Dennis Bassboss
Member
But keep in mind that when the bonus draws are held the announcers saddly lose a lot of their significances...
I've been watching for that 23 also - since we have seen the 17 recently. I like itDennis Bassboss said:Also keep in mind that the 49 is an excellent announcer for 42-23 and 31 all of this is for single announcers....
Dennis Bassboss said:Interesting stats Martek but the most obvious stats of all is the 49 announced 49 a total of 62 times...and that should fill a position...Interesting to notice that the 09 was announced 49 times by 49....
Yes Beaker with bonus included obviously...I think it is better to use them with bonus included and I will tell you why....Because of the pairings that you can build after..martek said:Is that 62 times including the bonus? According to my stats 49 announced itself only 42 times (excluding the bonus).
Dennis Bassboss said:Or it might be even better to use both stats following this reasonning... But nothing beats the double announcers...
Single,pairs,group of numbers you name it...It is awesome very often...And it shows the way...But there's a twist to hit you have to look at history within history in order to find the real gems... In other words you have to look at the actual context and at what it was once...martek said:Double announcers as in a pair announcing another pair or another number?
Nick, can you do 30 cycles and post the results?Nick Koutras said:Some Cyclical comparisons on the Can 6/49 game.
Based on appearances between 50 Cycles (Long) and 15 Cycles (Sort)
I don't know how this got posted earlier - could be headed for a crash hereBeaker said:Nick, can you do 30 cycles and post the results?
I won't do it in 6/49 this time again....I will stick to my long term set...Because of the bonus draws....these numbers previously posted are being study ... I saw a few hits for them using my normal approach...but I'm not ready to say that this is where I will go...I put them here to cross references with the others here in the hope that one of us will spot something for these...often in the past by doing so whether it is Beaker,Peter or someone else...(I like players like Market being involved like he is now...) we were able to catch more numbers by exchanging our thouhts...luckyhorse said:dennis so far i have 49-42-23-31 as your favorite numbers for next draw what about these that you posted 6-12-13-15-16-25-26-46-47 will you be posting a surefire set once again???? i will be looking forward to it
Dennis Bassboss said:Nick this look interesting...Do you have some backtesting on this?
Nick this look interesting...Do you have some backtesting on this?Nick Koutras said:Some Cyclical comparisons on the Can 6/49 game.
Based on appearances between 50 Cycles (Long) and 15 Cycles (Sort)
the following are noticeable based on some power function.
NB Power
37 7.6
06 6.8
38 6.2
15 6.0
21 5.7
26 5.4
49 5.2
16 5.1
32 5.0
07 4.8
08 4.5
47 3.9
45 3.6
05 3.6
48 3.5
33 3.4
30 3.0
29 2.8
13 2.7
46 2.3
11 2.1
34 1.8
04 1.6
12 1.3
14 0.5
35 0.0
01 -0.1
44 -0.8
41 -0.8
22 -0.9
40 -1.0
19 -1.0
02 -1.1
24 -1.2
18 -1.2
10 -1.2
20 -1.4
28 -1.4
39 -1.5
31 -1.7
25 -2.3
23 -2.9
03 -2.9
36 -3.1
27 -3.2
09 -4.7
43 -5.7
17 -6.2
42 -9.5
The above can be used to select
your numbers based on that function.
The greater the Power is the more
probable the numbers shall be, based
on the duration of the analysis.
Dennis Bassboss said:Nick this look interesting...Do you have some backtesting on this?
Beaker said:Nick, can you do 30 cycles and post the results?
martek said:Nick, how is this power function calculated? Or is it top secret?
I'd like to help out with backtesting and crunch some data for us all to see how reliable this is.
Nick Koutras said:There is not any secret here.
Step 1. Calculate the Average & SD
for the Long Cycle
Step 2. Calculate the Average & SD for the Sort Cycle
Step 3. Calculate the Upper Limit of the Graph using 1 Standard Deviation for both cases.
Step 4. Calculate the Difference between Sort & Long Cycles.
Step 4 is the answer of the Power Rating.
We could use different confidence levels but for this kind
of analysis one Standard Deviation is sufficient.