Answer: B) What actually happens
This is in contrast to theoretical probability where it's what we think would happen, or predict to happen.
For example, the theoretical probability of getting heads on a coin is 1/2 = 0.50 = 50%. If you flip a coin say 1000 times and get 400 heads, then the empirical or experimental probability of getting heads is 400/1000 = 0.40 = 40%. The experimental probability is what actually happens when you do the experiment. The theoretical probability is what you predict what would happen given some list of assumptions and rules about the object.
According to the law of large numbers, the experimental probability should get closer and closer to the theoretical one assuming the theoretical probability has been calculated properly (and all assumptions are correct). If we got say 475 heads instead of 400, then we might have fairly good evidence that we have a fair coin (since 475/1000 = 0.475 is fairly close to 0.500); however, if we got something like 300 heads then we might suspect that the coin is biased toward tails than it is toward heads. The more trials you do, the better a picture you can get of it.
