Here is a summary of the most surprising theoretical poker result I’ve seen in a long time. (I haven’t played poker for money in almost four years, and I rarely think about it seriously, so take that only for what it’s worth.)

In brief: if you are playing heads-up NLHE with 300 BB stacks, and if the SB open-raises to 4 BB with his whole range, you are better off defending your whole range than folding your whole range. Andrew says that he constructed a sim in GTOWizard and that every flop he examined gave the out-of-position defender at least 3.5 BB of equity (out of the 8 BB pot).

So, it’s not even close.

I have very little to say here except “wow,” because I (like Andrew) would have guessed that folding is significantly better. A few misc. thoughts:

1. I wonder how much that changes if you play merely well after the flop (instead of perfectly).
2. I hope it’s obvious that this does not entail that you should actually defend every hand in this situation! Calling everything is only better than calling nothing.
3. This is a good example of why Andrew is still succeeding at poker. A lot of people who have sustained success at poker have both a diverse set of intellectual skills and the metarational ability to know what sort of investigation is appropriate at a given time. Poker has changed a lot in the time I’ve known Andrew, and his approach to poker training has changed more than anyone else’s I can think of.