Table Selection

[TJD]

Cross Posted from HUSH forum.

We make money from the other players' mistakes. The NET result at every table is that the players lose the rake.

This means that whether the table is 6 brilliant players or 6 appalling players the "table" will lose the same. The 6 exellent players will be making virtually NO mistakes. The 6 bad players will be making loads.

I think this must mean that the more bad players there are at a table, the more mistakes they make in total but the actual value of each mistake is less. So we get a formula M x V (number of Mistakes x average Value of mistake). This is logical since on a table of bad players there will be better odds available for their "poor" decisions. In fact some of them might even be right

We then need to subtract from these mistakes R for the rake.

So we get MV-R as the profit available to the table.

In a table of excellent players, MV will be close to zero, so the players lose R approximately and the table becomes unplayable for profit.

However, as Gary Carson says "Just ONE bad player" can make a table playable. The implication here is that the bad player has a high M and that each V is high.

I have always believed that I would prefer a table of 5 bad players but if they make smaller mistakes and sometimes even make a correct decision because of the looseness of the other bad players, how much extra advantage do I actually gain?

What are peoples views on this? Other than ensuring that there ARE bad players to sweeten the pot, how mcuh do we think this changing value of "V" affects our table selection?

T

[poker-penguin]

MV-R is the profit available to the whole table.

However,

Lets say you have one winning player (who makes no mistakes, and hence has an MV of 0) and five fish. The profit available to the winning player is 5MV-R

On a table of five equally winning players and one fish, the profit available to each winning player is MV-R / 5

MV is going to have to go up a LOT to make it a better idea to play with 5 good players. Yes, one bad player with high M (or V) can turn the available table profit into a positive number, but he would have to be spectacularly bad to make more mistakes than 5 fairly bad players (also in Limit holdem at least, the V of each mistake is pretty much capped at a few BBs - even for a spectacularly bad player).

Yes, Morton's Theorem (that's the schooling effect, right?) probably reduces the M (mainly through loose calls that become correct) or the V or both.

But let's be charitable and say that the schooling effect is the same as if one of the five bad opponents is as good as you.

This halfs your share of the profit, but 3MV-R / 2 is still likely to be bigger than MR-V / 5


Sorry if I sound like I think you're trying to argue that it's better to avoid schools of fish (if that's not what you're saying).


I think you have hit on a nice distinction between M and V. It's something I hadn't thought much about, but it's pretty important.

For example, a guy who always calls the river, even when he has no chance, has a high M but they cost 1BB each.

A guy who wrongly folds on the river in a juicy pot once a session has a low M but a high V. But you might not be the one who benefits from it.

High M low V games or players = fairly reliable medium sized profits
High V low M games or players = some big profit ocassions and many small or no profits.


I hope I haven't made a complete monkey of myself here.

[TJD]

I think it must be right that as the players gets looser, the M value must reduce since some calls will become correct. Also, V must reduce as well because the odds will be better for the "chasers".

I genuinely have no fixed view on this. I have always assumed that more bad players must be better. However, if M and V will vary according to the table then how MUCH do they vary?

If we cannot find an ideal table with 5 bad players, how much do we give up if we accept playing on one with only 3 bad players.; 2 etc? At what point does the reduction in our profit become so bad that we are better to watch TV or move down in limits?.......

If I "fiddle" an example to show my point it may stir discussion of "likely" values of M and V.

Let's assume 5 bad players. Two players enter preflop with slghtly -EV hands -0.1 say. I think it is generally accepted that all pf mistakes are small losses.

After the flop, one player who would have been making a mistake in a smaller pot/fewer-tighter opponenets now makes no mistake. The other 4 all make 0.2 errors. This gives a total error of 0.1 + 0.1 + 4 x 0.2, which the 2 non-error players share for 0.5 each

In a game with 2 better players and 3 poor players. The preflop error is not made. However, now the pot is smaller and there are fewer opopnents post flop to make the errors correct/smaller and they remain as bad errors. If the value of these errors were 0.5 each, then the total errors are 1.5 and we divide that amongst the beter players giving the same 0.5 for us.

The crux is just how often the poor players play "correctly" and by how much the size of their mistakes fall.

I really have no idea how to approach this question.

Cheers

T