A head lies on a pillow.

Soooooo, to rephrase the original question... Has anyone found bias in the attack odds generator?

The reason for my original post was the observation that it's only natural for people to watch the changing odds and try to press the attack button so as to get the best odds. I think this would be the first thing that everyone would do. It's only after a while that I began to question this method and tried shutting my eyes to make the selection more truly random. Once I started doing this, it seemed that my outcomes on average became better. That prompted me to make the initial post to see if anyone else noticed this effect.

The true way to resolve this is to keep a record of the outcomes using both methods and do a full statistical analysis of the results. I have no real interest (or ability) in doing this.

It's certainly a time honored method of introducing randomness in a computer system by having the user press a key. With computer actions in the millisecond and microsecond time range a humans response time is truly random. However, because of the perceived difference in the outcomes using these methods I began to suspect that there was a bias and that it was set up to "punish" those that tried to time hitting the key to get the best odds.

Objectively thinking about it makes this seem silly and paranoid. Who would program this in such a way and how did they correlate it to ensure the bias was in the "right" direction? In reality the true odds of either method are probably the same. However, by closing my eyes and ignoring the display I'm truly certain that I'm getting random results. If I watch the display and correlate my input to something visually on the screen I can't be positive my results are random. I think the best results one can expect long term are the truly random results. Therefore, the only thing I can do that guarantees the best long term outcomes is to close my eyes. Anyway, that's my story and I'm sticking to it.