...however, I think it's important to inform whoever is voting that the probability of finding is the same for 24 hours with 50ph and 12 hours with 100ph (with everything else being equal)...
From a calculation point of view, this may be true, but it is only half the truth. The most important and limiting factor is "
time" and this is often neglected or completely ignored in the overall view. Thus the imagination is impaired and leads to blurred statements. The biggest limitation by far is our life time, which is limited. You probably won't be able to mine for longer than 60 years. Possible previous limitations would be serious changes in the Bitcoin protocol, or economic consequences that change the price/performance ratio, or even natural disasters or wars. There are countless possible limitations, but let's keep the best case scenario of our lifetime, because hopefully many decades will pass before this limitation arrives.
If one were to leave the above quoted sentence as it is, then one could think to mine with a CPU at 1 MKey/s in the long run. The CPU miner could hit a block in 4,822,698,348 years, yeah. But that doesn't matter, because before that all the limits mentioned above would be reached and you wouldn't live anymore. You will never be able to experience this block hit. You could also mine at 1 GH/s hashrate, that's not better either. It doesn't matter with which hashrate you mine, if you completely leave out the word "luck" in the overall consideration.
It is probably clear to everyone that the chances of hitting a block are immensely (!) greater for Antpool than for the home Solominer with 1 TH/s. When we talk about chances, the word luck has no place. Probabilities can be calculated mathematically, this is not rocket science and on each place on whole planet earth the result of such a calculation is the same. If someone wants to solve the puzzle, whether he is direct mining or solving a Bitcoin puzzle, then there are only two factors that matter:
TIME and
POWER (Hashrate in case when mining, or Keys/sec when brute-forcing). In both mining and Bitcoin puzzles, you try to be the fastest (
TIME!). If you can't do that, you can't crack the block or the puzzle. This is the only reason why the chance also increases with increasing performance.
If the time would not matter, then yes, you could try to mine a block at 1 MKey/s on a CPU with a current difficulty of 36.95 T
Stuff like "
luck" or "
investment costs" have no place in such a consideration and have absolutely no bearing on the outcome. Everything is defined by
time and
power, including the probability/chance.