QQ from upfront early in Warmup...Is this ok?

[uphigh_downlow]

[ QUOTE ]
Responding to a bunch of different posts:

If you have some other way to adjust for this problem, more power to you. What works for me when deciding when to play across events is to consider total chip outlay in relation to max payoff.

[/ QUOTE ]

First off I'll start by saying that people who say baltostar is making false statements arent completely right.

But I hasten to add that baltostar ( IMO) hasnt made clear his base assumptions and model. So it is in fact a half truth.

The way I see it, what baltostar is proposing is an alternate methodology to compute implied odds, because magically computing the possibility of btn reraise etcis not possible.

While it might be a good strategy to employ for a bot, humans can do better, and thats why we use cost to call. As humans we can adapt better when more information becomes available.

Again your method solves the apparent paradox of using cost to call, when trapped in a min-raising war between two other stacks, but it does not perform well in most situations.

If you design your algorithm to work well for all cases, and make special concession for fringe cases which constitute 5% (shooting from the hip figure, but something ridiculously low) of all cases, then, quite often it will be a poor strategy.

It will be much better to identify the fringe cases, and deal with each separately and accordingly. Treating everything uniformly is a cop-out in my opinion. You seem to have given up and want to do no better.

So you are bent upon creating this uniform decision making model, completely ignoring improvements that can be made on the basis of empirical evidence.

eg: the raising war will happen very infrequently. WHy?? Because empirical evidence tends to indicate so.

Now if you were writing a bot, you could use cost to call in most cases, and total cost in special cases, that are identified using some basic heuristics. I'm pretty sure you would gain equity overall, despite losing some equity while identifying problem cases.

[NYWalker]

[ QUOTE ]
[ QUOTE ]

The "implied odds", which is really the implied expectation, are what you expect to win if you hit, multiplied by the percentage that you hit. It's really that simple. And flopping a set or better is 7.5:1...


[/ QUOTE ]

Does this mean if villain had you covered you would call mostly any amount as long as there were 7.5:1 ratio of what you can win and how much it is to call? This can't be right.

I am trying to develop a shorter hand way of figuring out when to call with small pairs for set odds (I only play live so don't fiddle w/numbers as much as I should). How does your stack and villain's play come into this equations (I realize with smaller stacks and w/a villain that won't stack off w/1 pair calling is less desirable, but how do you account for them mathmatically?).

[/ QUOTE ]

I think 7.5:1 is required implied pot-odd for set-mining.

Stacksize, villain's plays and other factors are definately in consideration.

Edit: As for this hand, this is a call to me.

[Pudge714]

[ QUOTE ]
If a decision criteria for playing across event risk is to be useful, it should not incur radical swings in validity when successively applied to similar scenarios.

[/ QUOTE ]
What does this mean?

[Requin]

[ QUOTE ]
The real problem with only concentrating on cost-to-call is that players tend to get pulled into assuming excessive relative stack risk. Opportunities tend to look better than they really are when rated on a relative basis.

[/ QUOTE ] This is basically saying that players who only consider cost-to-call tend to underestimate how often they will be required to put more money in the pot, right?

You say players 'tend to get pulled in' and situations 'tend to look better' than they are. These are not arguments for your point. A competent player will NOT tend to get pulled in, and will not tend to see better oppotunities than actually exist. That's our whole point: if you grant us reasonably accurate predictive abilities as to our opponents tendencies, then we do not get 'sucked in' to -ev situations. And we don't ask that this assumption be granted for no reason: time and again experience has borne out our assumption.

[dumbndumb]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]

The "implied odds", which is really the implied expectation, are what you expect to win if you hit, multiplied by the percentage that you hit. It's really that simple. And flopping a set or better is 7.5:1...


[/ QUOTE ]

Does this mean if villain had you covered you would call mostly any amount as long as there were 7.5:1 ratio of what you can win and how much it is to call? This can't be right.

I am trying to develop a shorter hand way of figuring out when to call with small pairs for set odds (I only play live so don't fiddle w/numbers as much as I should). How does your stack and villain's play come into this equations (I realize with smaller stacks and w/a villain that won't stack off w/1 pair calling is less desirable, but how do you account for them mathmatically?).

[/ QUOTE ]

I think 7.5:1 is required implied pot-odd for set-mining.

