Combined Statistics: Expansion of filters?

[Meisterkeks]

Hello Stan,

i have done very extensive tests, where I created the filters individually with the combined statistics.
For the quality of a forecast, these combined filters really have a very high hit rate. But it is so that you have to create a large number of filters in order to achieve a high reduction of the tickets.

Unfortunately, it is not possible to select the following filter in
Tools -> Options -> Statistics -> Combined Statistics:

Average Distance
Ticket Interval
Number Intervals
History StdDev (diffs)
Standard Deviation
Ticket Index Movement

With these additional filters, the forecast could be significantly improved again in different combinations.

Thanks for checking this out.

[Ralphj]

Hello Stan,

i have done very extensive tests, where I created the filters individually with the combined statistics.
For the quality of a forecast, these combined filters really have a very high hit rate. But it is so that you have to create a large number of filters in order to achieve a high reduction of the tickets.

Unfortunately, it is not possible to select the following filter in
Tools -> Options -> Statistics -> Combined Statistics:

Average Distance
Ticket Interval
Number Intervals
History StdDev (diffs)
Standard Deviation
Ticket Index Movement

With these additional filters, the forecast could be significantly improved again in different combinations.

Thanks for checking this out.

Hello
I think you are using Combined Statistics to do what many of us do with Complex Filters. Complex Filters are much easier and you can drag and drop almost all of the filter list if you need to. It is also easy to adjust the filter values in a Complex Filter, simply select the filter and adjust the filtering values.
Ralphj

[Meisterkeks]

Hello Ralphj,

Thank you, but my goal is that two or more filters are simultaneously considered.

Example, Lotto Germany 6/49, Complex Filter:
Filter 1 = Odd / Even Reject 4/2
Filter 2 = Low / High Reject 2/4
Filter 3 = Sum Root Reject = 9
Package total = 13,893,816 tickets
Package after 3 filter = 7.006.370 Accepted

However, I would like to eliminate only the tickets that meet all three filters (= conditions) at the same time:
Filter Combined Statistic with 3 conditions; Reject because drawn in the last draw
Package total = 13,893,816 tickets
Package after filter Combined Statistic = 13.885.922

The main difference is that filters that eliminate many rows are not good for the prognosis. It is much more efficient to create many filters that eliminate a few rows but have a high probability of the jackpot hit on the remaining tickets.

This is very complex, but only possible with the combined statistics !!

[Falcon]

Meisterkeks - you say...

However, I would like to eliminate only the tickets that meet all three filters (= conditions) at the same time:

I interpret this as a number of filters joined together in an AND type statement.

Combined Statistics can combine up to 5 filter types in and AND combination. Is this what you are after?

cheers

falcon

[Meisterkeks]

Yes:

Hello Stan,

i have done very extensive tests, where I created the filters individually with the combined statistics.

Unfortunately, it is not possible to select the following filter in
Tools -> Options -> Statistics -> Combined Statistics:

Average Distance
Ticket Interval
Number Intervals
History StdDev (diffs)
Standard Deviation
Ticket Index Movement

[stan]

Combined stats can only show those stats that fit into a well known list of discreet values. For example StdDev stats values are floating point numbers so there's almost endless list of possible values therefore it isn't possible to combine StdDev with other statistical properties.

[Meisterkeks]

Ok, thank you, I understand.
However, it would be absolutely sufficient if you restrict or adjust the flowing values so that the combined statistics are always rounded up to 2 or 3 digits after the comma.

However, the following values are not endlessly long, but, like other values, either integers without decimal places or maximla 2 decimal places:

Average Distance
Standard Deviation

Could you at least allow this in Cobimed Statistics?

[stan]

Well, technically it is possible but I'd have to redesign those stats from scratch.
Is that really something that a complex filter and customer analyzer cannot solve?

[Meisterkeks]

Hello Stan,

I'm making an example of how powerful the Combined Statistics can be, Germany Lotto 6/49; Drawing on 23.11.2016.
Filter 1: Arithmetic Complexity 10 = 3.078.578 Tickets
Filter 2: Number Movement Position 1: +3 = 850.668 Tickets

Now I want to exclude for the forecast of the next draw on 26.11.2016 all tickets, which simultaneously the two filter conditions
Arithmetic Complexity = 10
and
Number Movement Position 1: +3
fulfill.

These are 189,312 tickets !! These tickets will not be drawn with a very high probability on 26.11.2016.

And with the filters
Average Distance
Standard Deviation
It is also possible to significantly improve the forecast by various combinations.

So if you could add the two filters still, that would be very good.

[stan]

Ok, but why do you need those stats in combined statistics? Why can't you filter from the charts or use corresponding filters?

[Meisterkeks]

Hi Stan,

I have now created a variety of different combined statistics. So I can already remove between 20 and 40% tickets from the 13.9 million tickets, as these are not likely to be pulled in the next draw. With the filters Average Distance and Standard Deviation I can again create a high quality in the ticket reduction.

Alternatively, I have found no way to achieve this result of the very good ticket reduction on the charts or individual filtering. It is crucial that certain filters always work together, even if fewer rows are removed. But if you create many Combined Statistics, many rows are removed.

A decisive advantage of the combined statistics is that they can be integrated into one or several complex filters and can thus be automated and easily applied for each drawing.

If it is necessary, I can send you by mail some examples (by screenshot) - publicly I would not set this.

Thank you if the extension would be possible.

[stan]

Hi Stan,

I have now created a variety of different combined statistics. So I can already remove between 20 and 40% tickets from the 13.9 million tickets, as these are not likely to be pulled in the next draw. With the filters Average Distance and Standard Deviation I can again create a high quality in the ticket reduction.

Alternatively, I have found no way to achieve this result of the very good ticket reduction on the charts or individual filtering. It is crucial that certain filters always work together, even if fewer rows are removed. But if you create many Combined Statistics, many rows are removed.

A decisive advantage of the combined statistics is that they can be integrated into one or several complex filters and can thus be automated and easily applied for each drawing.

If it is necessary, I can send you by mail some examples (by screenshot) - publicly I would not set this.

Please do email me. I'm sure the current Expert Lotto features should work for you fine.

[Meisterkeks]

Ok i have sent mail