Small +EV edges: Simulation vs. Reality

[BrassMonkey]

gulon, I had the exact same question. If you have no bloody idea what percentage of the time a person is going to call you, especially since this number changes throughout a game, based on the table's perception of you, how much stock can you really put in the SNGPT numbers? In some cases, it is very clear what must be done, but those aren't the cases in which we're really interested. Perhaps I'm not using the tool right, but I don't think I'm getting as much value out of it as most of the posters on here seem to. What is the right way to approach use of this tool?

Also, do most people use the tool during play? Maybe I'm just a slow mofo, but it doesn't seem like there's enough time for that.

-Brass

[microbet]

Do the math and figure out how big a mistake it is if you are off on your range in either direction. Usually it is not symmetric.

[protoverus]

[ QUOTE ]
gulon, I had the exact same question. If you have no bloody idea what percentage of the time a person is going to call you, especially since this number changes throughout a game, based on the table's perception of you, how much stock can you really put in the SNGPT numbers? In some cases, it is very clear what must be done, but those aren't the cases in which we're really interested. Perhaps I'm not using the tool right, but I don't think I'm getting as much value out of it as most of the posters on here seem to. What is the right way to approach use of this tool?

Also, do most people use the tool during play? Maybe I'm just a slow mofo, but it doesn't seem like there's enough time for that.

-Brass

[/ QUOTE ]

If you have no idea what a person's calling range is during a game then, no, sngpt isn't going to help you.

However, you can (and must) have SOME idea of what a persons calling range is in any situation. Experience and observation will allow you to deduce the general calling ranges of an average person. Then, you can shade the range one way or another based on a specific players specific plays at the table.

Understanding your opponent's calling ranges during any given situation is so almighty important that you must WORK HARD at doing it! This may include taking copious notes, pouring over hand histories, 1-tabling so you can actually observe and think about what people are calling/pushing with...etc.

You'll never be totally accurate, but the point is that the better you are at it then the better your decision will be even if it isn't perfection. Further, playing with sngpt will allow you to see how much changing the ranges affects your decision in a given situation.

Also, no, the tool isn't supposed to be for use at the table. Eastbay deliberately made it so.

Be well.

[modnareno]

What I found is that it is usually its a bigger mistake to make a bad push into a loose caller then it is not make the push. This make me pass up a lot of "close" pushes. I'm not sure if I'm playing correctly by doing it, but I really hate pushing junky hands and getting called by A2 and busting out.

[BrassMonkey]

Thank you Proto for that insight. I haven't played enough of these things, unfortunately, to have encountered too many players more than once (Party), but you confirmed my suspicions that a key facet to success is in knowing your opponents inside and out (I had a hunch, of course, since that's what most of poker is about, but it was good to hear you say it, anyway).

Modnareno, I'm am finding the exact same thing to be true. A lot of people seem not to play "by the book" (i.e., as SNGPT would have you play), even if you're pushing a lot (and they're justified in therefore being SOMEWHAT looser). It's sometimes easy to lose track of stack sizes when you're trying to decide whether you can nuke the remaining players out of the hand with a push. In other words, instead of thinking "do I really NEED these chips right now?", it's easy to sometimes just take the "can I GET these chips right now" mindset when you're not entirely sure if your push is +EV.

[BrassMonkey]

Proto, I have a question for you, as a follow-up to what you said in your post. Would you say that SNGPT's primary purpose, then, is to hone one's instincts? In other words, while you never know exactly what the expectation of your moves are while you're playing (even if you know your opponent is tight, average, or loose), working with SNGPT continuously sharpens your instincts so that your hunches get better and better? These may sound like dumb questions. Maybe they ARE dumb questions. Do you think there are people out there that can make +EV moves almost every time? Ugh, I hope not. Heh.

-Brass

[blackize]

[ QUOTE ]
Would you say that SNGPT's primary purpose, then, is to hone one's instincts? In other words, while you never know exactly what the expectation of your moves are while you're playing (even if you know your opponent is tight, average, or loose), working with SNGPT continuously sharpens your instincts so that your hunches get better and better?

[/ QUOTE ]

Yes

[ QUOTE ]
Do you think there are people out there that can make +EV moves almost every time?

[/ QUOTE ]

Yes

[morgan180]

I think one of the biggest problems is that when you run these simulations through SNGPT or prego you don't give people enough credit for calling with a wide range. you really need to run the same simulation at various calling ranges to dial in where your EV is at for a given push.

then you need to watch your table and look at the general patterns of primarily the people 1 and 2 to your left to get the best possible feel of their calling ranges.

[protoverus]

[ QUOTE ]
[ QUOTE ]
Would you say that SNGPT's primary purpose, then, is to hone one's instincts? In other words, while you never know exactly what the expectation of your moves are while you're playing (even if you know your opponent is tight, average, or loose), working with SNGPT continuously sharpens your instincts so that your hunches get better and better?

[/ QUOTE ]

Yes

[ QUOTE ]
Do you think there are people out there that can make +EV moves almost every time?

[/ QUOTE ]

Yes

[/ QUOTE ]

[HesseJam]

[ QUOTE ]
gulon In some cases, it is very clear what must be done, but those aren't the cases in which we're really interested.
-Brass

[/ QUOTE ]

You'd be surprised to what extend people get these clear cut cases (push, no matter what and call, no matter what) wrong. If you see somebody not pushing his SB where it is an atc push, you have a read right there.

Also, in many cases being slightly off with your pushes is not catastrophic. Being off with your calls is.

Finally, about observing and deducing general calling ranges of your opponents: The problem for a beginner is that he pushes and awaits with tension the reaction of his opponent(s) and then BHAM! His T9o gets called with KKs. So, he got burnt and it is natural to think that he has made a mistake. If he is a fast learner he will not push this hand any longer in this spot and BHAM! he just learned the wrong thing! If he gets called by T8o and wins the hand, he might think "wow" that pushbotting really works" and BHAM! has he "learned" the wrong thing again. This is what SngPT is for. SNGPT will tell you that a) was a good push and b) might have been a bad push, independently from the result.