Stacksize, villain's plays and other factors are definately in consideration.

Edit: As for this hand, this is a call to me.

[/ QUOTE ]

I guess I am nittier than this. I just felt a while ago I bled too much set-mining and have tried to find spots where implied odds are worth it.

BTW, I would call this had as well.

[baltostar]

[ QUOTE ]
[ QUOTE ]
The real problem with only concentrating on cost-to-call is that players tend to get pulled into assuming excessive relative stack risk. Opportunities tend to look better than they really are when rated on a relative basis.

[/ QUOTE ] This is basically saying that players who only consider cost-to-call tend to underestimate how often they will be required to put more money in the pot, right?

You say players 'tend to get pulled in' and situations 'tend to look better' than they are. These are not arguments for your point. A competent player will NOT tend to get pulled in, and will not tend to see better oppotunities than actually exist. That's our whole point: if you grant us reasonably accurate predictive abilities as to our opponents tendencies, then we do not get 'sucked in' to -ev situations. And we don't ask that this assumption be granted for no reason: time and again experience has borne out our assumption.

[/ QUOTE ]

Not necessarily EV-. Could be marginal EV+ significantly below the avg opportunity expected during the current M-bracket.

I think this is a bit difficult to see if you concentrate on relatively deep-stacked online tournaments, such as the FTP $1M or $750K with starting stacks 5000,3000 respectively.

If you concentrate on the shallow-stacked tournies, such as the FTP $24+$2 or $69+9, or SnGs, all with starting stacks of 1500, this problem of being pulled into excessive sub-par opportunities becomes very clear.

It doesn't take very many unanticipated min-raises, or even 3x bb raises, following which you again call for set value, or combo-draw value, etc., until you typically end up with significant damage to your stack utility.

The criteria of using cost-to-call to calculate implied odds given to play across the flop event can easily pull you into squandering significant stack utility on multiple significantly sub-par opportunities.


Sidenote: I'm really amazed that there are people on here who think that required implied odds for set-mining are equal to the base odds of hitting a set. This is a ginormous humongtic leak, way way larger than the subtle leak for which I am proposing a solution.

[Exitonly]

[ QUOTE ]
Sidenote: I'm really amazed that there are people on here who think that required implied odds for set-mining are equal to the base odds of hitting a set. This is a ginormous humongtic leak, way way larger than the subtle leak for which I am proposing a solution.

[/ QUOTE ]

Yea, i was going to correct this too, btu saw you posted and was dying to read what new things you've said.

i actually think your latest post is the most coherent you've made, but your whole system seems to be in place to prevent you from [censored] up, and like requin said, better players dont need these guidelines.

[NYWalker]

[ QUOTE ]

Sidenote: I'm really amazed that there are people on here who think that required implied odds for set-mining are equal to the base odds of hitting a set. This is a ginormous humongtic leak, way way larger than the subtle leak for which I am proposing a solution.

[/ QUOTE ]

Can you explain to me the difference between the two?

I'm sorry I haven't read thru the entire thread. Maybe you had said that before, but I'd appreciate if you can explain to me (again, if you had done so).

(BTW, I've been reading about poker computation at 2p2's probability forum and this site, http://www.math.sfu.ca/~alspach/

However,you brought up something different and new.

[Exitonly]

Because everytime yo hit a set you wont get payed off for your full stack, sometimes your set wotn be good, sometimes they'll just c/f and you wont get 1 bet... etc etc.

[djk123]

[ QUOTE ]
[ QUOTE ]

Sidenote: I'm really amazed that there are people on here who think that required implied odds for set-mining are equal to the base odds of hitting a set. This is a ginormous humongtic leak, way way larger than the subtle leak for which I am proposing a solution.

[/ QUOTE ]

Can you explain to me the difference between the two?

I'm sorry I haven't read thru the entire thread. Maybe you had said that before, but I'd appreciate if you can explain to me (again, if you had done so).

(BTW, I've been reading about poker computation at 2p2's probability forum and this site, http://www.math.sfu.ca/~alspach/

However,you brought up something different and new.

[/ QUOTE ]

you aren't going to stack someone every time you hit your set. you have to account for boards that will slow down your opponent and for the rare but very costly set over set situation. also you might lose to a flush or straight.