Yes, I agree with this as well... although no matter what polls always do have bias. It's just about recognizing what form it takes and weighing the results accordingly. Even if it was in-game, you'd… (View Post)
The difference in win-rate looks huge, but since it's such a small sample size, is not actually considered statistically relevant, according to Fisher's exact test... ( used http://graphpad.com/quick… (View Post)
This is a great idea / great point. It might not be super easy to do, but it's one of those polish things that would definitely help make it feel better. The health bar segmentation is another thing … (View Post)
