. The Scoville scale was designed to measure concentration. It was never designed to predict the architecture of the experience — where the heat lands, how long it stays, what it does to everything else on the plate. That's the gap this entire post has been describing. Tropical is just the clearest single example of it.

The Fruit Matrix Problem

Tropical tested lowest despite containing the same pepper varieties. The difference: eight different fruits creating a pectin-heavy, fiber-rich matrix. HPLC testing requires extracting capsaicinoids from the sauce using a solvent — high-pectin fruits can trap capsaicinoid molecules in the fiber structure. If the extraction isn't sufficiently aggressive, some capsaicin remains bound to the pulp and never reaches the detector.

Additionally, pineapple contains bromelain, a proteolytic enzyme that survives into the finished sauce. Bromelain doesn't interfere with the HPLC detector — the extraction process uses heat and solvents that would denature it well before that point. Its effect is sensory: bromelain breaks down proteins in the mouth, which can affect how capsaicin binds to tissue and how intensely the heat is perceived. It's a perception modifier, not a measurement one.

Why 762 SHU Doesn't Feel Like 762 SHU

Tropical tests at 762 SHU — technically "mild" by the scale. But people eating it don't experience mild heat. They experience layered complexity the scale can't capture:

  • Phased burn architecture: Jalapeño creates immediate front-mouth heat (nordihydrocapsaicin content). Habanero delivers delayed throat bloom (capsaicin dominance). Sequential activation feels more intense than the sum of parts.
  • Bourbon and capsaicinoid release: Capsaicin is highly alcohol-soluble. The bourbon enhances the solubility and rapid release of capsaicinoids on the palate, increasing the speed and intensity of flavor perception beyond what concentration alone predicts.
  • 45% aroma amplification: The capsaicin enhances perception of tropical fruit aromas through increased saliva flow and trigeminal sensitization. The heat makes the fruit taste more intense — and the fruit makes the heat feel more integrated.
  • Ester resonance: Habaneros contain hexyl and methyl esters chemically similar to the ethyl hexanoate in pineapple and mango. The brain experiences these as unified rather than separate, creating flavor complexity that registers as heat-intensity.
  • Acid sensitization: Citric acid from fruits increases TRPV1 receptor sensitivity, making the same capsaicinoid dose feel more pronounced.

The machine measures 762 SHU. The human experiences something closer to 5,000+ SHU in terms of perceived complexity — through mechanisms the scale doesn't measure. Understanding heat location and duration changes how you think about how to pair a sauce with food.

I came from restaurant work — wine pairing, describing dishes to customers, explaining why a sauce elevated one protein and overwhelmed another. Heat was never the point in that context. Heat was one element in a composition. When I started making sauce, I wasn't thinking about Scoville numbers. I was thinking about whether the heat enhanced the flavors or fought them. Designing for perception instead of concentration is the decision at the center of what makes Salamander different — location, duration, suppression, amplification. Concentration is just one variable, and it's the one everyone measures.

Why Milk Actually Works (It's Not Just Fat)

Everyone knows milk helps with spicy food. Most people attribute it to fat solubility — capsaicin is lipophilic, fat dissolves it. That's part of it. But Penn State research by Farah, Hayes & Coupland (2023) found something more specific: milk proteins physically bind capsaicin molecules.

Casein proteins sequester capsaicin, reducing the free capsaicin concentration available to activate your TRPV1 receptors. It's not just washing it away — it's physically grabbing the capsaicin molecules and holding them.

This is why skim milk works better than water. The protein content matters more than the fat content. Water spreads capsaicin around. Milk sequesters it.

Why Capsaicin Burns Skin (And How to Stop It)

Capsaicin is lipophilic — fat-soluble. It penetrates skin within seconds and activates TRPV1 receptors in your epidermis, triggering the same pain response it causes in your mouth. Touching a cut habanero and then rubbing your eyes creates immediate, intense burning because the same neurochemical mechanism is activating in a much more sensitive location.

The problem compounds: capsaicin doesn't wash off with water. Water is polar. Capsaicin is non-polar. They don't interact meaningfully. Rubbing water on capsaicin just spreads it.

