Two locks, two numbers, and the numbers have nothing to do with each other. The ABUS listing on this page leads with “ABUS Security Level 10”. Kryptonite runs a security scale of its own that also reaches 10. Those two tens are not comparable, are not meant to be comparable, and there is no conversion between them — they are two companies grading their own homework in two different classrooms. If you have ever stood in a shop trying to work out whether a Level 10 beats a rating of 9, the answer is that the question does not have one.
Kryptonite is at least open about what its scale is. Its own support documentation describes “a 10 point system with 10 being ultimate security and 1 being deterrent security”, and it tells you how it intends the number to be used: not as a measurement, but as a recommendation keyed to where you leave the bike. “Kryptonite makes security recommendations based on where you are locking and the length of time that your bike or vehicle will be unattended” — a 1 is for “a low risk area such as out on the trails”, a 10 for “a densely-populated metro area”. Read that way, the rating is a siting guide. It is not a claim about how long the lock survives an angle grinder, and Kryptonite never says it is.
Which raises the obvious question: is anyone grading these things who does not also sell them?
There is an independent test house, and neither listing mentions it
Sold Secure was established in 1992 by Northumbria and Essex Police with Home Office backing, and is now run by the Master Locksmiths Association from a purpose-built test facility in Rugby, Warwickshire. It assesses security products by “manual attack testing”, and it grades them in words rather than numbers:
- Bronze — “Offering theft resistance against a basic tool list (aimed at preventing opportunist crime)”
- Silver — “Offering theft resistance against an enhanced tool list”
- Gold — “Offering theft resistance against a dedicated tool list (aimed at preventing dedicated attacks)”
- Diamond — “The highest level of theft resistance including use of specialist tools (aimed at preventing the most destructive attacks that could include angle grinders)”
Notice what those definitions are made of. Not minutes, not steel grades, not force in newtons — tool lists. Sold Secure says its specifications “vary in the type of attack, tools used, and time spent during the attack”, and the grade tells you which class of tool the product held out against. That is a genuinely different kind of statement from a manufacturer’s index, and it is the only one on this page produced by someone with nothing to sell you.
Two caveats we have to give you honestly. Sold Secure does not publish attack times or numerical thresholds on the page we read, so the grades are ordinal, not quantitative. And it is a British scheme — which matters for a US buyer, because neither Amazon listing on this page mentions a Sold Secure grade at all. We are not going to imply either lock has one. We are pointing out that the most useful rating in the category is the one that survives the trip to a US product page least often.
What the theft research says thieves actually do
The US Department of Justice’s COPS Office publishes a problem-oriented policing guide on bicycle theft (Guide No. 52, 2008, by Johnson, Sidebottom and Thorpe). It is written for police rather than for cyclists, which is precisely why it is worth reading: nobody in it is trying to sell a lock. It sets out six techniques offenders use against locked bikes, and the list reorganizes how you think about the purchase.
- Lifting — “Thieves lift the bike and lock over the top of the post to which the bike is secured.” The lock never fails. The thing you locked to does.
- Levering — thieves use “the gap between the stand and the bike left by a loosely fitted lock to insert tools such as jacks or bars to lever the lock apart”.
- Striking — “If a cyclist locks a bicycle leaving the chain or lock touching the ground, thieves may use a hammer and chisel to split the securing chain or lock.”
- Unbolting — “If a cyclist locks a bike by the wheel alone, then it may be all that is left when the cyclist returns.”
- Cutting — “Thieves are known to use tin snips, bolt cutters, hacksaws, and angle grinders to cut their way through locks and chains.”
- Picking — “For locks requiring keys, thieves can insert tools into the keyhole itself and pick the lock open.”
Read that list again and count how many are defeated by buying a more expensive lock. Two. Cutting and picking are properties of the lock. Lifting, levering, striking and unbolting are properties of how and to what you locked the bike— the post you chose, the slack you left, whether the lock was resting on the pavement, whether you captured the frame or just a wheel. The guide is blunt about where the losses actually come from: “A consistent finding is that most stolen bicycles, regardless of theft location, are either not locked at all or are secured using a lock that requires little force to break or remove.” And it does not let an expensive lock off the hook either: “it has been proposed that all locks can be overcome if the opportunity is present and the offender is suitably equipped”.
One more figure from the guide, because it changes what a lock is for. At the University of Wisconsin, “for 22 percent of cycle thefts, only the lock and secured wheel remained at the crime scene”. And offenders have a financial reason to work that way — the guide records one who reported getting “10 percent of the list price for a whole bike, but 25 percent of the list price for individual parts”. Locking the frame is not a refinement. It is most of the job.
The paperwork is part of the product
Here is the pairing that convinced us this page needed to exist, and we have not seen anyone else put these two documents next to each other.
The DOJ guide, on why recovered bikes do not come home: “Surveys indicate that most cyclists do not know their bicycle serial number, nor can they provide legal evidence of bicycle ownership, such as a purchase receipt.” The consequence it describes is grim — a bike is recovered, the owner cannot prove it is theirs, and the suspect “may be released without charge and may be given the stolen bike on release”.
Now Kryptonite’s Anti-Theft Protection Offer, which is the closest thing in this category to a guarantee. Kryptonite is careful about what it is: “ATPO coverage is NOT bicycle insurance and there is no affiliation with an insurance company. The offer is administered by Kryptonite.” What it does is defined tightly: “If your bicycle or power sport vehicle is stolen because of the opening or breaking of the Kryptonite lock by force, we will pay you the base cost of your bicycle or powersport vehicle (excluding taxes and accessories)” — up to a published maximum that varies by lock. It is optional, it costs a registration fee (Kryptonite lists 34.99 USD for one year and 54.99 USD for two, pricing updated September 2025), and “You must register within 30 days of purchasing your lock.”
And the registration list asks for, among other things: an itemized lock purchase receipt (Kryptonite specifies it must come from a bike or powersport dealer, retail store or authorized online retailer, and says “We do not accept third party receipts”), the key number stamped on the key handle, your bike’s make, model, base cost and serial number, an itemized bill of sale for the bike, and a photograph of the actual bicycle — “not a photo from the internet”.
So: the research says most owners cannot produce a serial number or a receipt, and the guarantee is built entirely out of serial numbers and receipts, collected within thirty days. The lock you buy is one decision. Whether you spend twenty minutes with a phone camera in the week it arrives is the other one, and it is free.
We will not tell you whether an Amazon order satisfies Kryptonite’s “authorized online retailer” wording, because Kryptonite does not answer that on the page and we are not going to answer it for them. Ask them before you count on it. That is the honest version of this advice and the useful one.