ECE 22.06 vs Snell M2025: what each motorcycle helmet standard actually tests
ECE 22.06 is a government type approval; Snell M2025 is a voluntary lab standard with retail sample testing. Both now include an oblique (rotational) test, DOT does not. Here are the test speeds, who vouches for each label, what independent labs found, and how many helmets in ALLR's catalog carry each (ALLR data, 2026-10-24).
Nobody pays to appear in this guide. Where a link leads to a retailer or a brand store, ALLR earns no commission on anything you buy through it. How ALLR makes money
ECE 22.06 and Snell M2025 are two different kinds of promise. ECE 22.06 is UN Regulation No. 22 (06 series): a government type approval, labelled on the chin strap, that a helmet must hold to be sold in Europe and the UK. Snell M2025 is a voluntary standard written by a private foundation, backed by testing of helmets pulled from retail stock. DOT (FMVSS 218) is the US legal minimum, certified by the manufacturer itself. The big change in this generation is that ECE 22.06 and Snell M2025 both added an oblique (rotational) impact test, and DOT still has none. In ALLR’s catalog, 51.1% of complete helmet pages state ECE 22.06 and 3.5% state Snell (ALLR data as of 2026-10-24, spare parts excluded). This guide goes through what each test actually measures, what independent labs found, and how to choose. For the four labels side by side, start with the certifications overview.
The comparison in one table
| Standard | Who vouches | Main impact speeds | Rotational test |
|---|---|---|---|
| ECE 22.06 | Approval authority | 7.5; 8.2 high, 6.0 low | Yes (8.0 m/s oblique) |
| ECE 22.05 | Approval authority | 7.5 m/s (5.5 chin bar) | No rotational limit |
| Snell M2025R | Snell + retail tests | 8.20 flat, 7.75 edge, 7.70 hemi | Yes (8.0 m/s oblique) |
| Snell M2025D | Snell + retail tests | 7.75 m/s, all anvils | Same oblique test |
| DOT FMVSS 218 | The manufacturer | 5.8–6.2 flat, 5.0–5.4 hemi | None |
Do not rank the standards by speed alone. Each uses its own headforms, anvils and pass lines, so “8.2 beats 7.75” is not a safety ranking. What the table does show is which questions each standard asks: how hard the helmet is hit, and, in this generation, whether it limits the twisting a glancing blow puts through the head.
What ECE 22.06 tests
The 22.06 series was adopted by the UN World Forum for Harmonization of Vehicle Regulations in June 2020 as document WP.29/2020/60. The test set below is as summarised in a peer-reviewed comparison from Strasbourg University, presented at IRCOBI 2024 (Bourdet et al., IRC-24-69).
- 01Linear impacts.At the standard points the helmet is dropped at 7.5 m/s onto both a flat and a kerbstone anvil. 22.06 adds a high-energy drop at 8.2 m/s (peak acceleration ≤275 g, HIC ≤2,880) and a low-energy drop at 6.0 m/s (≤180 g, HIC ≤1,300) on the flat anvil, plus at least three extra test points chosen from a pool of 12.
- 02Oblique impacts.Five glancing drops at 8.0 m/s onto a 45° anvil covered in P80 abrasive paper. Pass lines: peak rotational acceleration ≤10,400 rad/s² and a Brain Injury Criterion (BrIC) ≤0.78. This is the new part. 22.05 looked at tangential force but never at rotational acceleration.
- 03Everything else.Visor and field-of-vision requirements, retention-system tests and labelling carry over, with new high-speed particle and photochromic visor tests added in 22.06.
Timing matters when you shop. Authorities stopped granting new 22.05 approvals from 2022, and countries may bar older-approval helmets from sale; SHARP notes that after 2024 all new helmets entering the UK market had to comply with 22.06. Helmets approved to 22.05 are still on shelves in many countries, which is why the label (the “06” on the approval label) is worth reading, not assuming.
What Snell M2025 tests, and how it changed from M2020
Snell’s explanatory cover letter is unusually direct. It says M2025 is two separate standards because ECE 22.06 “appears to be incompatible” with previous Snell standards: M2025D continues the line for helmets that must also meet DOT and JIS, while M2025R is “expected to enable headgear which will also meet ECE R 22-06 as well as DOT and JIS.” Apart from a new set of oblique tests, both are largely unchanged from M2020.
The oblique test drops the helmet at 8.0 m/s onto a flat surface tilted at 45° and faced with 80-grit paper, on a headform treated with the same silicone coating FIM uses. Two helmets are tested at ambient temperature, up to three impacts each. The rotational acceleration limit is 10,000 rad/s² (raised from 9,000 in earlier drafts) and BrIC must stay under 0.78. The M2025R text also adds a flip-up rule: on the first oblique impact the chin bar is locked closed and must not release. The linear tests are still Snell’s own heavy set: M2025D drops at a nominal 7.75 m/s on every anvil, with a 275 G peak limit on the common headform sizes, and a second impact at each site.
Two non-test differences are worth knowing. Snell certification is voluntary and requires the maker to join a random-sample programme in which the foundation buys helmets from the same supply as retail stores and tests them. And the transition was short: Snell began shipping M2025 labels on 1 April 2024, the standards took effect on 1 August 2024 (the first day helmets bearing M2025 labels could be made available for sale), and production of M2020-labelled helmets was scheduled to stop on 1 April 2025 in the cover letter (a later Snell newsletter says by 1 March 2025).
