Why CoinPulse rated Ledger 9.82 in hardware wallets
Hardware wallets are scored on four criteria: Track record (40%), Status transparency (22%), Latency (20%) and Security record (18%). A device in a drawer has no value locked and no uptime of its own, so the profile asks how long the company has run, whether the companion service it depends on can be audited from outside, how that service responds, and whether a hack is on record. Scale of value trusted, Security audits, Multi-chain reach and Peg stability are not scored in this category; their absence is neither a strength nor a weakness. The full rules are on the methodology page.
Ledger's points add up as follows: Track record 10 × 40% = 4.00, Status transparency 10 × 22% = 2.20, Latency 10 × 20% = 2.00 and Security record 9 × 18% = 1.62, for a total of 9.82. Every criterion was measured, so coverage is 100% and no weight was redistributed. The 0.18 points it did not get all come from the one hack entry.
Nothing in this score comes from the devices themselves. The rating does not test a Ledger device, its firmware, its secure element or its recovery options. It measures the company's operational signals from public data.
Where Ledger scored well
Status transparency: the only parseable status feed in the category (10/10)
Ledger publishes its incident history at status.ledger.com on Statuspage, a format the rating reads automatically. That earns a 10 on Status transparency, against a category median of 2: most hardware wallet companies publish no status page at all, and the peers that do (Trezor, Blockstream Jade, OneKey and NGRAVE among them) publish pages without a feed a script can parse, which score 5. At 22% weight this criterion alone is worth 2.2 points, and it is where Ledger separates from every other service in the ranking.
The feed also makes Ledger the only hardware wallet with incident context on its card. Over the 80 days the feed covers (it caps at 50 incidents, short of the 90-day window the collector asked for), Ledger logged 50 incidents of every severity, 18.75 per 30 days, with a median repair time of 26 minutes from report to resolution. 44 of those were labelled major or critical by Ledger itself, which leaves 86.76% of the window free of self-declared major outages. These figures are measured, not scored. They cannot be compared fairly because most services publish no feed, and logging granularity varies: a company that logs every degraded blockchain node will show more incidents than one that logs nothing. The point rewarded here is that the record exists and can be checked by anyone, including against this rating.
Track record: 12.3 years live, full marks
The rating counts 4,494 days of operation since Ledger's 2014 launch, taken from the public record and linked to ledger.com. That earns a 10 on the heaviest criterion in the profile, worth 4.0 points. It is a common result: the category median for Track record is also 10, and the median hardware wallet in this ranking has been live 3,033 days. Longevity shows the company has kept operating through several market cycles; it says nothing about how well any one device protects keys.
Latency: 81 ms median against a 352 ms category median
Our probe timed Ledger's live site at a median of 81 ms to first byte, using two of three samples with the slowest discarded, from a single location. That scored 10, one point above the category median score of 9, and faster than any peer named on this page. The comparison is made only within this category. It measures how quickly a public web page answered one probe location; it does not measure the companion app, firmware updates or transaction signing.
Where Ledger lost points
Security record: one 2020 phishing entry at DefiLlama (9/10)
DefiLlama's hack database, checked on October 19, 2026, holds one entry matched to Ledger by exact name. As DefiLlama lists it: dated November 6, 2020, technique recorded as phishing, a reported amount of $271,000, and no returned amount. No other incident in the 1,307-entry database is matched to Ledger.
The Security record criterion marks each hack on what users lost after any funds were returned. A loss under $1M falls in the band that scores 7. Incidents older than three years count two points less against the service, capped at 9, and this one is almost six years old, so it scores 9. It is the only entry, so it sets the score. That is one point below the category median of 10 (every other hardware wallet named on this page has no hack on record) and costs 0.18 points, the whole gap to a perfect score.
The score reflects DefiLlama's record only: the date, the amount and the technique label. The rule counts nothing outside that database, so incidents that DefiLlama does not record are not scored, and the entry does not grade how the company responded.
Is Ledger safe? What this rating can and cannot tell you
This rating measures operational signals from public data: how long the company has run, how its site responds from one location, whether its status feed can be audited, and what DefiLlama's hack database records against its name. It does not test any Ledger device, its firmware, its secure element, its seed-phrase generation or its recovery features, and it does not assess exposure to phishing or supply-chain tampering. Those are the questions most people mean by "is Ledger safe", and the score cannot answer them. It is not a certification of safety, legality or solvency.
What the measured evidence does show: a company operating continuously since 2014, a status feed that logged and resolved 50 incidents over 80 days with a 26-minute median repair time, a fast public site, and a single $271,000 phishing entry at DefiLlama from 2020. Device price, supported assets, the companion app, customer support and the company's own disclosure history are not scored for any hardware wallet. For company background, see About Ledger; for how the site is paid for, see how CoinPulse is funded.
How Ledger compares in the hardware wallets ranking
Scores in this run range from 6.2 to 9.82 across 15 services, and every one of them was fully measured. Ledger leads by 0.92 points, and the whole lead comes from the status feed.
Trezor and Blockstream Jade, tied at 8.9 in #2 and #3, match Ledger on Track record and Latency (10 and 10) and beat it on Security record with no hack on record (10 against 9, +0.18). But both have a status page without a parseable feed and score 5 on Status transparency, which is −1.1 against Ledger. Net: −0.92, the exact gap. Jade's 77 ms reading was taken at an API endpoint rather than a website, so it is scored on the API bands; it still lands at 10.
OneKey at 8.7 (#4) has the same status gap (−1.1) and the same clean security record (+0.18), plus a slower site: 352 ms scores 9 against Ledger's 10 (−0.2). Net −1.12. NGRAVE at 8.3 (#5) follows the same pattern with a 699 ms site scoring 7 (−0.6), for a net −1.52.
The pattern is the same all the way down: no peer in the category beats Ledger on any criterion except Security record, and the 0.18 points that costs is far less than the 1.1 to 1.76 points every peer loses on Status transparency.
What the Ledger rating means in practice
On the evidence this rating can see, Ledger is the longest-running hardware wallet company in the ranking after Trezor, the only one whose incident history can be checked by a script, and the fastest website timed in the category. The single mark against it is a $271,000 phishing entry from 2020 in DefiLlama's database, old enough to weigh lightly. The score describes the company's public operational footprint. Whether a particular device, firmware version or recovery scheme suits a given holder has to be judged on evidence this rating does not collect.