This was from August last year and that's the only data I found in my old PMs, but this continued for several days the same way.

I know, I know, it's a gaussian formula or w/e, it's not a clean lesser SPAL bounces vs greater SPAL sort of thing. And this is a just a sample (even though, as I said, this continued for days), so it's like the guy who says he didn't make any missiles in 1500 turns - just bad luck with the random factor. But lesser SPAL shouldn't succeed, no matter what the random factor, at 60% on a harmful op like bomb structures against a better SPAL'd country. The formula works for intel ops but for harmful ops it just doesn't.