What Actually Works for Skin Burns

  • Dish soap + vinegar — Surfactants break down capsaicin's oil structure while vinegar's acidity helps dislodge it. Scrub for 30+ seconds, then rinse.
  • Dish soap alone — Surfactants break down capsaicin's oil structure. Scrub for 30+ seconds.
  • Rubbing alcohol (70% isopropyl) — Dissolves capsaicin on contact. Apply, wait 10 seconds, wipe off, repeat.
  • Vegetable oil followed by soap — Oil dissolves capsaicin (like dissolves like), then soap removes the oil.
  • Milk (for eyes/mouth) — Don't use alcohol near your eyes. Milk proteins bind capsaicin. Flush with whole milk.
  • Baking soda paste — Alkaline compounds partially neutralize capsaicin. Mix with water, apply, wait 5 minutes, wash off with soap.

What doesn't work: Water alone (spreads capsaicin), ice (numbs temporarily but doesn't remove it). Prevention: wear nitrile gloves when handling peppers above 100,000 SHU. Capsaicin can remain on your hands for hours after handling — which is why people touch their face long after they've finished cooking and suddenly experience burning.

What Fermentation Does to Heat Perception

Capsaicinoids are remarkably stable during fermentation. The concentration doesn't change significantly. But perceived heat often does — fermented sauces frequently taste less aggressive than fresh-pepper sauces with the same Scoville score.

The mechanism isn't capsaicinoid degradation. It's matrix complexity. Fermentation produces organic acids, volatile aromatics, and flavor compounds that create a more integrated sensory experience. The heat is still present. It's embedded in a larger composition instead of isolated at the front of the palate.

That's why Tabasco — three years in oak barrels — doesn't taste like raw cayenne pepper mash even though capsaicinoid levels are similar. Fermentation mellows perceived heat without changing underlying concentration. The full story of what months of fermentation actually change in a sauce's sensory profile is more complex than the Scoville number suggests.

The Extremes: Pepper X and Superhot Territory

Pepper X — officially recognized by Guinness World Records in 2023 as the world's hottest pepper — averages 2.69 million SHU with peaks reaching 3.18 million. That's above most pepper spray formulations (2–5 million SHU) and approaching pure capsaicin territory (16 million SHU).

Ed Currie bred Pepper X over a decade through structural optimization. The pepper's extreme ridges and curves maximize placental tissue surface area — since capsaicinoids are synthesized in the placenta (not the seeds), this morphological adaptation creates more synthesis space per pepper.

Superhot Pepper Rankings

  • Pepper X: 2.69M average, 3.18M peak (2023 record holder)
  • Apollo Pepper: ~2.5M estimated (Pepper X × Carolina Reaper cross)
  • Carolina Reaper: 1.64M average, 2.2M peak (previous record)
  • Trinidad Moruga Scorpion: 1.2M average, 2.0M peak

At these concentrations, capsaicinoids become a physical hazard. Currie described eating a whole Pepper X as causing immediate, brutal heat lasting three and a half hours, followed by severe abdominal cramps that laid him out for approximately an hour. The capsaicinoids penetrate skin within seconds and trigger neurogenic inflammation. Pepper X isn't sold for raw consumption and exists primarily in heavily diluted commercial extracts.

How I Approached This Research

I started with a question that seemed obvious but turned out not to be: why do some peppers burn so differently from others at similar Scoville scores? The obvious answer is "different concentrations." But a jalapeño and a serrano overlap in SHU range and feel completely different on the palate.

I went through the HPLC methodology to understand what the Scoville scale actually counts — the ASTA formula weighting capsaicin at 1.0x and dihydrocapsaicin at 0.82x, discarding the other 18+ detected capsaicinoids. That led to Krajewska and Powers' mouth-location mapping research from 1988. Then to the Ohio State pungency suppressor study from May 2025 (Capsianoside I, Roseoside, Gingerglycolipid A). Then to Yang et al.'s 2021 work on aroma amplification — 45% enhanced perception, 92% increased saliva flow.

