Welcome Guest
You last visited December 6, 2016, 9:00 am
All times shown are
Eastern Time (GMT-5:00)

So close, so far away

Topic closed. 6 replies. Last post 3 years ago by adobea78.

 Page 1 of 1
Florida
United States
Member #135615
November 27, 2012
405 Posts
Offline
 Posted: April 11, 2014, 11:32 pm - IP Logged

So I’m currently testing out various patterns. They treat me better than predicting based on number stats. However, I seem to keep hitting the same wall. There are so many patterns that work well but I keep having trouble with covering the straight. Let me give an example. For no good reason other than it’s the code I currently have up, I’m using Florida’s numbers (Day and Eve) for 2010. Take a simple pattern like doubles. Here’s an example…

In 2010, the first double is on 1/2 Mid: 2-2-5

This easy pattern is to take the double “2” and the other number “5” and look for it to repeat. In this case, it hits on 1/5 Eve 5-1-2. To get a hit, I need to play the 2 and 5 and then cover the last position with all numbers. If I play that boxed, that will give me 10 sets. If I play it straight, that gives me 54 numbers. Either way, I’ll end up losing money. I coded this up so I could see exactly what’s going on. Here’s the breakdown:

In 2010, 214 doubles occur. If you use the above system and played for 5 days, you’d get 112 hits. If you played for 10 days, you would have gotten another 109 hits (211 total). 15 days would get you another 115 hits (324 total). It keeps going out from there, 20 days = 101, 25 = 97, 30 = 120, so on and so forth. So let’s just pick a stopping point for good measure so we can see what the cost/win amounts are. I’ll go with playing 5 days out. So when a double would hit, I’d play the double, the off number, and then 0-9.

The cost of playing boxed for 5 days would be 10 (draws = 5 days) * 10 (sets) * 214 (this is the number of times a double occurred and thus, how many times I would have had numbers to play) = \$21,400. As for wins, in Florida, a 6-way win will get you \$80. I would’ve had 112 hits using this pattern. So \$80 * 112 = \$8,960. I may have gotten some 3-way wins in there which pay out \$160 but I’m going to go ahead and assume that wouldn’t have made up the difference.

The cost of playing straight for 5 days is where it gets interesting, or rather, possibly interesting. To cover a straight, one needs to play 54 numbers, because we don’t know the order/position they’re going to hit in. So doing the math 54 (sets we play) * 10 (draws = 5 days) * 214 (this is the number of times a double occurred and thus, how many times I would have had numbers to play) = \$57,780 (cost to play). Straight winnings would be \$500 (what we get in Florida for a straight) * 112 (number of hits the pattern would’ve made) = \$56,000. Still a loss… Keep in mind that some years perform better than others so doing anything to tilt the scale toward the net positive may have worked for 2010 but not another year.

To make this actually work, there are two options that I can think of. One could either find a way to put the sets in order (this way, we’d play 10 sets at a time instead of 54) or not play the full ten numbers as the last position (if we knew which number to play in the last position, we could cover it for a straight and only need to play 6 sets). I haven’t found a way to do either in a way that works enough to give me a profit worth playing for.

If anyone is still reading this little novel, I’d love to hear any ideas on how to find the last position and/or ordering the positions. Either way will result in a system that is profitable.

United States
Member #128790
June 2, 2012
5431 Posts
Offline
 Posted: April 11, 2014, 11:47 pm - IP Logged

So I’m currently testing out various patterns. They treat me better than predicting based on number stats. However, I seem to keep hitting the same wall. There are so many patterns that work well but I keep having trouble with covering the straight. Let me give an example. For no good reason other than it’s the code I currently have up, I’m using Florida’s numbers (Day and Eve) for 2010. Take a simple pattern like doubles. Here’s an example…

In 2010, the first double is on 1/2 Mid: 2-2-5

This easy pattern is to take the double “2” and the other number “5” and look for it to repeat. In this case, it hits on 1/5 Eve 5-1-2. To get a hit, I need to play the 2 and 5 and then cover the last position with all numbers. If I play that boxed, that will give me 10 sets. If I play it straight, that gives me 54 numbers. Either way, I’ll end up losing money. I coded this up so I could see exactly what’s going on. Here’s the breakdown:

In 2010, 214 doubles occur. If you use the above system and played for 5 days, you’d get 112 hits. If you played for 10 days, you would have gotten another 109 hits (211 total). 15 days would get you another 115 hits (324 total). It keeps going out from there, 20 days = 101, 25 = 97, 30 = 120, so on and so forth. So let’s just pick a stopping point for good measure so we can see what the cost/win amounts are. I’ll go with playing 5 days out. So when a double would hit, I’d play the double, the off number, and then 0-9.

