Originally
posted by
Cable:
Originally
posted by
Xinhuan:
You're calculating something wrongly then.
No I am not calculating anything wrong as I have no other problems calculating every other hit I make. But thanks for the constructive comment, its called bad luck. No fiffrent then running 60 turns and have earthquakes twice in those 60 turns. Or as I stated above many other random fluffty things to go wrong, I have made 0 missiles all set and seen one PCI boom and 2 Work booms that I didnt pay for in 1400ish turns. It all comes down to "luck" and that 1% seems to find me every week.
My 99.25% attack success rate over 3 years disagrees with you. Every time I fail a grabbed, I was able to identify why I failed after the attack. (Typo, or SS instead of PS, or forgetting to multiply in something, or target's NW or ally NW changed, etc).
If a standard gaussian bell curve was used and, IF it was centered at the exact break, you would have a 50% chance of breaking, but you don't, you have less 50% chance of breaking. If you undersend by 1%, you have a ~10% chance of breaking. I've done a number of tests and haven't found any chances of failing if you oversend by over 10%, though it could be so small (1 in 1000) that the tests didn't pick it up.
One of the failure reasons is people not understanding how the def bonus, weap tech, allies and govt multiply/add into each other, and the order of operations. If you just multiplied everything together, and the target is a Republic, you will fail sometimes if he has D allies, because D allies aren't affected by any multiplier.