Each paper answered one question and raised three more. What emerged across all of it: heat perception is multidimensional. The Scoville scale measures one dimension accurately. It ignores the others completely. That's not a flaw in the measurement — it's a limitation of reducing complex chemistry to a single commercial number. A January 2026 systematic review confirmed what the individual studies already suggested: researchers found no direct relationship between capsaicin levels in food and reproducible sensory measurements.

Sources

The Bottom Line

The Scoville scale was designed in 1912 to standardize capsaicin concentration across batches — a pharmaceutical problem. It does that job accurately. But heat perception isn't just concentration. It's composition, ratios, location, duration, suppression, amplification. The scale accurately measures one variable. It ignores the rest.

The science just tells you the number is incomplete. The harder question — the one the data can't answer — is why an industry with access to all of this research chose to simplify heat into a single number and sauce into a single formula. That's not a measurement problem. That's the industry decision that simplified everything about how hot sauce gets made.

The number tells you how loud the heat is. It doesn't tell you where it lands, how long it lasts, or what it does to everything else on your plate. Those are the questions that matter when you're designing a sauce to work with food instead of just registering on a scale.

The Salamander Standard

When we set out to make a better hot sauce, we refused to compromise. Here's what we measure ourselves against — and what every bottle delivers:

  • ✓ Flavor and fire working together to elevate your food
  • ✓ Vegetables or fruit first, not vinegar and water
  • ✓ 10+ flavor ingredients vs. 2–4 in mainstream brands
  • ✓ Ingredients you recognize — see the full list
  • ✓ Made in small batches in New York’s Hudson Valley
  • ✓ 50mg or less sodium per serving
  • ✓ No xanthan gum or artificial thickeners

Every bottle. Every batch. Since 2012. See exactly what’s in each sauce →

Frequently Asked Questions

Is the Scoville scale accurate — and what does it reliably measure vs. what it actually can't?

Accurate for what it was designed to measure: capsaicin and dihydrocapsaicin concentration in a dried sample via HPLC. That part it does precisely. What it can't do is predict the eating experience. HPLC detects 20+ capsaicinoids but the Scoville formula counts only two or three — nordihydrocapsaicin, capsaicin, and dihydrocapsaicin per the ASTA standard, with some labs using only capsaicin. The other 17+ get measured and discarded. It doesn't account for capsaicinoid ratios that determine where and how long heat hits, three natural pungency suppressors that modulate perceived intensity without changing concentration, the fresh vs. dried measurement gap that makes raw pepper SHU and sauce SHU incomparable numbers, or capsaicin's documented 45% aroma amplification effect. The number is precise. What it predicts about the experience is limited in ways most people don't realize.

Why do jalapeños and habaneros feel so different — is it really just more capsaicin, or is the composition completely different?

Completely different composition, not just more capsaicin. Jalapeños (C. annuum) run a 2:1 capsaicin-to-dihydrocapsaicin ratio with meaningfully higher nordihydrocapsaicin content (6% vs. 2%). Habaneros (C. chinense) are far more capsaicin-dominant at 3:1 with minimal minor capsaicinoids. That difference in composition produces different heat in different locations at different speeds: jalapeño creates immediate front-mouth heat (0–5 seconds) that fades in 1–3 minutes, driven by its nordihydrocapsaicin content. Habanero delivers delayed mid-mouth to throat heat (15–45 seconds onset) that persists for 5–15 minutes, driven by capsaicin dominance. Same scale. Completely different architecture. The Scoville number captures concentration. It doesn't capture any of that.

What are pungency suppressors, how do they work, and why doesn't the Scoville scale account for them?

Pungency suppressors are naturally occurring compounds inside chili peppers that reduce perceived heat without changing capsaicinoid concentration at all. Ohio State researchers identified three in 2025: capsianoside I, roseoside, and gingerglycolipid A. They work by targeting TRPV1 receptor signaling — reducing capsaicin's binding affinity, downregulating receptor expression, or competing at the receptor site. The result: two peppers with identical capsaicinoid levels can feel completely different if one has higher suppressor content. The Scoville scale doesn't measure them because it was designed before anyone knew they existed — and because the HPLC formula is built around capsaicin concentration, not the full chemical profile that determines how that capsaicin actually registers in your mouth.