The cost of playing boxed for 5 days would be 10 (draws = 5 days) * 10 (sets) * 214 (this is the number of times a double occurred and thus, how many times I would have had numbers to play) = \$21,400. As for wins, in Florida, a 6-way win will get you \$80. I would’ve had 112 hits using this pattern. So \$80 * 112 = \$8,960. I may have gotten some 3-way wins in there which pay out \$160 but I’m going to go ahead and assume that wouldn’t have made up the difference.

The cost of playing straight for 5 days is where it gets interesting, or rather, possibly interesting. To cover a straight, one needs to play 54 numbers, because we don’t know the order/position they’re going to hit in. So doing the math 54 (sets we play) * 10 (draws = 5 days) * 214 (this is the number of times a double occurred and thus, how many times I would have had numbers to play) = \$57,780 (cost to play). Straight winnings would be \$500 (what we get in Florida for a straight) * 112 (number of hits the pattern would’ve made) = \$56,000. Still a loss… Keep in mind that some years perform better than others so doing anything to tilt the scale toward the net positive may have worked for 2010 but not another year.

To make this actually work, there are two options that I can think of. One could either find a way to put the sets in order (this way, we’d play 10 sets at a time instead of 54) or not play the full ten numbers as the last position (if we knew which number to play in the last position, we could cover it for a straight and only need to play 6 sets). I haven’t found a way to do either in a way that works enough to give me a profit worth playing for.

If anyone is still reading this little novel, I’d love to hear any ideas on how to find the last position and/or ordering the positions. Either way will result in a system that is profitable.

I don't get it, can you do Florida?

Park City, UT
United States
Member #69864
January 18, 2009
993 Posts
Offline
 Posted: April 11, 2014, 11:55 pm - IP Logged

So I’m currently testing out various patterns. They treat me better than predicting based on number stats. However, I seem to keep hitting the same wall. There are so many patterns that work well but I keep having trouble with covering the straight. Let me give an example. For no good reason other than it’s the code I currently have up, I’m using Florida’s numbers (Day and Eve) for 2010. Take a simple pattern like doubles. Here’s an example…

In 2010, the first double is on 1/2 Mid: 2-2-5

This easy pattern is to take the double “2” and the other number “5” and look for it to repeat. In this case, it hits on 1/5 Eve 5-1-2. To get a hit, I need to play the 2 and 5 and then cover the last position with all numbers. If I play that boxed, that will give me 10 sets. If I play it straight, that gives me 54 numbers. Either way, I’ll end up losing money. I coded this up so I could see exactly what’s going on. Here’s the breakdown:

In 2010, 214 doubles occur. If you use the above system and played for 5 days, you’d get 112 hits. If you played for 10 days, you would have gotten another 109 hits (211 total). 15 days would get you another 115 hits (324 total). It keeps going out from there, 20 days = 101, 25 = 97, 30 = 120, so on and so forth. So let’s just pick a stopping point for good measure so we can see what the cost/win amounts are. I’ll go with playing 5 days out. So when a double would hit, I’d play the double, the off number, and then 0-9.

The cost of playing boxed for 5 days would be 10 (draws = 5 days) * 10 (sets) * 214 (this is the number of times a double occurred and thus, how many times I would have had numbers to play) = \$21,400. As for wins, in Florida, a 6-way win will get you \$80. I would’ve had 112 hits using this pattern. So \$80 * 112 = \$8,960. I may have gotten some 3-way wins in there which pay out \$160 but I’m going to go ahead and assume that wouldn’t have made up the difference.

The cost of playing straight for 5 days is where it gets interesting, or rather, possibly interesting. To cover a straight, one needs to play 54 numbers, because we don’t know the order/position they’re going to hit in. So doing the math 54 (sets we play) * 10 (draws = 5 days) * 214 (this is the number of times a double occurred and thus, how many times I would have had numbers to play) = \$57,780 (cost to play). Straight winnings would be \$500 (what we get in Florida for a straight) * 112 (number of hits the pattern would’ve made) = \$56,000. Still a loss… Keep in mind that some years perform better than others so doing anything to tilt the scale toward the net positive may have worked for 2010 but not another year.

To make this actually work, there are two options that I can think of. One could either find a way to put the sets in order (this way, we’d play 10 sets at a time instead of 54) or not play the full ten numbers as the last position (if we knew which number to play in the last position, we could cover it for a straight and only need to play 6 sets). I haven’t found a way to do either in a way that works enough to give me a profit worth playing for.

If anyone is still reading this little novel, I’d love to hear any ideas on how to find the last position and/or ordering the positions. Either way will result in a system that is profitable.

