Payout Percentage Results

[wutsit4]

This is in reference to my IGT Aristocrat Esprit, that with your help we have put back to work.  Thanks again.  My wife is developing a very strong right arm, as she seems to really like our new toy.  This may not work to my advantage.  My saving grace is that she is left handed.

Let me begin with the statement that this machine has no locks, and is only used for entertainment.  That being said, I started to write down the numbers from the counters inside the machine.  It's amazing how quickly the number of plays builds up.  I started thinking about how the machine was actually operating, relative to how it should operate.  My wife finally hit the "jackpot" and the 3 SEVENS all lined up on the pay line.  It only took a bit over 16,000 tries to hit the jackpot.

I have attached a copy of the log sheet that I am using to record and calculate the session totals.  Just wondering if the numbers look close enough to what I should expect them to look like.  The left columns are the date and each meter reading.  The cashbox meter is not counting - have not begun that quest yet.  As I have never looked at anything like this before, I have no reference as to what to expect.  The program chip was labeled as 90.??% if I remember correctly.  Might have been 90.86%. 

Just wondering if someone with a stronger background and understanding could let me know if I am operating within "commercial tolerance".  Thought someone might find this interesting.
Thanks in advance.
   Keith

[stayouttadabunker]

You just gave Statfreak a Christmas gift!
He's gonna love you!!!

[rickhunter]

The sample size is WAAAAAYYYY too small to get anything out of your logs.  You need several hundred days in order to have any kind of statistical validity.  What you are seeing is the volatility of the title you are logging.  It is normal to have wild swings in such a short sample.  The 90% holdback payback [Oops!!! Thanks Stat] is calculated over millions of pulls.   How many virtual stops per reel does your game have?  What is your reel chip?

[StatFreak]

The sample size is WAAAAAYYYY too small to get anything out of your logs.  You need several hundred days in order to have any kind of statistical validity.  What you are seeing is the volatility of the title you are logging.  It is normal to have wild swings in such a short sample.  The 90% holdback is calculated over millions of pulls.   How many virtual stops per reel does your game have?  What is your reel chip?

First, what he said ^^ (except that it's "10% holdback" ). But having said all that (oh wait, I didn't actually say it, Rick did ), the figures look quite normal and average to me. In fact, I don't see any wild swings at all.

Second, nice job setting up your spreadsheet. 

Your average bet is 2.989 coins and you are 176 coins short of full max play, which I would also consider average for a home machine, based on what I have seen on my own machines when others play them. You played 12 out of 20 days, and appear to average roughly two hours a day, which seems a bit light to me for someone with a new (or newly working) machine...
...But then, I'm an addict.

It's interesting that you seem to play a lot on Tuesdays (12/29/09, 1/5/10) although you skipped 12/22, and never play on Sundays. Do you or your wife have Tuesdays off? * 

*If this seems wrong, consider that I have only three weeks of data, which is not enough information to make these inferences. It's basically the same when trying to look at the slot statistics after only 16,000 spins.

[Bobodude]

The sample size is WAAAAAYYYY too small to get anything out of your logs.  You need several hundred days in order to have any kind of statistical validity.  What you are seeing is the volatility of the title you are logging.  It is normal to have wild swings in such a short sample.  The 90% holdback is calculated over millions of pulls.   How many virtual stops per reel does your game have?  What is your reel chip?

Rickhunter is right on. I read some where that the odds are figured after 48 miillion pulls.

[rickhunter]

Actually the odds are mathematically calculated based on probability outcomes (remember from highschool, permutations and combinations) .  In order to have a Statistically significant sample, you need the millions of pulls, the more pulls, the closer you get to this "theoretical" outcome as calculated by the manufacturers.

I believe for the jurisdictional approval, once the programmers finish with the programs and provide these calculated holdbacks, they are then required to run a simulation of millions of pulls, so that they can demonstrate with a statistically significant sample, that their published holdback will hold VERY close to the simulated play.   

Say that it's 16 million spins, you would have to play the machine nonstop for 1,296 days and roughly 8 hours assuming it takes 7 seconds on average  between spins (I added a couple of seconds to factor in winners and the delay it adds).