Does capsaicin make food taste better — and how does it amplify flavors that have nothing to do with heat?

Measurably, not just perceptually. Yang et al. (2021) documented a 45% increase in aroma perception in capsaicin-containing solutions vs. controls, with saliva flow increasing 92%. Capsaicin activates trigeminal pathways that heighten sensitivity in the orbitofrontal cortex — the brain region that integrates taste and smell — making everything on the plate register more intensely, including flavors that have nothing to do with heat. Good hot sauce is designed to use this. The capsaicin enhances the ingredients it's paired with rather than competing with them. A sauce built purely for maximum Scoville with minimal everything else is working against the very mechanism that makes capsaicin useful in food.

Why does milk work better than water for spicy food — and is it really about the fat?

Not primarily. Penn State research identified casein proteins as the primary mechanism — they physically bind capsaicin molecules, reducing the free capsaicin available to activate TRPV1 receptors. It's not dissolution, it's sequestration. Fat solubility plays a role, but skim milk outperforms water despite minimal fat content because the protein binding is doing most of the work. Water doesn't bind capsaicin at all — it just spreads it across a larger surface area, often making the sensation worse. The practical implication: the protein content of what you drink matters more than the fat content. Whole milk works better than water primarily because of casein, not cream.

Does fermentation reduce heat in hot sauce — and if the capsaicin doesn't change, why does fermented sauce taste milder?

Capsaicinoids are chemically stable during fermentation — concentration doesn't change significantly. But perceived heat frequently mellows, and the mechanism isn't degradation. Fermentation produces organic acids, volatile aromatics, and complex flavor compounds that create a denser sensory context. The heat is still there — it's embedded in a larger composition rather than sitting isolated at the front of the palate. Your brain processes heat differently when there's more to process alongside it. That's why Tabasco — three years in white oak barrels — doesn't eat like raw cayenne mash despite similar capsaicinoid levels. The capsaicin is the same. Everything built around it has changed.

Why do fresh-pepper hot sauces test lower on the Scoville scale — and how far off can the number actually be from real experience?

Fresh peppers are 85–90% water. The Scoville scale measures capsaicinoid concentration in a dried sample — so fresh peppers test at approximately 10% of their dry-powder equivalent. Southwest Bio-Labs documented this with the Carolina Reaper: fresh pods tested at 130,000 SHU, the same pods dried and powdered at 1.3 million — 10x difference, identical capsaicinoid content. For a finished sauce made with fresh peppers and other ingredients, the gap between what a dry-powder calculator predicts and what the lab actually returns can exceed 10x. And that lab number still doesn't tell you what the sauce tastes like — because matrix effects, capsaicinoid architecture, and delivery mechanisms all shape the experience in ways no number captures.

Why doesn't the Scoville number predict what a hot sauce will actually taste like — even when the measurement is technically accurate?

Because the number measures one variable — capsaicinoid concentration in the tested matrix — and experience is the product of many. Capsaicinoid ratios determine where heat lands and how long it stays. Three natural pungency suppressors can make a high-SHU pepper taste flat. The fresh vs. dried measurement gap means sauce SHU and raw pepper SHU are different measurement contexts entirely. Capsaicin's 45% aroma amplification effect means the heat interacts with every other flavor on the plate in ways concentration doesn't predict. A sauce testing at 762 SHU can deliver a perceived experience closer to 5,000 SHU — through mechanisms the scale wasn't designed to capture and doesn't measure.

Why does an online Scoville calculator show a completely different number than the actual lab result — and which one is right?

The lab result is right for the finished sauce. Online calculators are built on dry-powder pepper SHU values — the numbers published in charts for dried, powdered peppers. Three compounding errors explain the gap: calculators typically use peak SHU values (350,000 for habanero) rather than the average (~200,000 per the Chili Pepper Institute), they apply dry-powder values to fresh pepper weights when fresh peppers test at roughly 10% of their dried equivalent, and no formula accounts for ingredient interactions in the final matrix. Together these factors can produce a 10x overestimate. The calculator is doing what it was designed to do — it was designed for dried peppers, not fresh-pepper sauces. The lab measures the actual finished product.

