Another question for the maths experts...

[Woodeneye]

Hi Fellas, this is my 1st post and I'll keep it brief. Could someone please tell me

what the chances are of betting on Victor's "Parachute System" and not scoring a hit?

e.g. 1 even chance bet
       1 @ 2-1
       3 @ 5-1
       2 @ 8-1
       4 @ 11-1
       6 @ 17-1
      18 @ 35-1

35 bets altogether.

I'm thinking of giving it a go, but would like to think the percentage was close to being in my favor. (well as close as you're gonna get when there's a 2.7% edge against you)

Thanks in advance; I enjoy reading this forum immensly.

[Kon-Fu-Sed]

Hello Woodeneye,

(and all interested)

Your mathematical probability to lose this 35-bets chain:
"even" bet - dozen or column - 3 x 6-bet - 2 x 4-bet - 4 x 3-bet - 6 x 2-bet - 18 x 1-bet (I hope that's what you showed above)
is somewhere around 5.06425%

As a study, I asked my computer-friend to run it through "my" Wiesbaden spins... To see if reality is different from math.

The bets:
1 x Low
1 x 1st Dozen
3 x double-street 1 - 6
2 x corner 1-2-4-5
4 x single-street 1 - 3
6 x split 1-2
18 x single 1

Start at spin #1 every day
New "session" immediately after win (or loss)
No recording of broken sessions at the end of the day

Spin #  Hits       %      Sum      %    Sum %
---------------------------------------------
  1    63863   49.26    63863   49.26   49.26
  2    21776   33.11    85639   16.79   66.06
  3     7286   16.56    92925    5.62   71.68
  4     6237   16.99    99162    4.81   76.49
  5     5136   16.85   104298    3.96   80.46
  6     2757   10.88   107055    2.12   82.58
  7     2527   11.19   109582    1.94   84.53
  8     1713    8.54   111295    1.32   85.85
  9     1572    8.57   112867    1.21   87.07
10     1412    8.42   114279    1.08   88.16
11     1350    8.79   115629    1.04   89.20
12      828    5.91   116457    0.63   89.84
13      798    6.06   117255    0.61   90.45
14      713    5.76   117968    0.55   91.00
15      656    5.62   118624    0.50   91.51
16      613    5.57   119237    0.47   91.98
17      533    5.13   119770    0.41   92.39
18      301    3.05   120071    0.23   92.62
19      255    2.66   120326    0.19   92.82
20      273    2.93   120599    0.21   93.03
21      254    2.81   120853    0.19   93.23
22      249    2.83   121102    0.19   93.42
23      242    2.83   121344    0.18   93.61
24      252    3.04   121596    0.19   93.80
25      253    3.15   121849    0.19   94.00
26      213    2.73   122062    0.16   94.16
27      235    3.10   122297    0.18   94.34
28      219    2.98   122516    0.16   94.51
29      184    2.58   122700    0.14   94.65
30      174    2.51   122874    0.13   94.79
31      206    3.05   123080    0.15   94.95
32      186    2.84   123266    0.14   95.09
33      176    2.76   123442    0.13   95.23
34      178    2.87   123620    0.13   95.36
35      160    2.66   123780    0.12   95.49
---------------------------------------------
  0     5845    4.50   129625  100.00 (Rounding error due to only 2 dec's shown of 15)
---------------------------------------------
                       129625 sessions

The first % shows hit% in relation to the the number of sessions that reached at least this level (should be like 48.6% for the "even" bet, 32.43% for the dozen, 2.7% for the singles...
The second % shows hits in relation to all sessions (129,625)

So this shows that those spins lost in only 4.5% - better than math!

...only...



Good luck
KFS


PS. Read about roulette probability in the Reference Area.

[TwoCatSam]

Kon

Not a math person!  Are you saying that you will lose 36 units 4.5% of the time?  That would mean 95.5%  you would win at least 1 unit and maybe more.  Surely you will get enough hits on the bets that return more than 1 unit to make a profit.

If you are saying this, this is astounding news!

Sam

[ChickenDinner]

While the mathematical probability looks good. A thing to consider with parachutes is how do you select the number?

Some wait for 4 numbers to show in 36 spins and then target it (hot), while others wait for a number to sleep for about 200 spins or more & then target it as soon as it wakes in the hope that statistical pressure will make it hot. Which works best, I don't know. But what I do know is you that you usually have to monitor the spins for a long time (hours sometimes) before you begin play (there are some online casinos which give running stats, though).

