4-of-a-Kind
Join Date: Aug 2007
Posts: 3,068
|
|
Quote:
|
Originally Posted by bjsaust
I gotta admit, I dont know about the maths, but generally speaking you've got the right conclusion, which is that the less skill advantage, the more willing you should be to take small edges, or conversely the more the skill advantage, the less willing you should be to take small edges.
Of course this applies more to deeper stacks. The skill advantage becomes less relevant the shallower the stacks, and the advantage shifts to the player with the better push/fold game, which is to say if your edge is a better knowledge of when to push and when to call pushes, then giving up an edge becomes a bigger mistake because you're negating your own advantage.
|
Well, one place the maths don't work quite right is that any time you have 50% equity in the hand, this method says call, but you probably shouldn't call an all-in on the first hand unless you're somehow certain your equity advantage > skill difference advantage. Same for early game and mid-game before the stacks get shallow - don't call an all-in for a coin flip and the game when you have a 3 to 2 skill advantage. Wait for a hand that's likely to have his range crushed or have his likely hands dominated.
I set up a spreadsheet to get all the formulas working right, and I realized that the "chip switch SD difference" is based effective stacks, not Hero's stack. And I listed every stack size from t100 to t2900. Generally, the maths have us calling with a 40% equity hand when really short, folding when we have a decent sized stack, and calling with good-sized lead (say 2500 to 500). Which makes a good bit of sense.
I've compared the Hero in the BB, SB open-shoves, just to get the setup right. I plan to work on the Hero SB fpr 3xBB, with opp 3b shoveling. With more chips at risk, of course, I'm expecting the "all-in" call recommendations to be more liberal.
When I get some of the maths test-driven for accuracy and run some scenarios, I will post what I find.
Of course, there are some hurdles to overcome:
1. Does Hero really know his skill difference advantage? I would say "yes, to a point." We can always assumed based on history. For example, I've won 57% of my HU $6 matches over 75 or so attempts, so we could safely guess my SD is 55 to 45 against all be the top tier opponents at $6.
2. Can Hero really gauge his equity accurately enough to run these estimates? Yes, since it's the same process as ICM, and yes again, because most of the various all-in configurations like overcard/undercard (A6 vs QT) have well known probabilities. We can generally get a pretty decent idea of opp's shoveling range by seeing a couple of showdowns.
For example, I had been punishing this guy's limps, but I had just flatted a couple as well and saw he liked to limp Axo. About the 4th time I picked up a hand like AT and saw him limp, I raised him 4x. I could tell he was tilting. He thought about it, and I raced him to the middle with my stack when he shoved. Sure enough, he turns up A4o and my AT holds up. Game, set and match.
I'll post something with details in a few days, and see if y'all think it makes any sense.
|