To be fair, people really aren't too interested in Omega, at least not in conjunction with worst cases. Omega is more suitable for best cases, and that in turn

To be fair, people really aren't too interested in Omega, at least not in conjunction with worst cases. Omega is more suitable for best cases, and that in turn is slightly useless without any knowledge of how common it is. For instance, telling you that bubblesort is Omega(n) in the worst case isn't terribly useful, and telling you it's Omega(n) in the best case is somewhat more enlightening (you now have an absolute asymptotic lower bound), but still not really useful without knowing that the best case is going to be very rare (most of your n! possible input permutations have a lot of inversions). Theta is a bit more interesting, however. I think it speaks to the "tameness" of the algorithm.