However, I don't know if others use this technique, but what I do with parachutes is simply target the last number spun. Every time I get a hit, whether it be on ECs, dozens, lines, etc, I switch to that number (unless it's the same number I'm targeting). What I like about this method is that you rarely get stuck with a sleeper and are constantly moving around the table. However, I must stress that this is how I play, so I'm not suggesting that anyone else should.

If you don't mind betting online with RNG wheels, what I would suggest, though, is that you find a casino with the 'speed spin' option, or at least Air Ball speed. Otherwise, if you go through the parachute sequence without a hit (this happens more than you would expect), it can be very slow and frustrating trying to recoup losses. I usually raise the bet by 1 unit after a loss and keep playing like this until I've broke even. If you're unlucky enough to go through another parachute without a hit before breaking even, my advice, for what it's worth, is quit for the day.

CD

[Kon-Fu-Sed]

Hi Sam,

Don't you forget something? Each loss costs you 35u.

In 200 sessions you will lose 9 times = 315u (I use real numbers - not math; that's 10 losses... )
And in something like 97 sessions the "even" bet will hit giving you 1u.
Then there is that dozen also, giving only 1u - in like 32 times...
So in the rest of the sessions - 62 - you have to win at least 187u (186u + 1 net = more than 3u in average...)

Sorry; no astounding news
KFS

PS. Forgot to stress: The 4.5% is valid for THOSE spins only. Other spins may give other results (like a 5.5% loss... why not?)

[TwoCatSam]

KFS

Well, I said I was not a math person.  Here's how I read that chart.  Tell me in this first step where I'm going wrong.

When you start a parachute you will make it through 95.5% of the time, with a profit of at least one chip.

Is that wrong?

Now, look at this:

1    63863   49.26    63863   49.26   49.26
2    21776   33.11    85639   16.79   66.06

Does that not mean that 66.06% of the time, you will hit on low and if not low then on the first dozen?

Sam

[Kon-Fu-Sed]

Sam,

You read the table correct.
So, say that you hit those two 67 times in 100 sessions. That's + 67u.
You also know that you will lose ... OK 4 times (that's why I used 200 sessions to have 9 inst of 4.5 but that's OK)
Those four times cost you 4 x 35 u = 140u.
You had won 67u on Low and Dozen so that's 73u in the red. OK?

Now you have "used" 71 sessions on the Low hits, the Dozen hits and the losses.
So of 100 sessions you have 29 left and you have to get 73u + 1u net win = 74u. = more than 2.5u per hit.

OK... that's less than the three I said above but here I was also rather generous with the hits and mean(?) with the losses...

And again: This result was only valid for those spins in that size. <--- I stress this!

The math loss is 5 times in 100 sessions and - maybe I'm wrong here but - performing more studies will eventually be closer to that.

I dare promise you that.
KFS

[TwoCatSam]

OK, I'm clear now.  I should stay out of such forests!

Thank you.  Maybe I can return a favor; instruct you on how to fix your 'fridge!

LOL

Sam

[Kon-Fu-Sed]

LOL

No worries man - I'm happy I could sort it out for you!

/KFS

[Wildcard]

 That being said ...........  


I have a new one for you KFS, if you should be so nice and use some brain fuel on this. lol

My bankroll is so low that I could only post it on the HUMOUR section. But somehow YESTERDAY I managed to lower it by losing while flat betting.

Now, that "desperate" situation got me thinking. Using a good old Marty based upon 3 Lines ( Line = 6 numbers, right ? ), so using 18 numbers total, I came up with this:

Each of the 3 Lines run a separate Martingale at the same time so betting would be:

  1-1-1
  2-2-2
  4-4-4 and so on...     I haven´t set a limit, but you will see why further down.

Bet selection >> The last 3 lines that came out. We keep these 3 until a win.

After a win, we look back at the "new" last 3 Lines that came out and go from there. 

Any time we bet 1-1-1 and win, we begin another run on the same Lines.

So easy, yet so dangerous, right ? Yeah, I know that part, I am fully aware, don´t worry.

What this system does is look for a repetition on one of the " hot Lines " (not the dirty ones  ) either from repeating numbers or from warm/cold numbers WITHIN those Lines, so I think it makes a good mix as bet selection.

Now, here´s the big question... don´t really know how to put it....

How long can we go without a hit ... or how many steps of the Martingale would be necessary to win everytime ?

