The HuntGrade Standard · Methodology v1.5

Three grades. Fixed scores, relative letters. No single number.

Every anchor that produces a HuntGrade is published on this page, and every score is arithmetic you can repeat. Check our math — that's the point.

3 separate grades, never averaged Fixed score — the number never re-scores Relative grade — ranked against the field Open scale — no /100 ceiling Versioned — changes are logged, never silent
Why three grades, not one number

A composite hides exactly what you need to see

Two real scopes from the catalog. Average their three published scores into one number and they land on the same number — but they are opposite bets: a premium name with top-tier glass and handling carrying the only D warranty in the catalog, against a budget scope that's middling everywhere yet better-promised. The single number can't tell you which risk you're taking. The profile can.

Why lens size isn't graded

Lens size is a choice, not a grade

Almost every spec on a thermal has an obvious better: more resolution, longer battery life, less weight. Those go into the grades. Lens size is the exception — bigger glass picks up animals farther away, smaller glass shows you more of the field, and which one is "better" depends on whether you hunt open country or close timber. So lens size isn't included in the grade — instead, we show you exactly what each size trades away.

See farther, or see more — never both

Every thermal makes the same trade: a bigger lens picks up a coyote farther away, but shows you a narrower field of view. Each curve is one sensor class — choose a lens size and you've chosen your spot on it. A sharper sensor moves you to a better curve; nothing escapes the trade.

These yardages are our math, not the brochure's: the distance a coyote-sized animal first shows up as a warm spot on screen (textbook detection criterion, 12 µm pixels). Knowing exactly what animal you're looking at takes roughly 8× closer. Weather and sensor sensitivity can only shorten these. Lens sizes marked on the 640-class curve.

One direction of better — so it goes in the grade

No hunter wants a dimmer display or a heavier scope — there's no trade to weigh, so these specs go straight into the grades. Lens size stays out; it just moves you along the curve at left. Its one real cost is ounces: bigger glass weighs more, and the weight is included in the grade.

Image

  • sensor resolutionmore pixels is better
  • display resolutionmore pixels is better
  • sensitivity (NETD)a lower number is better

Handling

  • build & sealingtougher is better
  • battery runtimelonger is better
  • eye relieflonger is better
  • weight on the riflelighter is better

Warranty

  • published termslonger & transferable is better
How the specs make the trade

Three specs, one trade — and magnification isn't part of it

A spec sheet lists core resolution, pixel pitch, lens size, and base magnification as if they were four separate selling points. They aren't — they're one geometry. The core sets a fixed budget of pixels. Pixel pitch (the physical size of each pixel) and lens size decide how that budget is spent: across a wide field of view, or packed tight for detail at distance. Magnification just reports how big the result is painted in the eyepiece. The four steps below walk that ladder with real scopes from our rankings, measured two ways — the field of view (FOV: how wide you can see at 100 yards), and how many pixels land across a coyote at 200.

What those pixels look like

The detail column above is a number; here it is as a picture — the same coyote at 200 yards, drawn at each rung of the ladder. Every frame covers the same patch of ground; the only thing that changes is how many pixels land on the animal. Detail scales with distance: step out to 300 yards and 12 px falls to 8.

Simulated thermal view: 9 pixels across a coyote at 200 yards
9 px on the coyoteCollector 640 · 25 mm
Simulated thermal view: 12 pixels across a coyote at 200 yards
12 px on the coyotethe three-way tie · XQ50 Pro / XP50 Pro / Rattler 35-640
Simulated thermal view: 17 pixels across a coyote at 200 yards
17 px on the coyoteRattler V2 50-640
Simulated thermal view: 21 pixels across a coyote at 200 yards
21 px on the coyoteNocpix ACE S60R
Simulated thermal view: 26 pixels across a coyote at 200 yards
26 px on the coyoteSecutor LRF 75-640

And digital zoom can’t add to that count — it crops the center of the FOV and paints the same pixels bigger:

Simulated thermal view at base magnification: 17 pixels across the coyote
17 px, as capturedthe full FOV the sensor sees
The same 17-pixel capture at 2x digital zoom: identical pixels drawn larger
The same 17 px at 2× digital zoomidentical pixels, drawn larger — closer-looking, never sharper

These frames are geometry only — the pixels the sensor samples, simulated from core, pitch, and lens, with rendering quality deliberately left out. Sensitivity, glass, processing, and the display decide how cleanly those pixels reach your eye, and that is exactly what the Image grade below measures.

FOV (field of view) = the width you can see at 100 yards, in feet. Coyote detail = pixels across a 30-inch coyote body at 200 yards. Both are pure geometry from core, pitch, and lens — our math, not the brochure's.

