PDA

View Full Version : Eliminating collusion in SNG's


otctrader
03-29-2005, 10:52 AM
Simple idea which may have been suggested before - why don't the major sites just have you enter a queue/pool which is then randomly assigned a table instead of allowing direct table selection?

Since I don't play the high limits I don't know how feasible this would be for harder to fill tables i.e. Step 5's (goes without saying you would need a pool of 2 tables minimum to form to make random assignment useful) - but I think most would gladly sacrafice some convenience in favor of random assignment.

Thoughts?

Rolen
03-29-2005, 11:02 AM
Yeah I had a similar thought, not about collusion but simply because it would make joining tournaments easier - no more QUICK CLICK THE CHECK BUTTON BEFORE SOME MFER BEATS YOU TO THE SEAT, just add your name to a list and when 10 people are ready, start!

jackdaniels
03-29-2005, 11:04 AM
Ask yourself what this does to table selection, one of the most profitable things you can do as a player is try to play where at least some of the players at the table are worse than you. In the lower SNG's most prolly just click on any table that is still registering; at higher limits I'm guessing more table selction is being excersized.

otctrader
03-29-2005, 11:18 AM
Yes, you give up the benefits of table selection, but let's say you grab a seat at a 100 or 200 SNG that has two 60vpip fish present; what's the net value become if a team of 2 or 3 colluders joins the table? Bear in mind any sophisticated colluder likely datamines and will find it +EV to sit at the very same tables you seek out.

jackdaniels
03-29-2005, 12:02 PM
I've read in other posts here that the net effect of colluding is not as +EV as one would think. Essentially, for colluders to do well, they would have to be good players in the first place and as such are prolly better off playing on their own rather than collude. Not sure where the original post for this topic is but if you think of it in terms of preparing for an exam you can get the following analogy:

1. you can spend 4 hours writing out crib notes to use while in the exam (cheat)
2. you can spend the same 4 hours studying for the exam (which you are doing anyway in example one as you have to re-write the material on your cheat sheet).

While you are spending the same amount of time on both excersizes, in the first you need not commit anything to memory while risking the chance of getting busted and in the second you do have to memorize (study) but risk no chance of getting busted. This makes both cases about equal EV - thus you can choose either. Now, add a second person into the mix for the first scenario and you are doubling your chance of getting busted, while only slightly lowring your prep time (if you trust person 2 to do half the work, you may find that they aren't as good "cheaters" as you are - diminishing your success in the exam even if you don't get cought). This makes it more -EV than smply studying in the first place.

How does this relate to poker? If you were to cheat (collude) in poker, you would still need to spend some time "perfecting" your collusion technique (time you could have spent studying the game). When colluding you are doubling your chance of getting cought (risking your whole online bankroll). When colluding you are halving your expected return from the times you win (an even bigger negative compared to the exam situation since on an exam your success doesn't depend on others failing). Thus, you will find that most colluders are NOT sofisticated, don't play very good poker and more often than not, don't even have a good colluding strategy. To take away table selection from the honest players is too high a cost when faced with what is likely a break even EV situation against colluders, and more to the point, it would take it away even in situations where there are no colluders at all (as I doubt collusion is as rampant online as to necessitate taking these kinds of measures).

On another note, the rush to register for an SNG online (before the table closes) - especially at the smaller stakes - could be a proper motive for establishing a random system as described above, but I wouldn't want it to be the norm for ALL sng's.