I suspect this can´t be calculated.... It´s like betting an EC (18 numbers) but we have a mix of numbers INSIDE hot Lines.

Confused ? I hope so ...  he he he he

For what it´s worth I played it live online ... made 60* units in about 40 minutes or so (i wasn´t clocking the session) without going to high on the progression.

Before playing I did 2 tests on live spins and I was pretty impressed, because it didn´t go as far as one might expect.

Anyway, i´d like to hear from you or anyone else reading.  It´s simple and effective... HOW effective is what i´d like to know.


*EDIT: 60 units is actually 30 since i was playing 2 chips as "unit base".... so 30 units in about 40 mins. There !

[Wildcard]

Just finished another session... 30 mins. Bet selection is ultra-fast, of course.

Profit = 30 units  ( remember the way i played this means 30 units of 2 chips, so 60 chips profit )

I started all attacks with  2-2-2 and once it got to the 32-32-32 stage.

If it had been 1-1-1 as per the plan, the max amout would have been 16-16-16.

Anyway, it´s a Marty progression, so that really spells D-O-O-M..... Any time soon 

[Kon-Fu-Sed]

Hello Wildcard,

and all curious (if any)


Wildcard, I can't believe you are asking me this question... because I don't think you're going to like the answer. You obviously haven't read my text on probabilities (in the Reference Area)...

If you had, you would know my answer: You cover 18/37 so your probability to hit is 18/37. Always.
And it doesn't matter if you bet Red, three 6-bets (double-streets) or six 3-bets (single streets) or whatever makes up 18 numbers.
Or chase "hot", "medium" or "mild" numbers.

Or mix all that as you please, from spin to spin. 18 / 37. Always.


BTW: You ask for a SYSTEM-TEST. I don't like to do TESTS as I can as well use a calculator or Grabb's tools in a fraction of the time it takes for me to write the rules, send them to my friend, for him to write and run a program and send the output to me to check for errors...

All that time for what?

For the exact same result (+-3 SD ) that I get on my calculator or Grabb's tools in a few minutes.
For your system it was simple: 18/37 for each spin. Two seconds.

But if I just said that it's the same as betting Black - would you believe me?
Well, maybe YOU would but a whole lot of others would say that hot numbers hit like... And that I'm wrong.

Now, I'm an honest guy so I have to say it anyway: You bet 18 numbers so you have the same chance to hit as if you bet any other 18-numbers bet.


But please read on - even if you curse my narrow-minded mathematical thinking now... You see: I asked my friend to run your system through the Weisbaden spins I downloaded. This is the beginning of the file - spin by spin:

01.04.2003
Spin#  3 last DS   BetCount + 3 bets  Hit#  DS
------------------------------------------------------------
   1   0   0   0 |                      23   4
   2   4   0   0 |                       1   1
   3   4   1   0 |                      25   5
   4   4   1   5 |   1   4   1   5       4   1  <-- Hit    1
   5   4   1   5 |   1   4   1   5      16   3
   6   1   5   3 |   2   4   1   5      17   3
   7   1   5   3 |   3   4   1   5      19   4  <-- Hit    2
   8   5   3   4 |   1   5   3   4       0
   9   5   3   4 |   2   5   3   4      15   3  <-- Hit    3
  10   5   3   4 |   1   5   3   4      17   3  <-- Hit    4
  11   5   3   4 |   1   5   3   4      15   3  <-- Hit    5

First the spin, then the three last DSs.
Then the BetCount and the three bet DSs.
Then the hitting number for the spin and the DS it belongs to.
Last are the "hit-notes".

OK?

The system-test was performed like this:
Each day starts the same as the above: "Empty", waiting for three different DSs.
The first day (01.04.2003) is completely written to the file, plus 12 more spins from the next day (for control-purposes).
After that, a table is printed for the first day, making it possible to check that everything is in that table.

Then the test continues to the end-of-file (some 1900 days in total) and a complete table is written.

The first few lines of the 1900-days table:

Spin#    Hits   18/37 Tot hits   Real%   Tot %   Tot% from Grabb's tool
-----------------------------------------------------------------------
   1   154602   48.92   154602   48.92   48.92   48.64864864864865
   2    78578   48.69   233180   24.86   73.79   73.63038714390066
   3    40478   48.88   273658   12.81   86.60   86.45884745227332

The "18/37" column should according to math show 48.65 as that is the probability to hit 18/37 in percent. This is the true %.
The "Grabb %" column was copied from Grabb's "Hit or Sleep" tool for comparison.


