Notes On Anomaly Detection #1: The Dumbest Crime

“Barry, I have the urge to do all crime, or whatever.”

“Do you indeed, Larry. Me too.”

“Tell you what, Barry. Let’s go around to Harry’s and steal all his stuff.”

“Yeah! I mean, all his stuff has HARRY written on it in foot-high glowing luminescent letters, but OK. Then what?”

“Then we’ll say it’s ours.”

“Yeah! Of course.”

“And we’ll leave it right by our front door in a big pile.”

“Y… wait, no, pump the brakes. What if Harry comes past? He’ll know it’s his. Or his friends will. It’s got his name on it! They’ll tell him!”

“Nah, that’ll never happen.”

“It bloody well might. Can we at least keep it in the house?”

“If we kept it in the house, Barry, then no-one would know it was ours! It has to stay out the front of our house specifically, FOREVER. Do I have to explain everything to you, man?”


And so ends the brief illustrative saga of two criminals who are almost too stupid to breathe.

Their planned crime sucks, in fact, Larry and Barry are so bad at their jobs that I went with Lego figures over pictures of people — in case anyone thinks I’m being racist. “We could live with you calling us dishonest, but calling us that dumb is RACIST!”



This is how dumb plagiarism is.

Think about it.

  1. You must steal something that is already in public view, which means everyone else can see it too, or something that was sent to you privately, which means a record exists proving it was sent exclusively to you.
  2. It is 100% possible to conclusively prove. You did the deed. This isn’t a matter of ‘those results are unlikely’ or ‘this analysis is highly improbable’. It’s absolute. A = B, you = plagiarist.
  3. ‘Sloppiness’ is not really a defense even if it’s a possible explanation, which is rarely the case. “Whoops, I accidentally copied three-quarters of her chapter! How sloppy!” No.
  4. You cannot redact your work, remove it from the public record or alter it to make it acceptable. The genie leaves the box, and does not leave a forwarding address.
  5. For the deed to benefit you, a maximal amount of people must examine the evidence (i.e. must read what you ‘wrote’).
  6. If you get caught once, the rest of your work will be heavily scrutinised — the whole thing will snowball if it ever starts.

This is a really, really bad crime. Risk/reward, people! The inescapable conclusion is that plagiarists are pretty stupid.

Specifically: they do not understand the basic parameters of the transgression they have committed, or they wouldn’t do it once, let alone multiple times. Imagine your research getting famous on the back of you having ripped most of it off. Would you immediately be aware of the precarious position that would place you in? Absolutely! You’d become the narrator figure of the Tell Tale Heart, forever imagining the evidence of your crime beating under the academic floorboards.

No, the only way to live with it is to be wonderfully unaware of how badly you’ve exposed yourself.


While this is all good fun, here’s the dark side: when I discuss fraud detection and meta-science with people, they occasionally pull me to one side (or the internet equivalent) and in an oddly confessional tone, say something like:

“James, I can’t help but wonder… when we do fraud or anomaly detection, are we only detecting the bad frauds? Not morally, I mean, the people who are bad at it? I have this suspicion that we’re only catching the dumb people.”

To which I say, in quite a loud voice: yes, that’s exactly what’s happening.

I have a whole separate piece on the silliness of data fabrication, but rest assured for now that it is as similarly bone-headed as plagiarism if you understand the issues involved.

And therein lies the problem: we aren’t catching anyone else. The moment silliness is less flagrant, it’s much much harder to detect anomalies. Some of them may even be entirely undetectable. And to add to that, very few people are attempting to do detection.

Basically, science may be far more broken than we presently imagine.

Sleep well.