I see what you're saying, but maybe if you didn't go so fast you could have had properly tested code that didn't break something in the first place. I'd argue that not breaking it in the first place is better than speeding up to find the break.
Yeah I see what you're saying as well and I agree with that. It's hard not to break something when our release cycle depends on a QA person and on customers to break things that we can't always test for (and we're actively discouraged from writing unit tests and other automated tests).