But that's not all! As I suspect that someone will ask for it, I made (no - my friend did, I asked...) another test - the Single-Streets (3-numbers bets).
Here is the result, shown exactly as above:

01.04.2003
Spin#      6 last SS            BetCount + 6 bets             Hit#  DS
----------------------------------------------------------------------
   1   0   0   0   0   0   0 |                                  23   8
   2   8   0   0   0   0   0 |                                   1   1
   3   8   1   0   0   0   0 |                                  25   9
   4   8   1   9   0   0   0 |                                   4   2
   5   8   1   9   2   0   0 |                                  16   6
   6   8   1   9   2   6   0 |                                  17   6
   7   8   1   9   2   6   0 |                                  19   7
   8   8   1   9   2   6   7 |   1   8   1   9   2   6   7       0
   9   8   1   9   2   6   7 |   2   8   1   9   2   6   7      15   5
  10   1   9   2   6   7   5 |   3   8   1   9   2   6   7      17   6  <-- Hit    1
  11   1   9   2   6   7   5 |   1   1   9   2   6   7   5      15   5  <-- Hit    2
  12   1   9   2   6   7   5 |   1   1   9   2   6   7   5       3   1  <-- Hit    3
  13   1   9   2   6   7   5 |   1   1   9   2   6   7   5      12   4
  14   9   2   6   7   5   4 |   2   1   9   2   6   7   5      29  10
  15   2   6   7   5   4  10 |   3   1   9   2   6   7   5      24   8
  16   6   7   5   4  10   8 |   4   1   9   2   6   7   5      34  12
  17   7   5   4  10   8  12 |   5   1   9   2   6   7   5      24   8
  18   7   5   4  10   8  12 |   6   1   9   2   6   7   5      36  12
  19   7   5   4  10   8  12 |   7   1   9   2   6   7   5       5   2  <-- Hit    4


And the start of the results table looks like this:

Spin#    Hits   18/37 Tot hits   Real%   Tot %   Tot% from Grabb's tool
-----------------------------------------------------------------------
   1   152705   48.89   152705   48.89   48.89   48.64864864864865
   2    78079   48.91   230784   25.00   73.89   73.63038714390066
   3    39816   48.83   270600   12.74   86.64   86.45884745227332

If you compare the "Tot %" to "Grabb's" there's no big difference in none of the tests. And not mutually either. And those figures are the exact same as I get on my calculator in no-time.

So I ask the same questions as I do in my probability-text:
* Do you think math is very different from reality?
* Do we really need to do such tiresome, time-consuming empirical studies or can we use a calculator?


Now, this is a discussion area. There is a Testing Zone also so I post the complete output from both tests there.
You will find them.


Now, maybe you don't like the message but please PLEASE don't shoot the messenger...
KFS

[Wildcard]

I don´t shoot the messanger. I invite him to sit down and have a drink.

I LIKED the answer, KFS, it was brilliant, i was able to set aside a whole bunch of misconceptions i had until now, so i am nothing but thankful. Yes, sir. 

And you are right, i haven´t read your texts in the Reference Area, mainly because i´ve had little time to devote to reading, but i will start right away.... at least i can do it little by little and i am positive i´ll enjoy it.

I know it was similar to an EC system, remember i wrote:

"I suspect this can´t be calculated.... It´s like betting an EC (18 numbers) but we have a mix of numbers INSIDE hot Lines."

But i learned some more, and that´s healthy.

Now that i am done with bet selection, as it really doesn´t make a difference, it´s time to move on other grounds, such as money management.

You are a constructive/destructive person.... lol ... let me explain.... you destroyed my false ideas and gave me pointers to the ones i should be focusing on, so that´s constructive.

                                        [smiley=Best_of_fichiers/4_17_2.gif]

[Woodeneye]

A big thanks to KFS for all the calculations on the Parachute System; and to the rest

of you fellas for your input. Maybe I won't dive in with real money just yet!

ChickenDinner- may I ask how long you've been playing the Parachute? Is it your

preferred method of play and would you be in front from it?

Sorry to sound nosey.

[Kon-Fu-Sed]

@Woodeneye,
Glad to help.

@Wildcard,
Thanks for the drink
Bringing news like that is not always appreciated.
I'm relieved it was in this case.


KFS

[Kon-Fu-Sed]

-