Strategies Page (lots more to be added)
The
Odds Table
In approximately 5 draws
out of 100, all 5 draws will have been drawn in the last 6 draws. In
approximately 22 draws out of 100, 4 of 5 draws will have been drawn in
the last 6 draws. In approximately 22 draws out of 100, all 5 draws
will have been drawn in the last 10 draws it is on those days where one
can increase the likelihood of winning by 1300% and on 8/13/03,
approximately 19000% by following the three strategies
mentioned above to choosing from a limited subset of numbers. The
first two strategies can be observed by looking at the table
Anywhere in last 6 draws columns with the header '4'
or '5.'
This game has many, many
recurring trends that can be observed. When the short term average,
(defined in this game as being last 100 draws), falls below the long
term average, for that recurring trend then one can observe that same
trend being more popular; i.e. recurring more often until the short term
average becomes even with the long term trend (reversion to the mean). It is in combining these
trends that the game becomes interesting, more winnable.
Below is a Radar table where the
Max number is the draw that had the highest
"re-placement" number (i.e. how many draws back has the new Draw X been picked from or "re-placed" -- the five right-most columns of the
posi-tracker table map to draw 1 - draw 5 for a
daily view of series repeating). So the "4" three rows down
in the Max column shows a count of how many times that all 5 numbers were
previously picked from somewhere within the previous 4 draws. Similarly,
this table that was generated on 8/18/07 shows that the draw where all 5 numbers
will have already been picked in the previous 6 AND 8 draws is also due to
recur. By limiting the selection pool to only pick numbers from the
previous 4, or 6, or 8 draws (depending on the indicators) is where your
odds go way, way up. Predictions on the "re-placement" of the
posi-tracker table 5 right most columns can be made
by watching the pivot table for an
overview of all numbers repeating from any X days back.
| Max |
Last
100 draws |
Last
1000 draws |
All
n=4624 |
Percent |
| 2 |
0 |
0 |
2 |
0.04% |
| 3 |
1 |
5 |
12 |
0.26% |
| 4 |
0 |
9 |
37 |
0.80% |
| 5 |
3 |
17 |
88 |
1.90% |
| 6 |
1 |
17 |
96 |
2.08% |
| 7 |
5 |
38 |
152 |
3.29% |
| 8 |
2 |
37 |
186 |
4.02% |
| 9 |
7 |
51 |
214 |
4.63% |
| 10 |
7 |
62 |
242 |
5.23% |
| 11 |
4 |
58 |
255 |
5.51% |
| 12 |
8 |
54 |
251 |
5.43% |
| 13 |
5 |
55 |
272 |
5.88% |
| 14 |
3 |
45 |
238 |
5.15% |
| 15 |
4 |
54 |
268 |
5.80% |
| 16 |
5 |
57 |
235 |
5.08% |
| 17 |
5 |
34 |
189 |
4.09% |
| 18 |
4 |
38 |
197 |
4.26% |
| 19 |
4 |
35 |
191 |
4.13% |
| 20 |
5 |
39 |
162 |
3.50% |
| 21 |
1 |
23 |
136 |
2.94% |
| 22 |
1 |
28 |
136 |
2.94% |
| 23 |
1 |
29 |
135 |
2.92% |
| 24 |
3 |
21 |
107 |
2.31% |
| 25 |
1 |
15 |
92 |
1.99% |
| 26 |
1 |
25 |
90 |
1.95% |
| 27 |
5 |
22 |
77 |
1.67% |
| 28 |
2 |
13 |
63 |
1.36% |
| 29 |
1 |
18 |
65 |
1.41% |
| 30 |
0 |
12 |
54 |
1.17% |
Below
is a table of the distribution of Even versus Odd numbers being drawn of all draws since
the beginning until 8/11/07. Over the long run, it shows a slight bias towards odd
numbers, which makes sense since in the pool of 39 numbers there is one more
odd number than there are even numbers. See the daily dimensional
Odd or Even view table for
recent trends on odd versus even numbers; a commonly used and useful filter.
| Count
if Odd |
Percent
Odd |
Count
Of Odd |
Count
if Even |
Percent
Even |
Count
Of Even |
| 0 |
2% |
96 |
5 |
2% |
96 |
| 1 |
14% |
680 |
4 |
14% |
680 |
| 2 |
32% |
1579 |
3 |
32% |
1579 |
| 3 |
35% |
1718 |
2 |
35% |
1718 |
| 4 |
15% |
746 |
1 |
15% |
746 |
| 5 |
3% |
127 |
0 |
3% |
127 |
I recall a Math Professor in the
Economics department who once gave us an "expected value" for the
purchase of a ticket in the lottery. The table below extrapolates his
methodology for a five draw game with 39 numbers. The Unique Numbers 26,
20, 15 represent the smaller subsets that occur when picking with specific
criteria such as all 5 will be from the last 6, or 5 or even 4 draws.
