More thoughts on simplicity. lol

[KJ]

Ok, so I made this post on another site (wizard's) in a 'count discussion' and a couple people have told me it was my best post yet on the subject, that I made my points more clearly than usual, which I know is one of my weaknesses, so I have decided to re-post it here.

I know the simplicity opponents will dissect and dispute everything, but what the hell....it's a rainy and cold (55 degrees) day here in Vegas, so it will provide some entertainment value.


Really great posts by RS and Romes.

Here's the thing with card counting. The 'math guys' always want to squeeze every last drop of advantage. I understand that is their nature (being math guys).

But in the real world there is something called "diminishing returns" and it works like this: You capture the vast majority of advantage by doing X amount of work. Now by working much harder, which we will call XX or XXX amount of work, you increase your return by a very disproportionally small amount. Very 'diminishing' return for the extra work involved.

In Card counting this concept applies to two areas. Level 2 and 3 counts (and side counts) as opposed to simpler level 1 counts like Hi-lo and also in the number of strategy change or index plays that one chooses to memorize and employ. In the index play part of the conversation, something like 60% of the available gain is achieved with only the top 3 index plays, Insurance, 16 vs 10 and 15 vs 10. Something like 80-85% of all available gain is achieved by playing the top 20 or so index or strategy change plays (Don Schlesinger's Illustrious 18). Playing more than 20 or so (I actually play about 30) just begins to net you very miniscule gains. I call it chasing pennies. :/

But it is even worse than this because even a slightly higher error rate in either of these areas, will wipe out any small gain achieved on paper or in simulations. Proponents of higher counts and more index plays, usually refuse to acknowledge any such higher error rate, something that is scientifically proven when comparing simple tasks to tasks even slightly more complex. They just are refusing to accept the science, so they have these gains or improvements in results that show up on paper or in simulations that just aren't there in real world play.

Furthermore there are some advanced techniques that you can apply that correlate much better to a simple count like high low than higher level counts. One of the things that I do is tracking multiple tables and jumping immediately from a negative or neutral situation to a much more favorable situation. This allows me to see and play FAR more positive and max bet count situations which translates to significant increase (even allowing for increased error rate). We aren't talking 5-10% (which isn't there in reality), we are talking 50-75%, or more.

And this is just one example of such an advanced technique, there are others that I won't mention and almost all of these things work better with a simple type count. These techniques aren't available all the time, but when they present themselves they are actually worth the extra effort as opposed to the concept of higher level counts, which translates to pennies and only on paper (or simulations).

My final two thoughts are these: If you are looking to improve results by working harder, choose something that is going to make a significant difference......Dollars, not pennies. And two, there is a reason most professional players solo as well as teams, including most of the 'famous' teams everyone knows about employs Hi-lo and not some higher level count that can squeeze out a few more pennies for much more work. I mean surely the MIT team had the ability to have played a level 2 or 3 count with side counts. It's just not worth the effort. Diminishing returns.