Collusive Entanglement

[Sredni Vashtar]

[ QUOTE ]
I remember your Shania post and feel that the concept it discussed was explored more fully in John Vorhaus' book 'Killer Poker'.

[/ QUOTE ]

The main reason for the Shania post was because I thought it was fun. Heard many good things about the Vorhaus book (Tommy Angelo I think recommends it) but haven't read it yet. I've read so much poker theory/twoplustwo, etc that I rarely find anything new to me.

[ QUOTE ]
(Unless real collusion is going on, because the situation you describe with Cousin will repeat itself far too often for you to showdown enough winners.)

[/ QUOTE ]

See my post to Mr Angelo, if you would like to comment on that.


Thanks,

SV.

[Tommy Angelo]

Sredni, I think the concept you've laid out here is fantastic, fun, and useful.

"How about we see if we can agree on something?"

And here I thought we had already agreed to agree on everything.


"In a four player game (you and three opponents), the three somewhat skillful opponents are allowed to collude against you. You agree to this for experimentation purposes.Do you think you can make money in this game?"


It depends on how we define "collude." I believe item 2 is the one we're after.

1) If they are colluding by sharing a bankroll, and I know they are sharing a bankroll, and they know that I know, then I think my edge or disadvantage would be exactly the same as if they were all on their own money, minus whatever stabilizing effect that being diversified might have on any of them individually.

2) If again their only collusion is bankroll sharing, but I do not know they are sharing, then I would expect to lose against "somewhat skillful" opponents, the same opponents who, without the collusion, I would expect to beat.

3) If they share a bankroll, and they all know one another's hole cards, and I know they are signaling, then I'd lose a lot.

4) Same as item 3, except that I do not know they are signaling, then I'd lose a lot, a lot faster.


"And if the answer is "it depends on the skill level of the opponents" Would you agree that there would come a skill level where against three opponents that you could have a positive expectation when they were playing fairly (not colluding), but would not have a positive one when they were colluding?"


Yes! Yes! I agree! I always did! I always will! See item 2!


"And if you agree to that then:

Do you think that there is some probability that opponents strategies could inadvertently mimic those collusion ones above?"


I'd say it's past probable and into inevitable.


"Especially when you consider that opponents who tend to be aggressive are much more likely to produce "collusive entanglement" situations? Or have I missed your point?"


I'm not all too sure I had a point. What got me (and has me) financially interested is what you said about short-handed.

I'm starting to see two separate topics here. One is the definition of, and then evidence of, collusive entanglement. On that stuff we agree for sure. It exists, and it has been defined, and it's a nifty concept.

Where collusive entanglement gets hazier in my understanding-so-far is in the application. Tell me again about the part about shorthanded. And is there any distinction that needs to be made between B&M and online?

Also, in thinking of your questions above, is there a semantic element of confusion here, over the word collusive? In collusive entanglement, the entangled players gain some edge merely by being entangled. But in collusive cheating, their edge still depends on how they utilize their entanglement. Does that distinction help things along, or does it reveal my shallowness of understanding of CE?

Good stuff SV,



Tommy

[Mammux]

My experience is that collusion is most often intentional. Not that players are cheating by sharing information or bankroll; but in the sense of making actions based on your outspoken expectation of other players actions.

For example, is there is a mildly TAG twoplustwo reader playing at a family game with some semi-loose weak amateurs. After watching him take away their money for a while, they decide (above table) to start being looser and more aggressive when he is in the pot. Players will call even though they don't have pot odds because they expect the other amateurs to call as well.

-Magnus

[Dov]

[ QUOTE ]
Do you think that there is some probability that opponents strategies could inadvertantly mimic those collusion ones above? Especially when you consider that opponents who tend to be agressive are much more likely to produce "collusive entanglement" situations?

[/ QUOTE ]

Her's a situation I run into in home games all the time. (10-20 with a half kill and 15-30):

2 people who know each other well get into a raising war on on the flop, and after everyone else drops out, they just check it down the rest of the way. If I am in the hand, they will bet the turn and river.

I no longer play in these games, even though I know they are just being nice to each other and not splitting up the money later. The problem is that it feels so much like collusion, that I get taken off my game. This is what makes the game unbeatable for me. It is otherwise a good game with terrible players.

[ QUOTE ]
PS In the Phat Mack hand, while this is what I term collusive entanglement, even there, its beatable. AA would show a profit.

