Problem with your book

[damaniac]

No we don't agree. Just because something is likely to occur over the long run AT SOME POINT has no bearing on whether any such event is more likely than average to occur in x period of time. For example, you might expect to see 10H in a row on average every 5000 flips or something (not the real amount). But just because you've gone 5000 flips without seeing it, HAS NO BEARING ON THE LIKELIHOOD OF IT HAPPENING soon thereafter. You have exactly the same P of it occuring as before.