Thanks Chris, did we ever arrive on a 'ballpark' figure for the number of spins required to safely get close to T-RTP on slots?
(Accepting that even that figure will change massively depending on the variance of the slots played.)
The number of spins you need depends on how close to T-RTP you expect to get and, like you said, of the variance parameter of the slot.
You can approximately estimate how close certain number of spins gets you to T-RTP by using the equation below. SQRT() means square root, N denotes the number of spins and SD is the standard deviation parameter of the slot, which tells you how large variance the slot has. In the great work that Chris did Link Removed ( Old/Invalid) , he listed these standard deviation parameters for each of their slots, something no other software operator has done so far.
1 SD Range = SD/SQRT(N)
The above equation gives a 1 standard deviation range, which says that you will fall within to that range 68% of time and within two such ranges 95% of time.
I'll give an example. Let's take some medium variance slot such as Pinnacle's Lucky Lanterns slot which has SD = 6.1 and T-RTP of 97.45%. This SD parameter is also close to large number of Microgaming slots, so this calculation applies to very many typical medium variance slots.
If you play 1000 spins then the 1 SD range from the equation above is:
1 SD Range = 6.1/SQRT(1000) = 0.193 = 19.3%
This means:
68% of time your RTP after 1000 spins is within 19.3% of T-RTP
95% of time your RTP after 1000 spins is within 38.6% of T-RTP
So it doesn't look very promising after 1000 spins. Let's increase spins to 10 000:
68% of time your RTP after 10000 spins is within 6.1% of T-RTP
95% of time your RTP after 10000 spins is within 12.2% of T-RTP
Then for 100 000 spins:
68% of time your RTP after 100000 spins is within 1.9% of T-RTP
95% of time your RTP after 100000 spins is within 3.8% of T-RTP
And finally for 1000 000 spins:
68% of time your RTP after 1000 000 spins is within 0.61% of T-RTP
95% of time your RTP after 1000 000 spins is within 1.22% of T-RTP
So in short, if you want a 95% chance of falling within 1% of T-RTP then on a typical medium variance slot it takes more than 1 000 000 spins to arrive there.
However, what skiny wrote in post #14:
skiny said:Even if you spun a million times and kept track of your RTP the entire time a thousand other people are playing the same game so the game could be paying 97% and unlucky you is only running 82%.
Having a 82% RTP over 1 million spins is practically impossible. The above equation indicates that this would be a (97% - 82%) / 0.61% = 24 Standard deviation result. Even a 5 Standard Deviation result equals odds of 1 in 3.5 million so it easy to see that this sort of result is mathematically impossible - or the game is blatantly rigged.