Magnification is framing, not resolving. Base magnification falls out of pitch, lens, and eyepiece — notice it tracks the detail column — and digital zoom only enlarges pixels the scope already captured. Two scopes with the same lens and pitch capture the same picture even when one boots up looking closer: a 384×288 core behind the same glass is the literal center crop of the 640×480, so zooming the 640 to the same view matches it pixel for pixel. The bigger core adds width, never blur.
What's inside each grade

What each grade is made of

Each grade is built from a few specs, and the weights below are published and fixed. Every spec converts to points on a fixed anchor curve — each card publishes the conversion, and the worked example below repeats the arithmetic end to end so you can check it. Price isn't part of any of them — an expensive scope earns nothing just for costing more. What you get for your money has its own score, further down the page.

Image

What you actually see through the scope. Three specs set it: the detail the sensor resolves, the faintest heat difference it can show, and the screen that puts it all in front of your eye.

sensor 50%NETD 30%display 20%
50%Sensor resolution

Resolution does more for what you see than any other spec, so it carries half the grade. Every core is scored by its total pixel count on a scale that flattens as it climbs — each doubling buys a little less image than the last. The class is just the bracket; the points come from the exact core inside.

30%Sensitivity (NETD)

Sensitivity earns its keep in humidity, fog and the dawn hour, when everything reads the same temperature. Below about 25 mK you stop noticing the difference on a clear night, so the points flatten as the number drops — a 40 mK core with good processing still reads as a strong field image.

20%Display

Every minute behind the scope is spent looking at this small internal screen, so its quality counts. But a good OLED at 640×480 is already a clear sight picture, so the floor sits high and the extra resolution on premium panels refines the view rather than transforming it.

Handling

Whether you can live with it on a rifle, all night. The grade covers the housing that takes recoil and weather, the ounces it adds to the rifle, the hours a battery lasts, and the eye relief that keeps your brow safe.

build 35%weight 25%runtime 25%eye 15%
35%Build & sealing

The one score that's judged rather than computed, set in three tiers from the maker's own documentation: MIL-STD or machined billet earns 96, IP67 rifle-rated 90, weather-resistant IPX7 71.

25%Weight

Weighed as the scope sits on the rifle, ready to fire — where a maker's published weight leaves out the mount you're forced to buy, we add a flat +5 oz reference mount, the same constant for everyone, and keep the listed figure for audit. Ounces convert on a curve where 16 oz is anchored at 85 and each doubling costs 30 points; lighter is always better.

25%Runtime

Hours on one battery convert on a curve — 4 h is anchored at 29 points, 8 h at 70, 13 h at 99 — because swapping batteries at 2 a.m. is the most-cited frustration in owner reports.

15%Eye relief

Millimeters convert on a straight line — 27 mm is anchored at 20 points, 60 mm at 76 — because eye relief that's too short on a recoiling rifle is a safety problem before it's a comfort problem.

Warranty

What the maker promises in writing for the day the electronics die. The written terms are the only part of the promise you can hold them to, so the grade comes from those terms alone.

30
2 yr
50
3 yr
65
4 yr
80
5 yr
88
6 yr
Years of electronics coverage

The famous "lifetime" warranties usually carve out the thermal electronics, so what counts is the years the electronics are covered: 2 years is anchored at 30 points, 6 at 88, on the stepped ladder above.

Transferable: +8 on top

Coverage that follows the scope to its next owner is worth real money on resale. Warranties you have to register for count at full value — registering takes five minutes, so we assume you will.

From spec to grade

Worked example: how one real scope earns all three grades

Every number above is published, so each grade is just arithmetic you can repeat. Here are all three for the N-Vision Halo XRF, computed start to finish — a proven 640-class core behind a modest display, a short eye box, and a standout warranty. One scope, three separate reads.

Value is a separate question

Price never touches the grade

A $2,000 scope and an $8,000 scope are graded on the same curves. What you get per dollar is its own score — shown as a "value pick" tag and a sort, never mixed into the grades.

Image score÷street price, per $1,000=Value — image points per $1k
Honesty rules

Where the data comes from — and what we won't claim

Every number says where it came from

Each figure is tagged by how we got it. Where a maker won't publish a spec, we work out the right number ourselves and label it — we fill the gap, we don't guess silently or punish the blank.

PUBLISHEDCOMPUTEDESTIMATEDMEASURED

Sensitivity is the most-gamed spec

Brands quote sensitivity (NETD) two different ways, and the after-processing version can't be compared across brands — each brand grades its own homework. We use the comparable figure, tell you which one it was, and flag the uncertainty instead of hiding it.

