Making random fight against itself

[Colbster]

Here is another system that I have been hashing out lately, one which is showing great initial promise (although we have all been there before, just in time for someone to give us that session that kills our hopes and dreams!).

The concept behind this is akin to inertia: once spins are in a trend, they will continue on that trend until random can manage to throw a complete stop and go back the other direction.  To begin with, we wait for random to generate a short-term and an intermediate-term trend that are in agreement.  For this system, I am using trends of 3 and 7 spins.  When both 3 and 7 spins show a majority towards 1 side of an EC, we wait until a single spin tries to buck the trend.  For instance, if we record HHLHLHHH, we have both 3 and 7 spins showing an advantage towards H.  We continue patiently tracking until we get an L, then we bet against it for the larger trends to continue.  Our new series of HHLHLHHHL is the trigger we need to bet H, trusting that random cannot pivot on a dime and head back the opposite direction of where it was already heading.

Obviously, we will lose some spins, although fewer than we would expect.  I have been playing this with the clean-up-before-you-move-up progression that is found here:
link:://rouletteforum.cc/index.php?topic=4727.0

I have not had a session yet exceed 3 losses before complete recovery, although this is early going.  I have attached my tracker spreadsheet with conditional formatting.  Enter your spins into the cells along row 1.  The tracker will already make the necessary tracking and adjustments to trigger bets.  After you have entered a spin, if the 3 cells directly below that spin are all green, bet the appropriate amount on High.  If they are all 3 orange, bet the appropriate amount Low.

Based on my initial results, flat betting would also be a winnner, but I like the increased returns using the progression that GLC gave us.  I have played this with a session bankroll of 20 units, but I haven't even threatened 10 yet.  I am expecting that 20 should be more than enough.  I am playing until I get bored, with +10 being a piece of cake.

[Nickmsi]

I like this kind of bet . . . tracking the short term trends.

Flat Bets Only:

I did a quick 10,000 spin test run, but I used all 3 EC's instead of just the High/Low.

It averaged 100 bets per 100 spins so a lot of betting for those that like action.

It showed a profit of 133 Units, which for those who follow Van Keelen, this passes his test of over 100 units profits for 1000 placed bets, which means this is a superior system, so far.

Nice job. .  . will continue to test . . . Nick

[birdhands]

"It averaged 100 bets per 100 spins so a lot of betting for those that like action."


Did you mean to say 100 bets per 1000 spins?
Thanks for your testing Nickmsi; I for one really appreciate it.

[Nickmsi]

Yes, Birdhands . .  you are correct, my fat little fingers skipped a zero.

100 bets for 1000 spins.

Thanks . . .

[marivo]

After you have entered a spin, if the 3 cells directly below that spin are all green, bet the appropriate amount on High.  If they are all 3 orange, bet the appropriate amount Low.

Do you bet just once and wait for new trigger?

[Colbster]

Yes, there is no continuation bet here.  Either a win moves the short term (1-spin) against our rules or a loss moves the medium term (3-spin) against us.  Either way, we wait for the next trigger.

[Nickmsi]

Another 10,000 spin update.  Attached graph now for 20,000 spins.

Still looking good.  Another 173 unit profit totaling 306 for both 10,000 sessions.

The 200 sessions show 93 Losing and 107 winning sessions, 53.5% .

Remember: This testing is Flat Betting.  No progressions.

Nick

[warrior]

Yes, there is no continuation bet here.  Either a win moves the short term (1-spin) against our rules or a loss moves the medium term (3-spin) against us.  Either way, we wait for the next trigger.

Could you give an ex.how to play this thanx.

[Nickmsi]

Hit a speed bump . . .

3rd 10,000 session was a loss of 143 units.

Will keep you posted.

[Colbster]

That is better than 2-to-1 winnings. I can handle that, especially since that us flat betting. Money management and a positive progression can improve that ratio. That we are up over 30,000 spins is a nice indication to me that this has merit.

[vundarosa]

Colbster, thx for sharing this....
could you show some spins on how you play this

vundarosa

[Nickmsi]

4th 10,000 session . . . another loss of 18 units.

Will do one more session to see best of 5.

I believe there is merit in using moving averages to determine short term trends and we should continue testing any and all aspects to see what we can find.

[Nickmsi]

5th Session of 10,000 spins showed another small loss of 8 units.

I'll now start checking our some possible tweaks that might improve our results and will report anything positive.

Moving averages .  . . great topic.    Nick

[superman]

Hi guys

I have spent this afternoon testing this method, I have trie dmany dofferent ways of progression and it keeps bombing sadly, the progression that came closest to being the best was the Oscars grind method, +1 on a win same bet on a loss BUT let's say for example we lose 5 times and are down 5 units we get a win so the bet is now 2 units, we win again (we are down 4 units now) +2 means we are down 2 units, next bet is 3 units as this will put us 1 unit ahead, if we get to a point where we are down 3 units and th eprogression is going to put us ahead more than 1 unit then we drop the progression so if we win we only make 1 unit.

Its like a safety brake so the progression doesnt get too out of hand, heres the code for those that can understand it

  If $currentbank <= $highestbank Then
   $progression = $progression+1
   If $progression + $currentbank > $highestbank+1 Then
    $progression = $highestbank-$currentbank+1
   EndIf
  Else
   $progression = 1
  EndIf

I know it wont help with this particular bet selection but someone may think of somewhere else they can use it, the problem with this bet selection is working on 3 and 7 may not be the right sample size, I did try different variations of the sample size they all come up with a loss.

[Colbster]

Thanks for testing, guys! I still like the concept - I may try integrating the two concepts and using two time frame moving averages to identify trend and still bet against random reversing itself.