This week I wanted to cover the belt ranking system that has been developed by the brilliant people behind the lockpicking subreddit. For those of you who are as of yet still unaware, this belt ranking system is an excellent resource for those in the US and Europe, which ranks a wide selection of locks into very rough categories. If you’re able to provide proof of having picked a lock at a certain ranking, you get the corresponding belt. There are some further complications and hoops as you climb, but that’s the basic principle, and this explanation will suffice. The page itself explains these things more fully.
I’ll dig into some of the misconceptions around the ranking system, and talk about some of it’s limitations.
Why create a ranking system?
This is the first question we should ask. There are a variety of answers, too. Firstly, having a ranking system facilitates comparison. It can facilitate ranking locks, or pickers, or potentially both. Being able to compare locks helps us make choices about which locks we want to use, and being able to compare lockpickers helps build on the “sport” aspect of “locksport”. But it’s important to be very clear about the aims of our particular ranking system, and it’s limitations.
Ranking Locks
When it comes to ranking locks, there are a few issues. First of all, no two locks are the same. This is why some people can pop open a given lock of a certain make and model very quickly, but struggle with others of that same make and model. The binding order is determined by the inaccuracies in manufacture, and this is basically impossible to meaningfully account for. There are also factors like bitting, which can affect which rakes – if any – are effective, or which pick profiles might be best suited to picking the lock. A further complication is that some methods may be effective against a given lock, and others not. This is why sometimes one lockpicker at a table struggles with a lock another picker can trivially open. Furthermore, locks change over time as they wear and as they are picked, so a lock that was previously easy to open, can become a totally different beast later on.
so here is the first problem: We can’t strictly rank locks by make and model. Instead, we can only rank individual, particular locks.
One possible approach would be to have a wide variety of lockpickers attempt to pick a controlled selection of locks over an extended period, and then compare how the locks held up.
The issue with this, however, is that this is basically useless. Unless you then wanted to purchase and install one of those locks (which would have been doubtless subjected to countless hours of abuse), this would offer basically no benefit to anyone.
The alternative is to accept a loss of accuracy in your lock ranking system, in order to give approximate rankings for lock models. This obviously has much more utility, and is also the fundamental idea behind the subreddit’s beltranking system. Locks are given relative rankings, as decided by a volunteer team of experienced lockpickers, with feedback from their community.
Ranking lockpickers
Ranking lockpickers also presents challenges, in large part due to the aforementioned factors that make it hard to rank locks in the first place. Again, the key here is to take cumulative times across a wide selection of locks (instead of a wide selection of pickers, this time), and then rank the lockpickers according to that.
There is also one more factor that I haven’t already mentioned – the locks must be unknown.
That is to say, when we attempt to rank lockpickers, we don’t want to rank them based on the time it takes them to pick a lock they are already familiar with. The best way to explain this that I can think of is that comparing lockpickers on the basis of times that they achieve for picking a known lock (like videos on Youtube, where we might be witnessing take 1000, and not take 1.) would be like watching a contest between two people solving Rubix cubes from known positions. I’m sure there is a lot of beauty in seeing a well-choreographed performance, but I think there is a general consensus in the community that higher skill is shown by opening unknown locks, and that contests should focus on that.
Indeed, this is exactly how various contests between lockpickers work, from the Dutch Open which takes place at LockCon, to the German championships, to basically any other locksport contest. In fact, I can’t think of a single lockpicking ranking system that doesn’t work on that basis.
Only the toool.nl Suitcase Contest – to the best of my knowledge – works differently. Competitors are able to pick the same locks over and over again, getting familiar with the lock, and doing their best to improve their times. Unfortunately, it is only open to members of toool.nl.
The Subreddit Belt Ranking System
This brings me back around to the belt ranking system. Now that we understand some of the issues with ranking locks and lockpickers, hopefully what I say about the subreddit belt ranking system will be clearer.
Of course, it’s important to mention the intention behind the belt ranking system. It’s mainly designed to encourage people to improve their lockpicking skills. Of course, this doesn’t stop people from using it to judge which lock would be best for their front door, nor as a shopping list for the collectors among us. But I’d argue it does its job very well. I’ve benefitted from it myself, and I’ve seen countless others benefit from it too. It’s also indubitably the best resource I’ve seen for helping answer the question “Which lock should I try to pick next?”.
That said, it’s important to understand the limitations of the ranking system:
Firstly, the lockpicking subreddit’s belt ranking system doesn’t rank lockpickers. It fundamentally only ranks locks.
Secondly, It can’t distinguish between those lockpickers who were only able to defeat known locks, and those who are sufficiently skilled to defeat locks of a given difficulty without prior knowledge of the lock.
Thirdly, it puts almost 0 value on being able to open locks quickly – Although, since a volunteer needs to watch the video to verify it, it’s considered polite for the video to not exceed a few hours.
Fourthly, for a lock ranking system, it only generally considers non-destructive methods – specifically picking. The heavy emphasis on single pin picking as the method of opening also puts lower value on raking methods, which was one of the reasons I wrote my article in defence of raking.
And finally, the belt ranking system breaks down at the top end of the spectrum. I’d argue that the least well categorized locks are those at black belt level. There’s a real mix of locks there, from those that are just difficult to single pin pick but require no special tooling, to those that have not yet been publicly picked at all.
There are some other interesting effects, however, which I’ve only been able to notice thanks to conversations I’ve had with Holly Poer;
1. Most obviously, it’s resulted in a massive amount more lockpicking videos. The emphasis on gutting the locks also means that a massive wealth of knowledge on the internals of these locks is now publicly accessible.
2. Interestingly, it’s also given rise to very detailed methods for specific models of lock being openly discussed and actively developed. Medeco is probably the most prominent example that springs to mind for me, given the amount of details on specific pick shapes, direction of tension, and attacking the sidebar and then the pins, or vice versa.
3. The majority of people engaging with the belt ranking system seem to be motivated by social value. By this, I mean recognition from their peers for their accomplishments (instead of by, for example, money. This is certainly the case for some locksmiths who specialise in opening a small handful of particular locks quickly and efficiently, and keeping their methods secret from others is important to their business model)
These are details I hadn’t given any consideration previously, and I’m indebted to Holly for her astute observations (And corrections to the article too!).
Conclusion
Wrapping this up, I’d like to recommend that anyone who hasn’t done so takes a look at the belt ranking list on the subreddit. It really is a fantastic resource for finding challenging new locks, interesting lock mechanisms from around the world, and compiling wish lists of locks.
Just remember that it’s a compromise, and the system has limitations. Especially when it comes to making comparisons between lockpickers!