Originally Posted by YoshiYashiki
May get more responses if you post this in the STT forum.
He would most likely get his thread locked if he posted this in the STT forum (any post that isn't a strategy post seems to get insta-locked nowadays... not sure why).
As for you question, then 0.07% isn't actually that much to pass up on (especially if it's a call which will bust/cripple you if you lose). When SNGWiz first came out it would sometimes try to use a massive "silly" edge like 0.9% (which you should basically never pass up on), but passing up on anything below 0.1% isn't really gonna make a huge difference either way tbo.
One other thing to check is that you aren't just taking the SNGWiz default ranges as gospel and that is why it's telling you to fold AKo here. I know he has recently tried to reinstate his "empirical modeling" ranges which might be too tight for real purposes (hence why it could be telling you to fold AK here). It's essential to alter the ranges to something you yourself think is sensible and then run SNGWiz using them (or else the old computer saying: "garbage in garbage out" stands true...).
When it comes to choosing an edge yourself (and not just using whatever SNGWiz says) then you want to try to consider stuff like:
A) If I bust here, how much future equity gains am I giving up? If it's a table full of loose donks then quite alot probably. If it's a table full of aggressive regs then probably not that much.
B) How likely am I gonna be able to use a big stack to pwn these players? If it's a table full of loose donks then probably not that much. If it's a table full of tight regs then it could make a big stack worth quite alot more than ICM says.
C) How sure of my opponent's ranges can I be? If you are pushing with lots of fold equity then this doesn't matter so much as most times you won't be called. On the other hand if you are calling (especially if you will bust when you lose) then you need to be very sure of your ranges to take a tiny edge. Also, since the ICM function is non-linear you also need to realize that running SNGWiz with a call range of "top 20%" won't give the same results as the average of "top 10%" and "top 30%", and against "random" players you are most often picking a range that is representative of the average you expect to face rather than what the particular player will call with.
D) etc, etc, etc (search the STT forum for the term "edge" and you'll find lots of good posts on it).
After considering each of the important factors then you can then run SNGWiz with an edge of zero and then decide yourself if you think the push is worth taking or not.