Summary History Window - Simulation Tool

[LottoTrout]

I've been experimenting with the Simulation Tool in the Summary History Window. The Simulation Draws dialog box provides an option to use a filter with the simulation (see image below). I'm wondering why that option is there -- under what circumstances would you want to use a filter with the simulation as opposed to running the simulation without a filter..?

simulation_draws.gif

[stan]

There are two reasons one would use a filter in the simulation tool:
- You can select a full wheel and then choose a filter to see how the filter affects the simulation results. Naturally you can filter tickets in the Package, filter them and then run the simulation. But this way is a bit faster.
- There are too many possible combinations in Keno lotteries so it's necessary to reduce the full wheel a bit before running the simulation.

[LottoTrout]

Thanks for your reply Stan.

In your explanation you stated "You can select a full wheel and then choose a filter to see how the filter affects the simulation results". I would appreciate it if you would elaborate on this for me. I'm not understanding the logic of running the simulation against anything other than the full package (assuming it's not too large). Doesn't it make more sense to run a simulation against the full package..? Thanks.

[stan]

Thanks for your reply Stan.

In your explanation you stated "You can select a full wheel and then choose a filter to see how the filter affects the simulation results". I would appreciate it if you would elaborate on this for me. I'm not understanding the logic of running the simulation against anything other than the full package (assuming it's not too large). Doesn't it make more sense to run a simulation against the full package..? Thanks.

For example when you expect that the coming draw will consist mainly of low numbers then you would run Draw Simulation with Low/High filter to use only tickets with high count of low numbers. The shape of simulation histograms will change in all levels. In one level the peak will shift for example to the left. Then you'd compare this with your History estimates made for Draw Simulation with the full wheel and you may notice that your original estimated range is in the right-hand part of the new simulation histogram. It means that your estimate for that level is probably incorrect (or at least less statistically likely to fulfill) if the next draw really does contain a lot of low numbers.

[LottoTrout]

Thanks for the clear explanation !! More testing to do...