How do I get capsaicin off my hands — and why doesn't water work?

Capsaicin is non-polar and fat-soluble. Water is polar. They don't interact — running water over capsaicin just spreads it across more skin surface, often intensifying the burn. What works: dish soap scrubbed for 30+ seconds (surfactants break the oil structure), rubbing alcohol — 70% isopropyl — applied, left 10 seconds, wiped off and repeated, or vegetable oil followed by soap (oil dissolves capsaicin on the like-dissolves-like principle, then soap removes the oil). For eyes or mouth, use milk — the casein proteins bind capsaicin, don't use alcohol near your eyes. Capsaicin can remain active on skin for hours after handling, which is why touching your face long after cooking still causes burning. Nitrile gloves when handling anything above 100,000 SHU.

Is Pepper X dangerous to eat — and what actually happens to your body at 2.69 million SHU?

At 2.69 million SHU average with peaks at 3.18 million, Pepper X operates at concentrations comparable to law-enforcement pepper spray. Ed Currie — who bred it — described eating a whole pepper as causing immediate, overwhelming heat lasting 3.5 hours followed by severe abdominal cramps that incapacitated him for approximately an hour. At those concentrations, capsaicinoids penetrate skin within seconds and trigger neurogenic inflammation — the nervous system's response to what it registers as a chemical burn. Pepper X is not sold for raw consumption, seeds are not publicly released, and it exists primarily in heavily diluted commercial extracts. The Scoville number is accurate. The physiological response at that concentration is what makes it a hazard rather than a condiment.

Do thick-walled peppers like jalapeños have a larger fresh-to-dry SHU gap than thin-walled peppers — and does wall structure explain the heat difference between varieties?

Wall thickness is a real variable but a minor one. A jalapeño's walls represent roughly 32% of its total volume vs. approximately 18% for a habanero. That extra non-capsaicin dry mass slightly dilutes the per-gram concentration — modeling it out, giving a jalapeño habanero-thin walls would raise its SHU by approximately 1.15x, from ~5,000 to ~5,800 SHU. The actual gap between the two peppers is 40x. Wall structure explains almost none of it. The reason: capsaicin is synthesized in the placenta, not the walls. Peer-reviewed research confirms the placenta accounts for 65–95% of a pepper's total pungency while the walls contribute only 2–25%. The overwhelming driver of SHU differences between varieties is how much capsaicin the placenta actually produces — genetically determined, environmentally influenced. The exception is engineered superhots like Pepper X and the Carolina Reaper, where capsaicin synthesis extends into the pericarp walls themselves — a genetic adaptation that eliminates the dilution effect and turns structural mass into an additional production site. For normal culinary peppers, water content (~85–90% across all fresh varieties) explains the fresh-to-dry gap universally.

Try Heat That Enhances, Doesn't Overpower

Salamander sauces are designed for front-to-mid mouth heat that amplifies flavor instead of masking it. Real vegetables, real bourbon, no vinegar base, no xanthan gum. 25–50mg sodium per serving because fresh ingredients don't need salt as a preservation shortcut.

Shop Salamander Sauce

About Timothy Kavarnos

Timothy Kavarnos is the founder of Salamander Sauce Company, a Brooklyn-based hot sauce maker focused on fresh vegetables, real ingredients, and low-sodium formulations. Over fifteen years ago, he started making hot sauce in his kitchen because he couldn't find one that brought flavor and fire together. Today, Salamander Sauce is made in New York's Hudson Valley using the same recipes — fresh habaneros, real bourbon, and vegetables instead of vinegar. Timothy writes about hot sauce, ingredients, and flavor science on the Salamander Sauce blog.

The number tells you concentration. The experience tells you everything else.

Timothy Kavarnos https://www.salamandersauce.com
Previous

Hot Sauce and Health Conditions: What You Actually Need to Know
Next

Hot Sauce Fermentation and Sodium: Why Tabasco Has 35mg While Frank's Has 190mg