There have been some good threads about PokerEV, the game analysis graphs and the "all-in" luck graphs lately. I've been looking at Sklansky Bucks, how they're calculated, and what different types of graphs might be telling us.
Sklansky Bucks (or SB's for short) are simply EV calculations. They can only be computed retroactively after knowing opp's hole cards. There may be some way of estimating them for HH's in which we don't have a showdown, but I haven't found the details on that. There is no implied odds - SB's are calculated by multiplying EV by the amount you just bet.
Here's a sample hand to illustrate. I'm not trying to illustrate good poker - just keep the math easy and the hands simple. It's NL100, stacks are $200 each, table folds to SB where Hero picks up KK, raises to $5, gets 3bet by BB villain with AA to $20. Hero calls. Pot is $40. Flop is Q T 3 rainbow. Hero bets $30. Villain calls. Pot is $100. Turn is a K, hero checks, villain raises and Hero rr's all-in, getting a call. Pot is $400.
SB calculation:
Preflop: Hero has 18.75% equity and bets $20.
Flop: Hero has 10% equity and bets $30.
Turn: Hero has 86.4% equity and bets $150.
Preflop: Hero "earns" 18.75% of $40 of betting, or $7.5; but bet $20, so SB = - $12.50.
Flop: Hero "earns" 10% of $60 of betting, or $6.00; but bet $30, so SB = - $24.00.
Turn: Hero "earns" 86.4% of $300 of betting, or $259.20; but bet $150, so SB = + $109.20.
(BTW, I'm not certain this is the correct calculation. Actual EV calculations on each street would include the money already in the pot, but this would double and triple count the preflop $$. So I'm unclear on exactly what the calculation is and haven't found an answer, yet. BTW, my work above skews the results toward preflop action. The other way would make postflop play an even larger portion of the SB's total.)
In SB's, Hero earns $109.20 - $24 - $12.50 = $72.70. In real dollars, Hero takes down a $400 pot except when the Ace hits on the river.
Question: what does PokerEV do with the vast majority of hands where no showdown occurs? There must be some estimating procedure. Does anyone know what it is?
Absent that question, what the red SB line in PokerEV tells us is how often we're reading the table correctly and making good bets into +EV situations. SB's don't go way up when we stack someone with a coin flip hand. SB's actually go up (not down) when we get stacked by some miracle 1-outer on the river. Finally, since larger bets get made later in hands, SB's are weighted toward to postflop play, especially all-ins.
This is why, even though we might be opening only high quality hands, we can have a negative SB line. It means we're playing badly postflop. So looking at PokerEV, the key becomes seeing a steadily upward trend for the red SB line. The variance in poker is the measured by the distance our actual winnings diverge from the red line.
Here's a graph I posted in the thread that got me thinking more seriously about this:
http://xs225.xs.to/xs225/08103/ev20knl10327.jpg.xs.jpg
I know that the downward sloping red SB line correlates with a time I was multitabling too many tables at once - intentionally, as it turns out
:puke:
Here's the thing that interests me: I know I was opening "better" cards because my opening profiled dipped from 18/16 to more like 14/13 during that time. So I had better cards and, within reason, more equity postflop. But my decisions were spewy.
I can think of only two things that could make the SB red line trend downward over time (say 5k hands +): bad post flop play OR several "cooler" situations where the SB's come out negative (as they would with a flopped set/overset scenario). If you can't recall many/any situations where you had to just pay off the great hand if it's already hit, then a downward trending SB line means you suck at postflop poker. Like me.
Any thoughts?
