anastasios-tampakis.net

Hi LA

If we look the blue line, if its getting ready to climb then on average we can expect few hits in next draws depending on its GAT ID performance, what if we deliberately scan with minus few Draws to get the numbers for current Draw we are targeting ? The reason I am asking here is I was not able to find it on this forum and I am not sure if you guys have already discussed it earlier or not, what you think of it ?

Peace

Hi Rustamshah, if the blue line is about to start the climb, we expect more hits to come in the next few draws, otherwise it wouldn't go up. That is clear I believe, always based and judged on what the blue line did in the test range in a similar situation.
Now this

what if we deliberately scan with minus few Draws to get the numbers for current Draw we are targeting ?

how are you supposed to do this scan? If I understand this correctly, you just want to e.g. remove a few latest draws and with the remaining history to perform a scan that targets the draw in question? You cannot really do this in GAT. There was an old discussion about temporal effectiveness which stated that a particular prediction may be successful even if not played only for the next draw it was prepared for but it can also hit in a few more later draws. In my tests back then I found out this approach could provide some good hits but overall the performance was worse to using a prediction prepared for each individual draw (thus I don't include graphs for temporal effectiveness neither support this). Apart from this, I don't recall any discussion related to your question, primarily because you can't perform this type of scan natively in GAT. You can emulate this of course, just remove the last few draws and pretend they never existed.

So if I got it right, you are saying it could work but you are not recommending it ... Alright then.. however, I saw this happening quite frequently that numbers in a particular good GAT IDs are repeated in next few draws but the real work for me to correctly filter numbers for current Draw. For example, if I choose a particular GAT for next draw and if I choose 12 numbers, may be one of two may match the next Draw numbers but rest 10 or 9 numbers in my experience contains numbers for next three to four Draws.

Based on this observation I thought if I remove 2 to 3 last Draws and then scan GAT to a particular number and then check the numbers then in theory I will be able to see numbers which are already drawn and I may have less numbers to choose... I have not tried it yet, but I guess trying cost nothing, I will post results here if I find it something interesting.

Regards

So indeed we talk for the same thing. I name this behavior temporal effectiveness (you may search this term here at the forums for more). I did develop a specific GAT version to test temporal effectiveness with skip range 2-10 draws span between the last known draw in history and the draw to be predicted. I don't say you'll not get something out of this approach however in my tests the overall performance was lower hit-wise compared to the normal operation. The best production was in the range 2-4 of future draws span. Given this is complicated mechanism, I didn't include it in the system, at least for now.

Thanks LA, I will definately look for it. Just before I let you go do have one request,please have a look at this picture and suggests what numbers or GAT ID you think look promising ? Just your expert opinion

https://s33.postimg.org/dc4svraxr/GAAT.jpg

You use a small test range so these GATs can't show quite their dynamics. Anyway, to my eyes 583367 looks like having a small edge overall.

While I'm here, I might as well chime in. I did my own back tests, temporal effectiveness occasionally gave a 3-hit, while utilising Run Factor had a 50% success rate of scoring a 4-hit in 10 draws (RF10). Never had success with stat.data below 10.

HI Paperino

Thanks for your input. Indeed I did that too and was getting around the same results. Overall, the results for me are the same, very often even in normal scans I get 2 to 3 hits and couple of one offs. In my other approach where I open multiple GAT Engines and run them "Separately" and "Interdependently" I get usually better results. Just recently week ago, choose 18 numbers pool from these different GAT scans run simultaneously and got 5 correct numbers, and I combined them in 14 lines. Got 3 x 3 and 2 x 2 rights, here in the Lottery I am playing that is equal to around 80-90 Euro win which is not bad.

@ LA

I do have a question regarding Euro Millions Lottery, the Lottery commission use different Lottery Machines for Draws, for example couple of weeks ago they were using Machine number 11 and they used it for at least 10 draws, now from last three draws, they are using Machine number 03. Does using a different Lottery machine impacts the overall patterns or that particular Lottery signature ?

Regards.

Hi Rustamshah, everything that is not controlled somehow will affect the performance. Changing machines is probably one of the major randomization factors involved along with the balls swap too. I don't think you can do really anything about this, GAT's scanning process may be able to even distinguish internally between changing machines so to make the proposal for the next draw, however the more major randomization factors involved, this becomes less and less efficient. If you can remove some major randomization factors, by all means do it i.e. if you want to make a particular history for a specific machine and try to predict draws performed on that machine only, you essentially eliminate one major randomization factor. The engine underneath basically tries to identify the effects of the various randomization factors, (major and minor) without knowing which these factors may be. That's the whole point of the engine itself really. So, don't bother much about it if you can't find a way to address a particular randomization factor.

*5-hit* (not covered due to 4if5, but still)