Put yourself in the position of somebody who is trying to commit massive fraud. E.g., by purporting to sell a bunch of stuff on-line. Every bit of information you have about how PayPal combats fraud is helpful. Every time PayPal gives somebody a reason, it helps fraudsters avoid that particular wall or trap, letting them probe the defenses elsewhere.
It's a hard problem. PayPal solved it the wrong way, but they'll never get to full transparency.
Couldn't you say make the same argument of proprietary vs open source software?
Also, they've intentionally chosen a domain where full security isn't possible, mostly due to circumstances outside their control. Remember, they captured a huge new market mainly through the sophistication of their anti-fraud systems. So security through obscurity has to remain a major tool for them.
Basically if you work in fraud detection (or broadly any kind of crime detection) you keep your techniques secret because otherwise you're helping the criminals avoid you.
Look at the source code for Reddit or Hacker News, notice how in the publicly released codebases they've both chosen specifically not to publish the code to do with vote-rigging detection.
There's a strong argument for secrecy.
Then, you need to understand how open PayPal is to fraud. I can easily commit major crimes using PayPal, and security has nothing to do with it.
This is the challenge they face. If you think it's an easy problem to solve, ask yourself: why is there no one else out there doing this same thing? Why are all the other supposed opponents limited in such drastic ways that PayPal is not?
I'm not suggesting they are perfect, but your trite remark is meaningless and ignorant of the realities of the system.
Also don't want to help competing services.
Another data point - spammers used to check their work against spam assassin, fine tuning them.
Another data point - leaked password databases are a source of heuristics for generating word lists.