That's not supporting evidence of the actual important part of the claim though. This is like asking "Did the 2008 Ford Ranger have electrical issues due to cost-cutting measures after the housing bubble collapse and subsequent layoffs?" and pointing to some forum posts where individual instances of that vehicle had electrical issues. I could dig up those posts and point to people saying "I've never had those issues", and I guess we can count them pound-for-pound right? But that's not scientific, and neither is this.

Look, I'm not saying that 3rd party plugins never crashed, which would be invalidated by your links. I'm saying that it needs to be proven that they were widespread and/or an actual problem.

AND ALSO we should make sure to prove that the same problem doesn't still exist in the Javascript ecosystem. Which, well, they do. The multiple layers of Javascript confinement we've added to browsers, guess what, those would have worked for Flash too. Saying that it couldn't be secured because it's 3rd party is counter-scientific.