| Unique
Numbers |
Count Correct |
Odds@Unique Number Count |
Prize Amount (Average) |
PrizeAmount*ProbabilityofOccurring |
| 39 |
All 5 in 39 |
0.00000174 |
$50,000 |
$0.09 |
| 39 |
4/5 in 39 |
0.000295247 |
$400 |
$0.12 |
| 39 |
3/5 in 39 |
0.009708738 |
$15 |
$0.15 |
| |
|
|
Sum: |
$0.35 |
| |
|
|
|
|
|
|
|
|
|
| Unique Numbers |
Count Correct |
Odds@Unique Number Count |
Prize Amount (Average) |
PrizeAmount*ProbabilityofOccurring |
| 26 |
All 5 in 26 |
0.00001520 |
$50,000 |
$0.76 |
| 26 |
4/5 in 26 |
0.001597444 |
$400 |
$0.64 |
| 26 |
3/5 in 26 |
0.032258065 |
$15 |
$0.48 |
| |
|
|
Sum: |
$1.88 |
| |
|
|
|
|
| |
|
|
|
|
| Unique Numbers |
Count Correct |
Odds@Unique Number Count |
Prize Amount (Average) |
PrizeAmount*ProbabilityofOccurring |
| 20 |
All 5 in 20 |
0.00006450 |
$50,000 |
$3.22 |
| 20 |
4/5 in 20 |
0.004830918 |
$400 |
$1.93 |
| 20 |
3/5 in 20 |
0.066666667 |
$15 |
$1.00 |
| |
|
|
Sum: |
$6.16 |
| |
|
|
|
|
| |
|
|
|
|
| Unique Numbers |
Count Correct |
Odds@Unique Number Count |
Prize Amount (Average) |
PrizeAmount*ProbabilityofOccurring |
| 17 |
All 5 in 17 |
0.00016160 |
$50,000 |
$8.08 |
| 17 |
4/5 in 17 |
0.009708738 |
$400 |
$3.88 |
| 17 |
3/5 in 17 |
0.111111111 |
$15 |
$1.67 |
| |
|
|
Sum: |
$13.63 |
| |
|
|
|
|
| |
|
|
|
|
| Unique Numbers |
Count Correct |
Odds@Unique Number Count |
Prize Amount (Average) |
PrizeAmount*ProbabilityofOccurring |
| 15 |
All 5 in 15 |
0.00033300 |
$50,000 |
$16.65 |
| 15 |
4/5 in 15 |
0.016666667 |
$400 |
$6.67 |
| 15 |
3/5 in 15 |
0.142857143 |
$15 |
$2.14 |
| |
|
|
Sum: |
$25.46 |
| |
|
|
|
|
| |
|
|
|
|
| |
|
|
|
|
| |
|
|
|
|
| Note:
For the odds column the professor used 1 divided the odds at 5/39 as an
input to expected value formula ((1/Prob(5,39)) * TopPrize) |
| Note:
for each of the Unique Number scenarios, I'm excluding the expected value
for a payoff of 2/5, where a replay ticket is issued as the prize. |
The lead
developer has also had good luck (matched 4/5 twice for
about $900) with the frequency tables, some other web
site a few years back mentioned that looking at the
frequencies of all 39 numbers gave good insights into which numbers would
recur. These tables will soon go into a
multi-dimensional cube (in the near future) which will
run against several different algorithms in a data
warehousing (BI) database.
This game is also intended as a
public service to show would be players when NOT to place a bet on a
given day (see the Anywhere
in last 6 draws table for info/guidance on when to skip a
wager). In that table when the 0, 1 or 2 columns looks like it
will recur, (show a '1' meaning the condition is true). The 0
(zero) column = 1 for that day means that no numbers recurred from the
last six draws. The '1' column = 1 for that day means that one
number recurred from the last six draws (that occurs roughly 10% of the
time and is a good indicator of when to perhaps not play that day).
The '2' column = 2 for that day means that two numbers recurred from the
last six draws (that occurs roughly 26% of the time and is also a good
indicator of when to perhaps not play that day).