DOT: certified by the maker, no rotational test
FMVSS 218 is a short, tough-in-its-own-way test of its own: peak acceleration may not exceed 400 g, anything above 200 g may last no more than 2 ms and above 150 g no more than 4 ms; four sites, two impacts each, with flat-anvil speeds of 5.8–6.2 m/s and hemispherical-anvil speeds of 5.0–5.4 m/s; a penetration test and a retention test. There is no oblique or rotational requirement. The “DOT” label is, in the regulation’s own words (S5.6.2), the manufacturer’s certification that the helmet conforms. That is a legal baseline for the US market, not an independent grade.
What independent labs found
Two independent sources look at ECE 22.06 helmets and reach different-sounding conclusions, and both are worth holding in mind.
- 01SHARP (UK Department for Transport).SHARP has now tested over 50 helmets approved to 22.06 and says the vast majority score 4 or 5 stars, a substantial shift from the pre-22.06 years. Its June 2026 write-up credits the higher test speed in the new regulation. SHARP’s own linear tests run at 6, 7.5 and 8.5 m/s.
- 02IRCOBI 2024 (Strasbourg).Using the Certimoov protocol on 63 helmet models (47 approved to 22.05, 16 to 22.06; 1,134 impacts), the authors found only marginal reductions in kinematics and “very little improvement” in modelled brain-injury risk, and concluded that 22.05 helmets “do not look significantly different” from 22.06 helmets. The 22.06 sample was small and early, results varied widely between models, and a different protocol will weigh things differently.
The honest reading: a standard sets a pass floor, and individual models differ far more than the badge does. SHARP’s own FAQ says there is no connection between a helmet’s star rating and its price (we dig into that in do expensive helmets protect better?). A label tells you which tests a helmet passed; independent ratings tell you how much margin it had.
What the catalog shows (ALLR data, 2026-10-24)
ALLR records the certifications each retailer states for a helmet and shows them on the product page (as “confirmed on N of M retailer pages”). Counts below are complete helmet product pages in the Canadian view of the catalog, spare parts such as visors and plates excluded; a page can state more than one certification. The same counts are on the helmet certification availability tool.
| Stated certification | Helmet pages | Share of pages |
|---|---|---|
| ECE 22.06 | 7,112 | 51.1% |
| ECE 22.05 | 1,308 | 9.4% |
| DOT | 4,666 | 33.5% |
| FIM | 322 | 2.3% |
| Snell | 489 | 3.5% |
| Not stated | 3,628 | 26.1% |
Snell is rare and concentrated. Of the 489 complete helmet pages that state Snell, Shoei (148), Arai (130) and HJC (76) account for 72% of them (Bell has 63). 94% (460) of Snell pages also state DOT, but only 34% (168) also state ECE 22.06. The rest are mostly US-market versions: Arai’s Contour-X, Regent-X and Signet-X pages state DOT and Snell but not ECE, which matches Arai’s North American listings. Price tracks rarity: the median lowest landed price among Snell pages is CA$854, against CA$417 for ECE 22.06 pages and CA$405 across all priced complete helmet pages.
Helmets whose retailer pages state all of 22.06, DOT and Snell include the Shoei RF-1400, Shoei X-Fifteen and Bell Race Star Flex DLX. Arai is the cautionary case: US-market versions of the Corsair-X and Quantum-X are sold with DOT and Snell, EU-market versions with ECE 22.06, and some retailer pages state all three, so read the label on the version you receive (see the certifications overview). “Stated” is not “absent”: the catalog sorts helmets whose certification we have not confirmed below those we have, and never hides them. You can filter the helmets catalog by certification.
How to choose
- 01Riding in the US or Canada on the street.Look for DOT first, since it is the legal baseline in the US, and check your province or state for any added rule. Adding ECE 22.06 or Snell M2025R gets you a rotational test that DOT lacks.
- 02Buying in Europe or the UK.Look for the “06” series on the label. New helmets sold there are expected to carry it. A Snell sticker is a bonus, not a substitute.
- 03Track days and racing.Check your organiser’s rulebook: many accept specific Snell or FIM labels. Our track-day prep guide lists what US organisations ask for.
- 04Choosing between two certified helmets.Use the independent star rating (SHARP, where the model is tested) and then fit. A well-fitting helmet protects better than a better-rated one that does not fit; see head shapes and lifespan and care.
Common questions
Is a helmet with both ECE 22.06 and Snell safer than one with only ECE 22.06?
Does a Snell M2025D sticker mean the helmet is also ECE 22.06?
Are ECE 22.05 helmets illegal now?
Do I need to replace an M2020-labelled helmet?
Why does ALLR show “Not stated” on some helmets?
About these figures
Counts in this guide are a dated snapshot (2026-10-24) of the public catalog, spare parts excluded; live numbers are on the helmets catalog and the method is on the methodology page. To see two models side by side, try Shoei RF-1400 vs Arai Regent-X or RF-1400 vs Arai Quantum-X, each priced landed per retailer. For the full buying sequence, start at the buyer's guide hub.
Sources
- Snell Memorial Foundation: M2025D and M2025R explanatory cover (25 Apr 2024)StandardOblique test, 10,000 rad/s2 limit, transition dates
- Snell Memorial Foundation: M2025D final textStandard
- Snell Memorial Foundation: M2025R final textStandard
- Snell Memorial Foundation: HeadsUp no. 79StandardM2025 effective 1 Aug 2024; M2020 production stop by 1 Mar 2025
- 49 CFR 571.218 (FMVSS 218)
- IRCOBI 2024: Bourdet et al., IRC-24-69
- SHARP (UK Department for Transport): choosing a new helmetNo connection between star rating and price
- SHARP: Does SHARP work? (1 June 2026)
Watch the price of what’s in this guide
We’ll email you when the landed price drops (shipping, duties and tax included for your region). Free account, one tap.
