[sweetjazz]

Here are some actual numbers related to gehrig's calculation. Let's assume that I am playing a player I will refer to Shark. In a HU match Shark has a 1 BB/100 edge over me with an equal number of buttons on each side. Let's say that I win 5 BB/100 winrate while I have the button, while he has a 7 BB/100 winrate when he has the button.

If we just play 1 hand, my button, I will win 0.05 BB on average.

If we play 3 hands, my button twice and his button once, I will win 0.03 BB on average.

If we 71 hands, 35 buttons for him and 36 for me, I will lose 0.65 BB on average.

Okay fair enough, but let's compare those EVs to what would happen if I wait one more hand and play the button each time.

Well after 2 hands, his EV is 0.02 BB.
After 4 hands, it is 0.04 BB.
After 72 hands, it is 0.72 BB.

Notice that in each scenario I have gained an edge of 0.07 BB. Why is it douchey to do so in the first two cases, but not the third case? Is it solely because in the first two cases the Shark can perceive what is happening to him, but in the third case enough activity occurs in the middle for him to fail to realize the EV he just lost?

It seems to me that an argument can be made that a) and b) are douchey if and only if c) is (equally) douchey. What do you all make of this argument?