Accurate SD and EV calculation for record keeping

[Cardguy]

For the non-record keepers out there, and for the brief-post police, I apologize in advance. This is my effort to confirm that I’m keeping records accurately.
Here’s the game: 6D H17 DAS LS, 78% pen, 4 players, 80 rounds/hour, HiLo with 40 indices, Wong in at shuffle, Wong out at -2, Spread $25 to $400 (at TC +5).
When I sim this on CVCX, I get the following:
Backcount 67.1%, Avg bet $69.91, Win/round $0.68, Win/hr $54.76, SD/round $132.03, SD/hr $967.48

Now what if I play for an hour but the TC never gets above +2 so I never bet more than $100 per round. Also, I hit TC-2 earlier than usual and only play 40 rounds in the hour. Is my SD/round still $132.03 or is it much lower since I never got to my big bet? My assumption is that yes, the SD for rounds in that session are much lower, but in other sessions (where the count goes high more than usual) my SD/round will be much higher and in the long run my average SD per round will be $132.03 if I’m consistently playing the same exact game. As a result, I assume it’s OK to just plug in SD $132.03 for my records regardless of how the count actually went. Is this correct?

I'm fairly sure the win/round refers to win per all rounds (including those just observed but not played). Therefore, I assume the win per round actually played is $0.68 /0.671 = $1.01. This makes sense since I would be playing 67% of the 80 rounds per hour which is 54 rounds per hour and averaging about $1 per round which gives me the expected Win/hr of $54.

I also assume the SD/round applies to any round played and doesn't change whether I'm playing all 80 rounds or only 40 (due to Wonging). Correct? Then, to calculate SD/hour, I take SD/round and multiply by the square root of rounds played. So, if I average playing 67.1% of the 80 rounds per hour (53.68 rounds), my SD/hr would be $132.03 * SQRT (53.68) = $967.50 which matches the sim.

So, if I play this game for an hour and estimate that I played 40 rounds, this is what I enter in my records (regardless of how high or low the count went):
Hours:1 Rounds:40 EV/Rd $1.01 SD/round:$132.03 SD/session:$835.03 [$132.03*SQRT(40)] EV:$40.40 Winnings:$xxx
How does this all look?

[Cardguy]

Anyone? Bueller?

[Stealth]

You are comparing actual results to sim results and the fact that you didn't get to your max bet or the wong out was not as simmed is not relevant. The actual results you quoted are in the area of variance which is accounted for in your SD.

Using this session data to develop cumulative stats will smooth the concern about actual playing differences from the sim. Metrics like ZScore can be very enlightening in your analysis both by session and cumulative.

[Cardguy]

You are comparing actual results to sim results and the fact that you didn't get to your max bet or the wong out was not as simmed is not relevant. The actual results you quoted are in the area of variance which is accounted for in your SD.

Using this session data to develop cumulative stats will smooth the concern about actual playing differences from the sim. Metrics like ZScore can be very enlightening in your analysis both by session and cumulative.

Thanks very much Stealth. The z-score is the number of standard deviations any given result is from the mean (x-EV/SD), correct? So if my EV for a session is $100, my SD for that session is $1,000 and my actual result is $600 then my Z-score would be +0.5?

[Stealth]

Correct, ZScore calculation is [(Actual - EV)/Standard deviation]

[LoneWoLF]

Everything gets thrown out the window though when you visit Sands and Harrahs in PA, remember that.

[Stealth]

Everything gets thrown out the window though when you visit Sands and Harrahs in PA, remember that.

No, all your results are with in the variances for the games you are playing ASSUMING YOU ARE PLAYING THEM IN ACCORDANCE TO THE SIMULATIONS!

If they are outside the variances, then you are not playing them according to the simulation protocols.

Or, maybe they are cheating. LOL