Pro opinions needed, low variance and exceeding EV, analye my total play history

[hypercube]

Hi guys, I continue to have a good problem with results far exceeding simulation EV and at a much lower variance than I think should be expected.

I would like to resolve what could be causing this. The easiest explanation is not enough play to be statistically meaningful but the charted results are starting to seem that there actually maybe something to this.

I am also interested in your opinions on when to make a move to green chips. I feel like I am in the zone and comfortable with my play now and the results have been building my confidence that this really is possible. Yet, I am still worried of a drastic change in variance for the worse that many talk about that I have yet to experience. Ideally I would like to go through the experience of a bad negative variance at a lower chip denom before moving to larger denomination. This is one of the mental blocks in my head preventing me from jumping to green.

Tables $5(max spread $40) to $15 (max spread $60)
no wagers larger than $60 at any time.

DD H17, no surr, DAS, noRSA, 3:2, pen 50-65%

session wins:22
session losses: 8
percentage wins: 73.3%
median win rate/hr: approx. $50
no losing periods of longer than 6 hours in 116 hours of total play

The problem:
my sim results say I should be averaging $25-30/hr with an ev of about +1.7% to 1.9%

Please analyze this chart:

hypercube8_28.jpg

[Freightman]

You're overthinking. A 3 ish influence.

What's your bankroll, what's the pen. Cut straight to the chase - briefly.

Oh, and what's your proposed spread at green.

[hypercube]

You're overthinking. A 3 ish influence.

What's your bankroll, what's the pen. Cut straight to the chase - briefly.

the bankroll is on the left axis side of the chart started at about $1500 and has grown to about $5400 shown on the blue plot. The right axis is the dollars/hr for the red and yellow plots.

edit: the pen is either bad 50% in some locations to 70% in better locations. My sim uses 63% which i think is a realistic mean.

I think a big part of this could be that I start off the top with 2x2 units if multiple players at table and drop to one hand if deficit aces or negative count which is not accounted for in my simulation results.

[Freightman]

the bankroll is on the left side of the chart started at about $1500 and has grown to about $5400

What's the pen. What's the spread.
With 5400 roll and 1-8 spread, you're looking for trouble at green.
Caveat - is roll rep,enishable, and if so, at what rate.

[Freightman]

What's the pen. What's the spread.
With 5400 roll and 1-8 spread, you're looking for trouble at green.
Caveat - is roll rep,enishable, and if so, at what rate.

Finally see the pen. Lots if trouble at 50%. 65% with that roll ain't gonna be a walk in the park either.

[hypercube]

I have a professional level salary income, so yes the bankroll is easily replenishable (say about +1.5k/month that i used to blow on slots once upon a time to answer your question) but to keep wife happy, I would like the possibility of that to be remote.

proposed spread at green would be same i use now at $15 min: max spread 2x4units or 1x5units. 2x2units off the top, dropping to 1x1 units if deficit/surplus aces adjusted TC is <-2, playall

[Freightman]

I have a professional level salary income, so yes the bankroll is easily replenishable (say about +1.5k/month that i used to blow on slots once upon a time to answer your question) but to keep wife happy, I would like the possibility of that to be remote.

Part 1
Forget the 50% game for now. Your bankroll can't handle the variance. Add 15 k imaginary funds to your bankroll giving you a 20 bankroll. If you're down $1500 in any month, you're finished for the month - got it?

Part 2
Be damn good to the wife, because the imaginary part of the bankroll is money that she can't save or spend - got it?

Part 3
Don't make any mistakes between the fancy schmancy charts, excess verbiage and your actual ability. Be true to yourself - got it?

[Ryemo]

hypercube,

116 hours is still a very small sample size, so trying to make sense of your results right now is pointless. If you play this game long enough, you'll see a lot of crazy shit. I've be doing this for more than 3.5 years (going on 4 years now) and I've seen some massive swings in both directions just this year alone! I'm currently in the middle of 250 hour losing streak right now. My lowest point of that streak was -45K (not including living/travel expenses). But I've made some recovery and it also helps that I was up +50K in my first 50 hrs of the year. And I can tell you this much; my EV is nowhere near $1,000/hr, so clearly, I was running very well early on. The point is this: big swings will happen. You'll see long positives streaks, as well as losing periods that never seem to end. But if you play long enough, it'll all balance out in the end.

As Freightman said, skip the shitty 50% cut DD games, keep your RoR as low as possible, and just grind out those hours. You can start trying to make more sense of your results when you've got a few hundred hours more under your belt.

[Three]

The bad runs are coming. If you want to track your results to see if they are what you should expect use your EV and SD. Go to this bell curve:

https://static1.squarespace.com/stat...1/?format=750w

Then pick a number of rounds for a typical session to track your play by. Use the EV and SD for that number of rounds. Then plot the various half SD'd plus or minus from EV and note the percentages you would expect to be in that 1/2 SD zone in the bell curve. As time progresses you should find that your results mirror the win/loss distribution. If they don't you are creating session bias which is not good. If you don't leave without regard for win or loss you will create a session bias. You should leave for reasons other than results. Like you have shown your spread too much, conditions are deteriorating, you have won or lost too much (Remember to many casinos your win is determined from you lowest point during the session so a session were you get into the table for $2K and win $3.5K is a $7.45K win in the casinos eyes. Just like once they pay you it is your money so you aren't playing with house money but are playing with your money, to the casino once you lost it to them it is their money). You can end a session whenever you want. By always trying to post a win even when you are down you are showing a big win each time you barely make any money. It doesn't look good to win too high of a percentage of your sessions. A lot of the art to longevity is letting your history look worse than it is. The Floor may glance at your history and see a lot more wins than losses even though you have created a bias that has some monster losses and loads of tiny wins. You may be well behind but at a glance you look like a big winner.

I guess I am getting off track. The point is you should not expect your results to resemble the frequency bell curve until you have played for years. But it can help you spot session bias if you have created any from your exit strategy. It is all one big session so how you are doing should have nothing to do with your exit strategy except to keep the extreme wins and losses from being too extreme.

[Freightman]

3
I look at your answer, which is right.
I look at my answer, which is right.

You have to learn to keep it simple, brief, concise and to the point. A 5 year old can understand my answer.

[jimmybond007]

Far too small of a sample. You are experiencing some good variance but rest assured you have not seen anything yet. There will be losing streaks that will shake your faith in your skill and even make you wonder if counting even really works...
Enjoy the good variance and protect that bankroll. Its easy to feel invincible when you are winning. The losing is what really tests your game....

[SpikeBJ]

A 5 year old can understand my answer.

he even used the words fancy schmancy.

[ZenMaster_Flash]

IF you have no choice but to play H17 games with poor pen'
I suggest that you take up 3-D Chess or tiddly-winks.
More seriously, 65% is just acceptable. 50% is not.