The best option would be to play online where the house edge is only 10% versus 50%.

The median value for a 1 number from a draw to to repeat by position is 2 draws.

The median value for 2 numbers from a draw to repeat by position is 6 draws.

The median value to have all 3 numbers from a draw to repeat by position is 15 draws.

So you might be able to use that information to eliminate some of the 54 combinations to play.

Jimmy

New Member

United States
Member #154220
April 8, 2014
7 Posts
Offline
 Posted: April 12, 2014, 3:12 am - IP Logged

Here's what I did to limit the plays and maximize profits.
When testing th theory ( on a spreedsheet) I mad note of how the winners laid out.

In this example the 2 is the A column , 2 is the B col, 5 is the C
you shoud look for the "hit", lets say the hit was 5- 2 - 9
I would show this hit as CAx

a hit of  7-2- 5 would be xAC or xBC, a hit of  5 9 2 would be CxA

then count the hits. In my case (TEXAS) I found that limiting my plays to  AC*  and CA* (using my trigger "system") had close to a 40% Hit rate.

so when a play is triggered I play those 20 tickets. Play 10 games, the first five games for 25 cents (on line) games 6-10 for 50 cents

If you need to play (and lose) all 10 games at 20 tickets per game using the 25 cent and 50 cent amounts, the total loss is \$75.

25 cents straight produces a \$225 win, while 50 cents pays \$450.

Max consecutive losses is 4 (in my case)  Do the math and you should find a nice profit.

BTW, I also have a 2nd trigger system that I Box AC* (ten tickets)for \$1 & then \$2  and play only 7 games (on line box pay out is \$150 & \$300)

Also look at your hit % and if high enough you can double your bet after a loss and keep it at that \$ amount until you hit.

Good Luck

Florida
United States
Member #135615
November 27, 2012
405 Posts
Offline
 Posted: April 18, 2014, 6:54 pm - IP Logged

Jim, I'm afraid I didn't understand your suggestion.

Ron, I wrote a script that would tell me the order of hits. Depending on the date range, there was typically a clear winner (one order occurring more than the other). However, I didn't see a way for that to help me put the numbers in order. There are 6 order possibilities, so using that data, I could remove the order that's occurring the most and least. That still leaves me 4 possibilities. So 9*4 = 36. 36 sets per play to cover all. Or just the 9 and pick one and hope that's it. If I could get the pick right half the time, I could make money. I just don't think my luck is that good.

So since the \$500 win seemed to be something I could change. I tried Pick 4. I tried a pattern out. It's hitting well but producing 24 numbers for a box hit. Again, I don't know what order to put them in. The testing is a lot slower becaus I'm using an Excel sheet. I don't have Pick 4 data in my database... So trying to cover 24 numbers for Pick four would eat away at the profits just like with Pick 3. Trying to get luckly and hope I chose the right order doesn't sound very profitiable... Maybe I should go to a 5 number system. Order doesn't matter and you still get something on a miss...

Florida
United States
Member #135615
November 27, 2012
405 Posts
Offline
 Posted: April 18, 2014, 7:04 pm - IP Logged

For example, here's a sample for the pattern I'm currently working with. This is for P4, FL.

 9441 9444 9446 9411 9414 9416 9451 9454 9456 9461 9464 9466 9741 9744 9746 9711 9714 9716 9751 9754 9756 9761 9764 9766

Hits are good but I need to either figure out how to make sure that's a straight. To do that, these sets need to be further reduced or I need a way to order the positions. It doesn't have to be 100%, just enough for a profit.

United States
Member #116344
September 8, 2011
3926 Posts
Online
 Posted: April 18, 2014, 8:28 pm - IP Logged

For example, here's a sample for the pattern I'm currently working with. This is for P4, FL.

 9441 9444 9446 9411 9414 9416 9451 9454 9456 9461 9464 9466 9741 9744 9746 9711 9714 9716 9751 9754 9756 9761 9764 9766

Hits are good but I need to either figure out how to make sure that's a straight. To do that, these sets need to be further reduced or I need a way to order the positions. It doesn't have to be 100%, just enough for a profit.

Try 24 positional triads, eg 1234  set will be four combo 123,124,134,234, each combo will permutate to 6 positional triads 123, 132, 213, 231, 321, 312, your x could be the most frq digits  of last combined draws, from  the test, 9456  appears  to be the trend.  A repeating combo triads permutates 3 positional triads(for doubles),eg 944  will be two combos 994,944 for 6 positional triads 994,949,499,944,494,449, the bottom line  is  turn the patterns into positional triads.

 Page 1 of 1