[/ QUOTE ]

It is interesting, but the more I think about it, the more it seems to me that your concept of 'collusive entanlement' is actually one of the main reasons why good players win. For example:

Let's say that in a given hand if you want to prtoect your hand, you will need to c/r the flop. In order to accomplish this, someone else will have to bet after you check, preferably the player on your immediate right. If this player bets, he has actually helped you to beat the field by cutting everyones odds for their draws when you make it 2 bets back to them. This is something you couldn't do by yourself, and without his assistance you would lose this pot quite often, as well as lose a number of bluffing/semibluffing opportunities.

In conclusion, it seems like 'collusive entanglement' cuts both ways. If you can employ the help of the table, you are essentially getting them entangled with you, but if not, you will be entangled with them. The difference lies in the control of the entanglement. If you can control it, you'll win. If you can't, you need to get lucky to win.

I hope this made sense to someone else.

Dov

[Ikke]

Great post and very important subject.

Collusive entanglement happens IMO a lot in shorthanded play. An example:

Shorthanded and button open-raises. Button is on the weak-tightish side. Now, for the SB a good strategy can be to play on the loose side and let him lay down postflop. So he can 3-bet a wide range. But if BB is a very loose player who will defend his blind a lot a problem arises. BB basically makes a bad move by calling 2 cold with trash hands, but by doing so he gives the weak-tight button an advantage and you, in the SB, a tremendous disadvantage. Now his "stupid" play caused your initially maximizing strategy to be exploited by buttons weak/tight strategy! (you're now in a 3-handed pot, which is kinda protected by loose BB, so a tightish strategy is probably preferred)

IMO is collusive entanglement strongly related to the use a maximizing strategies, which, in itself, are exploitable as well. In a full game, standard play approaches a more optimal startegy and therefore is less vulberable to be exploited by a set of entangled strategies. In shorthanded, a maximizing approach is often far more rewarding, so there we see a vulnerability to collusive entanglement.

The example I gave is hard to overcome IMO. You probably have to play tighter, but by doing so you give up a lot, especially if it's 3 handed and blinds come around fast.

Thanks for the post...I haven't thought enough about this subject and your post is making me do so (even with this terrible hangover I'm having ;-) There's a lot more to it, but I really have to sit braindead on the couch now and watch some Southpark.

Regards

[Phat Mack]

In a full game, standard play approaches a more optimal startegy and therefore is less vulberable to be exploited by a set of entangled strategies.

Why would this be so? It seems that any time we step outside the realm of heads-up poker, we are potentially entangled. Wouldn't a full game present more possible entanglements? Or does a larger table offer more self-policing peer pressure that keeps entanglement at bay?

[Phat Mack]

2) If again their only collusion is bankroll sharing, but I do not know they are sharing, then I would expect to lose against "somewhat skillful" opponents, the same opponents who, without the collusion, I would expect to beat.

Wow. Are you saying that you would go from winner to loser if the same "somewhat skillful" opponents secretly started sharing their bankroll? That and nothing else? Not to kidnap a thread, but why?

Also, in thinking of your questions above, is there a semantic element of confusion here, over the word collusive? In collusive entanglement, the entangled players gain some edge merely by being entangled. But in collusive cheating, their edge still depends on how they utilize their entanglement.

It could be merely a matter of intent. But collusive entaglement, as I understand it, has the added feature of being able to change the composition and nature of the "colluding team" every time the action moves. This protean nature makes it harder to identify and deal with.

[Zeno]

And does a tight full game populated with nits, where the fight is usually in later positions and blind play, become, in essence, a short handed game with all the attendant collusive entanglements?

Or for that matter in any type of game where circumstances and situations always ebb and flow about some mean or, on occasion, suddenly ratchet up or down depending on the illogical whims and psychological predilections and of some players.

If nine-handed and the first four or five players fold then you are in a short-handed game – yes, or no.

If two maniacs populate a full game, with some skilled players trying isolation plays do collusive entanglements change not just outcome but the very nature of strategies employed?

Is it then best not to play poker and instead go fishing?

All rhetorical questions really.

-Zeno

[Ikke]

Yes, I stand corrected. What I said isn't necesarrilly true, but my initial thought was that a maximizing strategy is so sensitive to a certain players strategy that a slight shift in circumstances can easily make the maximizing strategy useless and due to this sensitivity I guessed that the maximizing strategy is more vulnerable to entanglement. I can't back this up though.

Thanks for your post, I was way too quick in drawing conclusions. This subject really is fascinating.

Regards

[andyfox]

"Which is more or less what I do more often than not."

One of my favorite sentences. Reminds me of the beginning of a response I got from one of my employees when I asked why a certain product was not inspected more thoroughly. She began:

Well, . . . . actually, . . . . probably, . . . ., maybe, . . ."