Since the formula is a simple probability algorithm, the closer your number attempts gets to infinity the closer the true 1/9 chance will show up.  So keep on plugging?

1.  I know how nukes work
2.  I know how probability works
3.  I had nukes successfully cast on my main buildings just last round

Maybe Bud is magic?

Update:
Dev's haven't been able to check the formula yet, but we were able to set up a scenario on to test multiple nukes.  I'll try to post results, but forums don't like excel copy pastes.

     90 Nukes    45 Nukes

CF         11        6
HF         6             4
LA          8             3
LQ          5     3
MF         11     8
OC         11     7
RS          7     0
RC          18     9
TO          13     5

First column is after 90 nukes, second was after 45.  While in a perfect world column 1 would all =10, and column 2 would all =5... probability isn't perfect.  Everything's being hit, and if I had time to cast 900 nukes those numbers would be closer together.

Step 2 is we're going to set someone up with loads of about 90% cfs and less than 2% of all other buildings, and see if we get any wildly differing results.  This will either require a dev, or a few real time days for us to get the buildings up.

Good to see this is being done if there is anything I can do to help let me know

Update:  No evidence has been discovered to indicate that there's any problem with nukes.

Same thing here (it's a 15% hit, not a 30% hit).  If it only targets buildings that players have, we'd have to significantly increase the difficulty.  The misconception here is that nukes are, ever have been, or were ever meant to be reliable.

Also, in regards to epik, I'm 90% sure I've had the convo with devs before and with the way the code works it would be difficult to have the op ping to check what buildings an individual before calculating what is hit, let alone set it so that it bases it's chances of hitting a building off the % of buildings an individual has.  Option A was doable, but any idiot could build 5 of everything on their hp and negate the change.  Option B would be extremely over-powered, and probably wasn't doable.

There have been preliminary talks about semi-targeted nukes, (3 different nukes each targeting 3 potential building types) but we'd have to figure out a way to balance the increased chance out.  Possibly something similar to destroy cash, and have it do 2-7% damage to whatever you hit instead of a flat 15%.

We're no where near testing on this (we're at the hey...what if we do this stage) but I'll keep this posted.

Nukes are one of the hardest ops to cast, so fails should be expected.

potentially most devastating.....

Point is undeath.. We did so many nukes and not one of them hit his CF's
It's not about the failing.. it's not about the statistics you are showing us.. it's all about doing 50 nukes and not hitting the CF's one freaking time. It looks fishy and you should agree with that instead of saying there is absolutely nothing wrong with it.
Round was won by them anyway.. we just wanted to go out with a blast.. we just wanted to have a final shot at it and proving that being 10% of their NW we could still hurt them really bad.

As Lightguns says, it's exactly about the statistics I'm showing you   I've tried multiple times to duplicate doing 50 nukes and not hitting cfs once, and haven't been able to do it under any circumstances.  Unfortunately you guys just had a run of bad luck, and that's all there is to it

I have some questions/suggestions about how nuking is working right now.

1) Does number of agents affect what buildings get nuked?  I was under the assumption that its the same regardless of agents used as long as the nuke lands.  Therefore I try to use the minimum amount of agents to nuke as long as it succeeds so i don't lose as many agents do to random failures.  I think trying to figure out what is the minimum amount of agents required to nuke adds bit more depth to the game instead of sending max amount every time.

2) if it really takes 1000 nukes or 500 nukes to see an average of each building hit 11% of the time, there is already a problem with the current nuking system.  Unless you are nuking everyday for half the round, its impossible to reach that amount of nukes, not to mention the random failures and better uses of cash and morale.  Realistically you should get an average nuking trend closer to 30-50 nukes or earlier (which is already a couple days of nuking assuming morale cost is 5)

3) I can only speak for myself, but I like to play this game because it is a strategy game, not a game of chance.  Technically speaking, it would be possible for someone to only nuke only the main buildings (which i do think would make nukes a bit too strong).    I would suggest making nukes so that once it hits a specific building, there is less of a chance for that building to be hit again until all other buildings are hit or 24hrs is up.