Week 4 Problem Session: Variance workshop

Recomputing the flagship hand's EV at a bigger pot, then measuring the variance and sample-size effects a bare EV figure never shows

15 March 2027

Before the session#

Read the lecture page. Bring the flagship hand’s 9-out, 19.6% equity figure — the session builds directly on it rather than re-deriving it.

In the session#

The group recomputes EV(call) for the lecture’s bigger-pot version of the flagship hand (pot $300, call $50), checking it against the lecture’s stated +$18.48 by hand. Then everyone computes the variance and standard deviation of that same call from the two-outcome formula, and sits with the fact that an $18.48 edge sits inside a $139 swing — most individual attempts at this exact call still lose the $50.

Second half: a show of hands on “if you ran this exact call 10 times and lost 9, would you believe the call was wrong?” before working out the standard error at N = 10 and seeing why that outcome is unremarkable.

Afterwards#

Nothing is marked this week. Going into week 5, what should be settled: computing EV and variance from a two-outcome payoff directly, not from memory of one worked example.