The brochure range is marketing

The spec-sheet "detection range" is measured against a standing man in perfect conditions. We compute our own ranges against the animals you actually hunt — coyote, hog, whitetail — at a stricter standard, and call them what they are: a best-case ceiling, not a promise.

Independence

No sponsored placements, no manufacturer money, methodology published. Where the site carries retail links they are clearly disclosed and have zero influence on any grade — the grade is computed before the link exists.

Reading the letters

The score is absolute. The grade is relative.

These are two different things, kept apart on purpose. The score is anchored and absolute — a 1024×1024 core is worth 92 today and in ten years, and nothing already scored ever silently re-scores. The letter grade is the opposite: it ranks a scope against the entire current field, so it answers "how does this stack up against everything on the market right now." As the field advances, a scope's score holds still but its letter slips — today's A is tomorrow's C, even though the number never moved. A top-tier scope from 2019 still earns its old score; it just no longer earns the A. The score is the measurement; the grade is the standing.

Versioned, in public

Changelog

Anchors are fixed; the methodology itself still improves. When it changes, the change is logged here with its reasoning — never applied silently. Two kinds of changes earn an entry: changes to the math, which bump the version, and changes to how the published numbers should be read, which don't. Design polish never qualifies.

v1.5 · Jun 2026Pixels-on-target explainer added — core, pixel pitch, and lens size collapse into two honest numbers: the field of view at 100 yards, and how many pixels land on a coyote at 200. A four-step ladder walks the trade with real scopes, simulated frames show what each detail level actually looks like on the animal, and the takeaway is that base magnification is framing, not resolving. Nothing in the engine moved — this changes how a spec sheet should be read, not how anything is scored.
v1.5 · Jun 2026Resolution points now follow total pixels — the five published anchors are unchanged, but a core between anchors is scored by its full pixel count, log-interpolated between them, instead of inheriting its class's headline number. Cropped and in-between cores land where they belong: a 16:9 640×360 earns 73.8 instead of riding the 640-class 80, and a 640×512 collects 80.6 for its extra rows. Recomputed across all 97 scopes; no letter grade moved.
v1.4 · Jun 2026Warranty letters made relative — the points ladder is unchanged, but the A/B/C/D now ranks the written terms against the current field, the same standing model as Image and Handling. Today's market is top-heavy — 59 of 97 scopes tie at the best published terms, so they all hold an A — and the letters spread the day someone ships a longer promise.
v1.4 · Jun 2026Score clamp removed — the engine quietly held sub-scores to a 0–100 window while the page said the scale was open. The page was right, so the clamp is gone; sub-scores now land exactly where the published curves put them.
v1.3 · Jun 2026Reliability verdicts withheld until the data is real — the brand track-record tiers were derived from reviewer and forum sentiment, and sentiment isn't evidence. Reliability comes off the page until the documented-failure database can back a verdict — three honest grades over four with a guess.
v1.2 · Jun 2026Grades made relative to the field — the 0–100 score stays absolute and anchored, but the A/B/C/D letter now reflects where a scope ranks against the current catalog (A = top 20%, D = bottom 15%). Scores never silently change; grades age as the field advances, so a scope can slip a letter without its number ever moving.
v1.2 · Jun 2026Anchor-curve gallery removed — the wall of plotted curves was more clutter than proof; the per-class point tables and the worked example carry the same information where you actually need it.
v1.1 · Jun 2026Image scale recalibrated to field-perceived difference — sensitivity (NETD) and display moved off steep/linear curves onto diminishing-returns curves with higher floors, because in the field a good core sits much closer to an elite one than raw specs imply. Sensor-resolution points are now published per core. Recomputed across all 97 scopes; rankings shift accordingly.
v1.1 · Jun 2026Reliability model reframed — the credibility multipliers became a plain evidence hierarchy topped by documented real-world failure data, not instrumented bench testing; the confidence rating is now explicit and reads "limited data" where evidence is thin.
v1.0 · Jun 2026Weight anchor curve adopted — 16 oz ready-to-fire = 85, each doubling −30, uncapped in both directions. Replaces an internal capped mapping; ounces are felt in proportion, and the open curve matches the rest of the scale.
v1.0 · Jun 2026Weight normalized to ready-to-fire — published weights audited per scope for whether they include a mount; +5 oz flat reference mount added where they don't. Listed figures retained for audit.
v1.0 · Jun 2026Ownership grade retired — the 50/50 warranty-reliability blend hid exactly what buyers need to see. Warranty and Reliability now stand on their own.
v0.2 · Jun 2026Fixed score scale locked — we rejected putting the underlying score on a curve against the current market, because it silently re-scored older scopes every time something better shipped. (The letter grade is deliberately relative — see v1.2.)