#1
|
|||
|
|||
\'The System\' improvements?
I have a head cold and can't focus enough to play, and was bored out of my skull, so I played 20 5+1 SnG's using 'The System' from TPFAP.
I won one, didn't reach the money in the other 19. Seems to me that The System is a bit too loose, did anyone ever get around to doing extensive sims trying to figure out the 'correct' pushes? (If it's folded to you on the button on level 1 it's a push with 54s according to the System.) Yes, it's obvious I do better playing my regular game. But I have a little boy's fascination with breaking things, and figuring out exactly how broken NL tournies are if lots of dead money play like this is fun. Especially when sick. |
#2
|
|||
|
|||
Re: \'The System\' improvements?
1) The system was not designed for SnGs
2) The system was designed for the WSOP, or other high buyin events, not 5+1 Sngs 3) I believe the system reccomends you wait through the early rounds and does not reccomend pushing with 54s on level one. 4) The system was not designed for SnGs I don't play in tournaments but these things seem obvious. |
#3
|
|||
|
|||
Re: \'The System\' improvements?
1) Duh.
2) Well, in WSOP today you'd be called plenty of times by crappy hands. The 2,000 or so dead money players would love to call you... I guess one could argue that it would have been good before the poker boom, but we'll never know, will we? 3) Yes, obviously one can wait through the early rounds. But if you do the math for the pushes 54s on the button when it's folded is correct according to the system. 4) Duh. There still should be a way to figure out the math for an optimal all-in strategy. Yes, it would be worse than optimal real play, but by how much? |
#4
|
|||
|
|||
Re: \'The System\' improvements?
[ QUOTE ]
1) Duh. 2) Well, in WSOP today you'd be called plenty of times by crappy hands. The 2,000 or so dead money players would love to call you... I guess one could argue that it would have been good before the poker boom, but we'll never know, will we? 3) Yes, obviously one can wait through the early rounds. But if you do the math for the pushes 54s on the button when it's folded is correct according to the system. 4) Duh. There still should be a way to figure out the math for an optimal all-in strategy. Yes, it would be worse than optimal real play, but by how much? [/ QUOTE ] I think like the poster above said you applied a system to something it wasn't designed for, thus you can't say it's broken. Further, there is an optimal mathematical strategy for all-in situations, but that doesn't mean it's 100% full-proof. In other words, if you have a mathematically correct situation to push all-in with 45s on the button and get called by aces, that doesn't mean the math wasn't wrong, that means you ran into a better hand and over the course of 20 tournaments (which is a statistically insigifagant sample) that will happen. Oh, and one other thing, the original system was not the same as the one you are using. The one you are using is David's modified system which is mathematically optimal. |
#5
|
|||
|
|||
Re: \'The System\' improvements?
No, it's not optimal unless it was by blind chance. Sklansky says so himself in the text, see page 132 of TPFAP. "Let me reiterate that the above guidelines are very far from perfect. A deep analysis, perhaps with the aid of a computer, would result in more precise and accurate criteria."
It's still a rough draft. So, which changes would need to be made for it to work in a sng? And more interesting perhaps, which changes need to be made now that basically all major tournaments have vastly multiplied the percentage of dead money more than willing to gamble it up? (You're going to get called rather often during the first three or four days of the WSOP main event.) In neither case do I believe that a system like this is better than optimal play, but it's very interesting to see how broken no limit as a betting system is. |
#6
|
|||
|
|||
Re: \'The System\' improvements?
Sklansky published the System which was pretty basic, then came out with a modified System that had you dividing the SB + BB + antes into your stack, (or if you had the biggest stack of who had acted or those who were left to act, the next biggest stack among that group).
Is this the system you are talking about? |
#7
|
|||
|
|||
Re: \'The System\' improvements?
Yes. Was wondering if someone had taken up the gauntlet on doing a more extensive system, and if not what changes would have to be made now that there are 1,500 gamboooolers entering the WSOP main event. When the field is filled with people happy to call an all-in with QTs you can't really rely on all-ins stealing the blinds and antes...
|
#8
|
|||
|
|||
Re: \'The System\' improvements?
One problem is that he advocates raising all in if there is a limper ahead of you, if the key number is right.
The problem is that lots of players like to limp with AA or KK in early position. Another issue is that he does not differentiate between say, KQ and KQ suited. Let's put all this in context. Imagine a PokerStars tournament. 8,000 chips left. Blinds are 400-800 with antes of 50. 9 players at the table. So, there is 1650 in the pot and 8 left to act. Your key number is less than 40. The system has you going all in in this situation with AA-22 AK-A2s AK-A2o KQ-K2s KQ-K2o QJ-32s QT-42s Any 2 suited cards In all, well over half of all possible hands. This is a little extreme, don't you think? Now, imagine the same situation, except you are in the CO+1, with 4 to act. Imagine no limpers. Here, the key number is less than 20, and you are supposed to go all in with any two cards. Again - too extreme. Either the scale needs to be adjusted, or the key number calculation does. |
#9
|
|||
|
|||
PokerStove analysis
I ran selected simulations on PokerStove v. AA-TT, AK-AQ (suited and off), hands that I routinely see call all-ins on Stars.
Note, in the System, Sklansky has you playing KQ suited and offsuit in the 100-150 Key number range, before you would play 99-22, before suited aces, and before suited 0-gap connectors down to 54. As you can see, 99-22 does better than KQs v. AA-TT + AK-AQ, and suited 2-gaps do better than KQ offsuit, much less suited aces, 0-gaps and 1-gaps. So, I think there could be some tweaking done to the System, not only on a key number basis, but on a hand selection basis. Win Rates v. AA-TT & AK-AQ From Poker Stove Pairs AA 84.6% KK 67.6% QQ 55.0% JJ 46.9% TT 40.3% 99 37.3% 88 37.3% 77 37.2% 66 37.3% 55 37.1% 44 36.6% 33 36.1% 22 35.5% Suited Aces AK 51.6% AQ 37.8% AJ 31.7% AT 30.3% A9 29.7% A8 30.0% A7 30.1% A6 29.9% A5 31.2% A4 30.9% A3 30.4% A2 30.1% Unsuited Aces AK 49.2% AQ 34.4% AJ 27.8% AT 26.3% A9 25.7% A8 26.0% A7 26.1% A6 25.9% A5 27.4% A4 27.0% A3 26.5% A2 26.1% Suited Kings KQ 31.5% KJ 33.1% KT 31.8% K9 30.5% K8 30.0% K7 30.2% K6 30.4% K5 30.2% K4 29.8% K3 29.4% K2 29.0% Unsuited Kings KQ 27.6% KJ 29.3% KT 27.9% K9 26.6% K8 26.0% K7 26.3% K6 26.4% K5 26.2% K4 25.9% K3 25.4% K2 25.0% Suited 0-gap QJ 31.6% JT 32.1% T9 30.7% 98 30.4% 87 31.2% 76 31.6% 65 31.9% 54 31.7% 43 29.8% 32 28.0% Suited 1-gap QT 30.3% J9 30.9% T8 30.0% 97 29.9% 86 30.5% 75 30.5% 64 30.4% 53 30.2% 42 28.4% Suited 2-gap Q9 29.1% J8 30.2% T7 29.1% 96 29.1% 85 29.3% 74 29.1% 63 29.0% 52 28.7% For comparison's sake, 72o is the worst hand at 22.1% win rate. |
|
|