"experiment 5 final report"

[stainless steel rat]

For those that remember, I started an experiment several months back, based on the premise of taking a $1000 bankroll and playing until it was gone, or until it was doubled.

The first four experiments ended like this:

test ending BR approx hands played
1 $2,102 1,000
2 $2,062 1,200
3 $2,008 450
4 $2,078 2,835

experiment 5 was just completed this morning on a stop returning from a business trip in New Orleans...

5 $2,015 5,000

Notice the difference in the five experiments with respect to approximate hands played. From a low of about 500, to a high of 5000, a factor of 10.

The last experiment saw my BR see-saw like this (I will give first min, then next max, then next min, until the end:

$1000, $35, $1145, $900, $1345, $852, etc. It swung up and down and up and down until I finally broke $2000 this morning and closed the book on experiment 5.

This was "close". I got all the way down to $35 (spreading $5 to $40, or sometimes $5 to 2x$25 if there was another player at the table) in almost exclusively DD games (a little SD action is mixed in which somewhat confuses the issue of course). When I hit $35 on a high count, I just happened to win, or the experiment would have ended at a busted bankroll. If you figure about 50 hours of play, at $20 per hour, this is right in line with what CVCX says I should earn at that spread for the game given (DD). I could have stopped broke, or I could have stopped 1/2 way through at +1800, but I made up the $2K rule when I started and stuck with it.

I am now going to pause on this and try something "shorter", using a BR of $500 for a while, to see what happens. Obviously if you keep good records, you could see what a BR of $500 would have done since anytime the previous experiments saw a BR of less than $500, they would have ended in bankruptcy. I will look back over them, but suspect that 3 of 5 would have easily doubled, while one would have been close and the last one would have been a bankrupt failure.

Gives some rough idea about variance. I've been asked before "CVCX says I should win $30 per hour betting 10-80 in this particular DD game, but I went down and played and I made over $300 per hour. How can that be? Answer: Variance. next time you might go down and make nothing or even lose. I spent several sessions with my BR well below the starting $1000 value...

Have fun, count accurately, play and bet correctly, and sidestep variance when you can get away with it. :) It will return, just like the terminator. :)

[stainless steel rat]

One more bit of trivia. In experiment 5, with 5000 hands played, I just noticed that at around hand 3000, my bankroll was back to almost exactly $1000. That is I played for about 3000 hands, and broke even. Then over the last 2000 hands, I doubled my starting BR.

The idea of playing 3000 hands for zero profit is a bit gross, when you think about it, and I'd bet this kind of variance is what drives most aspiring card counters back to flat-betting with basic strategy or playing the pass line on the craps table. :)

[jimpenn]

I have experiment playing perfect BS. No counting. I am finding it difficult and long to double. New change...set cash in at 75% BR...for last 10 days very positive results vs attempting to double. Seems like a do better grinding it out...lol

Stainless Steel.... What was your largest down period (markers.amt.) durig experiment. Ex. Did you ever have in excess of $2,500 in markers?

Thanks, Jim

[stainless steel rat]

Never had any markers at all. Each of the five experiments started with exactly $1000 in my pocket. That money was never touched, except to buy into games. The lowest one of those experiments hit was the +$35 point where I bet it all at a TC of +11, and happened to win. A loss would have ended that trial since I didn't allow markers. Also note that I did not eat or anything else out of that $1000 starting BR, nor out of any of the winnings. So technically there were some costs to earn that $1000 that are not included.

I spent about 10 hours playing Thursday night, friday and friday night, and doubled my starting $500 but it was close. At about one hour in, I was up $300, but things turned ugly. Before I knew it I was back to that magic $35 point but amazingly, I won that last bet and started a slow climb out of the hole that ended after about 10-11 hours of playing time. I suspect that this experiment is going to see me go broke more than once or twice, because variance betting up to $40 or $50 a pop at high counts can drain a bankroll of just $500 pretty rapidly.

I like the idea however, as I know of many people that go to the MS coast or MS indian casinos and take something like $500 for the weekend. I'm going to keep reasonable records to see how many trips, how many times I reach $1000, how many times I reach $0, to get some idea of what one could expect playing with $500.

More as time goes on.

these "swings" are pretty interesting to play through...