Originally posted by SecurityDrew
Thanks for the explanation. I better understand the tables now and how to use them in the end game.
One more question. How does SAGE fit in? Does it use ICM as well?
No 
But it is almost the same as HU Nash ♦
I guess that will be confusing well enough without further explanation, so let me go into it.
In HU, there is no "risk-aversion". There is no ICM. Risk-aversion exists because (basically, simply put) double the chips don't worth double the equity. A good example for this is a DoN tourney. If you get ITM with 1 chip, it will be just as good as collecting all the chips. The risk-aversion effect is much smaller in reguler SnGs tho.
So, there is this modeling technique which we call ICM. It is to measure how risk-averse we have to be, and how much we have to deviate from simple pot-odds (also known as "ICM tax").
Now, when two people collide in an SnG, the other people (who already folded) win equity with their stacks remaining unchanged. That's where the ICM tax "goes" - the colliding people pay a "tax" for risking dropping out of a tourney, from which the others (who have already folded) benefit. Just imagine the situation that you are short stack on the bubble, and one of the midstacks collides with the bigstack (or the other midstack) and someone is out of the tourney. You are ITM, you have won a lot! Even tho you didn't do anything in the hand.
Why am I explaining this when you asked about SAGE?
Because all this doesn't apply in HU - there is noone to benefit from you colliding with the opponent. If you double up in chips, you have double the equity. In HU all is just pot-odds. There is no risk-aversion, the "who-covers-who" thing doesn't matter either, not anymore.
So that's why I said No, it does not use ICM. In HU, it is just pot odds.
BUT...
Nash equilibrium (in general) has nothing to do with ICM. It can be used with an ICM calculator, or just with simple pot odds (in HU for example) as well. (edit: and for a million other non-poker related things and games in general)
And SAGE is basically a very very clever formula to memorize the strength of individual hands. If you check a Nash HU push/fold chart tho, you will see that it is very very close to SAGE, if I'm not mistaken. I think that in the article about SAGE they mention that it is a close-to-equilibrium strategy.
Cheers,
maybe I will edit this into my original post if it was useful
@chenny
thanks, now I won't have to write it down ever again