1. Yes it would be good to stick the exact length of the stargazing cycle into the math - you can't argue sensibly about the exact answer if you are doing the maths with approximate numbers.
2. Incomplete cycles - it's probably correct to include incomplete cycles as the dice is rolled at the start of the cycle. That's not going to make a big difference to the overall answer though.
3. What's still bugging me is the window of opportunity for abduction because it makes a big difference to the answer and because it's a basic fact about abduction. If a sim could do 11 hours of stargazing instead of 9 you'd be closer to 90% than 74% over the long term.
I came across posts about abductions from the electro dance sphere the other day. Is that for real or was there some long fixed bug causing sims to disappear from the dance sphere?
I think exact percentages aren't so important. It's now reasonably established that 5% odds of abduction per night is the correct figure, as advertised. Even if you do precision math on numbers extracted from game code to get a more exact figure, you'll still have operator error and intrinsic randomness. You
can argue sensibly with inexact math because abductions are coded to be random. In fact, it would be very unusual to get precisely 3 abductions for every 4 adult male sims who stargaze their whole lives. You will have players who go 4 generations without an abduction and then get 3 in the space of a sim week, and some will get an abduction in the first sim week. That's not borkage, that's the random number generator being suitably random.
I would argue that in practice a 90% chance of abduction over 29 days is indistinguishable from 75%. Why? Because human experience is simply not that sensitive to probabilistic outcomes. The player who got an abduction in the first week will have the impression that 5% odds per night really isn't so low, the player who waited 4 generations will feel that 5% is intolerably low, and it will not matter whether the exact figure was 4% or 6%. You may want to do the exact math to defend your credibility
but there isn't much more practical utility to be gained by it. In the end, it comes down to play style. What purpose does the player want abductions for, and will it be reasonable to wait possibly a very long time for it?
The electrodance sphere results in abduction and pregnancy if the sim had been abducted by telescope before. I use this often as a way to extend the fertility of stargazing males. Teens can get abducted but not pregnant, however they are eligible for sphere abductions as adults. If you prepare a CAS teen to stargaze through his full teen and adult life stages, the odds for abduction rise to about 90%, based on the same assumptions as above by Novotny. I've not tried a sphere abduction since BV, but it worked as of Seasons.
*Also of note, I suspect that YA sims cannot be abducted.