using rng(nltrn)

[DaycareInferno]

when i play, i use a random number generator to vary my preflop play a little, as do a lot of players. recently i started experimenting with making some random bets post-flop, and i was wondering if anyone had any thoughts on that. it seems to have improved my results so far, but it hasn't been over a long enough stretch to really mean much.

at first, i didn't really like the idea much, because you can obviously pick better than random spots to bluff. the downside to that though, is that when you bluff in prime bluffing situations, you get played back at more, because your opponent can also realise prime bluffing situations.

just to elaborate a little bit. i do not use the rng in raised pots. i only use it for determining when to bluff into an unraised pot that no one has bet into yet. i'll bet hands i would normally bet, +1/3 random bluffs on the flop oop or behind, 1/6 turn behind, and 1/3 river behind. (i won't bluff the river if i have a reasonable chance to show down the winner, of course) so far i have seen a lot higher percentage of hands get folded than when i would bluff roughly the same percentage picking my own spots.

of course, this is just a base. like anything else, you have to adjust to your opponent, but it still seems to have some merit. what do you guys think?

[ChicagoRy]

What level HUSNG do yo uuse this on?

[DaycareInferno]

using it at 10s currently. it might be marginally more useful against better players though, or at least some better players, since bad players take longer to figure out what you're doing anyway.

its kind of interesting though. its possible that it doesn't really help at all, but instead maybe just keeps you from curling up into a ball when you feel like nothing is working. its hard to say at this point. i do know that i've been able to exploit some players that used a very straightforward strategy of trying to check down with good high cards and use only rags for their bluffs in small pots though. using a little randomness would help keep that from happening to a player, not that you can't just change things up on your own.

i'd be interested in anyone's ideas to streamline it a little. obviously i ditch it in a lot of situations anyway, but i'm pretty sure the base usage could be better.

[DaycareInferno]

i should also add that i'm a small bets player, at least at these limits where people don't abuse you as much with position as they do when you move up. i'd imagine you'd have to use different frequencies for a larger betting style.

[soop]

I think a lot of people use cards as their rng. For example, if you wanted to bluff 33% of flops, you could bluff whenever the middle card on the flop was the highest (or something like that)

As you gain experience, you'll probably get better at picking spots to bluff and won't need the rng anymore.

[DaycareInferno]

oh, and if anyone wants it, i think this is the best random number generator for playing poker:

http://www.saccenti.com/randomnumber/randomnumber.htm

its freeware, very small, very simple. all it does is randomizes a number between 0 and x. most other rng i've seen are somewhat complex.

[DaycareInferno]

[ QUOTE ]
I think a lot of people use cards as their rng. For example, if you wanted to bluff 33% of flops, you could bluff whenever the middle card on the flop was the highest (or something like that)

As you gain experience, you'll probably get better at picking spots to bluff and won't need the rng anymore.

[/ QUOTE ]

how good can you realistically get at it when you're dealing with random cards in limped pots, though? the better a flop is for stealing, the less respect a good player will give your bets, because he/she will also be aware that the likelyhood you are bluffing on a K82r board is much higher than it would be on a Q98 2 tone board. i think a human could do marginally better than random, but at the same time, a human can also do a lot worse than random when things are going poorly.

[soop]

You might be giving your opps too much credit. I think you can be way better than marginal if you're good). As a simple example, you should bet more frequently if you're opponent is really tight. Read the Anatomy of a Continuation Bet thread for tons more information.

This doesn't mean you can't use a random number generator, but if you don't vary your bluffing percentages, you are giving up a lot.

[Fliff]

[ QUOTE ]
but instead maybe just keeps you from curling up into a ball when you feel like nothing is working

[/ QUOTE ]
yes

There is no way using rng's or game theory and such is going to be your best option at these stakes.

[DaycareInferno]

oh, sure. i agree with that 100%, but you will still always have a base that you adjust from, an idea of when you're going to bet before you know much about your opponent, and then work from there. i'm just debating whether randomness versus selectivity yields an advantage or disadvantage within that base level or within whatever adjustment that you make.

for instance, after you have so many matches under your belt, you begin to automatically make a note of the texture of the board each time that your opponent bets. if you notice that you're seeing a lot more than normal bets on coordinated boards, then you're usually going to conclude that your opponent likes to play draws agressively. that becomes a big factor later on when you're getting played back at with something like top pair/no kicker.

when it comes to straight steals, if you notice more bets than normal coming from your opponent on dry and paired boards, you will likely attack some of those bets, because not only is the board hard to hit, but but the likelyhood of a bluff combined with the likelyhood of a marginal hand that cannot call a raise is so great that its more than justified.

the advantages of randomness would be that it becomes impossible for an opponent to pick up on a pattern. also, you could construct profiles and frequencies that, while may not be optimal, could be much better than normal play when a person is not playing at the top of their game.

the obvious disadvantage is that overall, your plays will be much less successful in a vacuum. you will be betting in situations that are far from optimal for stealing. one less obvious disadvantage is that when using any sort of random action, you cannot counter-adjust to your opponents adjustment to your actions, because your actions are random. a lot of times matches seem to be won by virtue of exploiting a players adjustments to your actions, even if that adjustment is justified. (i.e. picking up a big hand when you've been playing agressively, etc.)

you are right about me giving my opponents too much credit, though. that's something i've always been bad about. i do think its an interesting topic though.