> Isn't that basically arguing "Don't talk to me about statistics, I want to talk about this one specific data point"? And why not? In this particular case, th

> Isn't that basically arguing "Don't talk to me about statistics, I want to talk about this one specific data point"? And why not? In this particular case, the statistics did not justify the expectation that this bridge would be at least as robust as the one it replaced. There's a general point to be made about engineering progress, but here we have a case which bucks the trend. Speaking in generalities when the topic is a specific incident is a way of missing the point.