Regression Betting

[GLC]

To all reading this, I am sorry about the charts not lining up very well.  I didn't write this.  It's just a copy and the formatting didn't transfer correctly.  If you study the charts a little, you should be able to recreate them if you're that interested.  I started manually lining up the 1st ones, but it's such a time consuming thing that I gave up.  If some of you are really, really interested it picking this apart with a fine tooth comb, let me know and I'll make sure you get the charts in understandable form.


Aside from the minuted details, there's still a lot of good points made by the author that should be considered by anyone who wishes to win at roulette.  Many of you will find them old news but some of the newer members should get a lot out of it.


Enjoy the read,


GLC


******************************************************************************




First, let me begin with a short preamble. I strongly recommend only playing on real
tables rather than online casinos whenever possible. Real tables, in a municipality
that inspects them on a regular basis â€" like Vegas, are harder to gaff. I’m a
programmer, and it is relatively easy to make a program behave non-randomly.
Second, only play on single 0 wheels; preferably with En Trap rules with the lowest
minimums and highest table limits you can find. This reduces the vig as much as
possible. Third, don’t forget in many casinos, you can designate any value you want
to the chips, as long as the total outlay meets the table minimum â€" say $0.05 per chip
if you have large outlays. This will help reduce the bank required and increase the
amount it will take to bust you. Finally, when playing on real tables, like any
mechanical device they are subject to wear when used incessantly. This can lead to
distinct patterns for individual wheels that compromise their randomness. Finding
and charting them is fairly easy, and doesn’t take too much time. Secrets of Winning
Roulette by Marten Jensen has an excellent explanation.


Up & Pull


This system was found in So You Wanna Be a Gambler! - Advanced Roulette By John
Patrick. © 1987. He is a eccentric writer who berates the reader for being stupid, but it
does have a couple of good systems.


Up & Pull is a regressive system that has returned fair results, but since it is an even
money system, on roulette it still has less than a 50% chance to win (48.64865% on a
single 0 without En Prison or La Partage rules), and Mr. Patrick admittedly relies heavily
on trends (aka series, or runs), however this can work.


Quote from book: “Up and pull is the ability to get the most that you can out of a winning
streak. Every winning hand or spin should result in a profit and an opportunity to increase
your bet against the house â€" hence the term “up and pull.”


Let’s take a typical series. You’re at a $5 table, and your first bet is $8, you win and get
paid $8. Your next bet is $5, as you want to regress your bet down to the minimum and
wrap up a profit of $3.


Now you’re in a real good position. Cause even if the next hand is a loss, you still have a
profit for that series. The important thing now is: what is your next bet? In this case, the
options are many. But what you should do is swing over to my system called up and pull.
Since you’re in a streak, and a streak begins with the first win, you wanna take advantage
of that run by increasing your bets after each score. Most of all â€" and I emphasize the
words â€" MOST OF ALL you have to pull back a profit.


If every person who gambled would take a profit every time he or she scored a win, the
casinos would have to resort to taking up collections to stay in business.


Go back to that example and follow these simple steps:
Bet $8 and you won;
Pull back $8 and regress down to $5;
Next hand wins, and you revert to up and pull. Up your bet to $7 and pull back $3 profit;
Bet $7 and again you win. Increase your bet to $10 and pull back profit of $4.
See what’s happening? You’re increasing your bets, but most of all you’re pulling back a
couple of bucks as you go along.”


If you are at the first round, and you loose, just re-bet again at level 1. Don’t chase
losses: in this system, you are chasing wins.


Here is a chart of two suggested progressions for a $1 min table:
3.1.2.3
Bank   TB   Profit  ReInvest   Round     Loss
-3     3      6          1      1         -3
2      1      2          1      2*         1
2      2      4          1      3          0
3      3      6          1      4**        0
5      4      8          1      5          1
8      5     10          1      6          3
12     6     12          1      7          6
17     7     14          1      8          10
23     8     16          1      9          15
30     9     18          1      10         21
38     10    20          1      11         28
47     11    22          1      12         36
57     12    24          1      13         45
68     13    26          1      14         55
80     14    28          1      15         66
93     15    30          1      16         78
107    16    32          1      17         91
122    17    34          1      18         105
138    18    36          1      19         120
155    19    38          1      20         136


3.1.1.2.3
Bank   TB   Profit   ReInvest  Round      Loss
-3     3      6           1     1          -3
2      1      2           0     2*          1
3      1      2           1     3**         2
3 2 4 1 4 1
4 3 6 1 5 1
6 4 8 1 6 2
9 5 10 1 7 4
13 6 12 1 8 7
18 7 14 1 9 11
24 8 16 1 10 16
31 9 18 1 11 22
39 10 20 1 12 29
48 11 22 1 13 37
58 12 24 1 14 46
69 13 26 1 15 56
81 14 28 1 16 67
94 15 30 1 17 79
108 16 32 1 18 92
123 17 34 1 19 106
139 18 36 1 20 121


Mr. Patrick strongly encourages money management, and suggests a series be abandoned
if there are 3 consecutive losses, and to find another table.


Up & Pull 5:1 variation with regression variation
This theory can be applied to any of the positions with a full hedge. I am partial to the
5:1 double street position for several reasons. If you play at a full hedge (5 of the 6
available positions), then the chances of winning any single roll is 81.0811% (on a singlezero table). Also, tests I’ve run that show the average win/loss ratio is around 4.278:1 which is a significant improvement over the 1:1 positions at 0.931:1 or 2:1 position at 1.831:1. Also, the occurrences of losing streaks of 3 or more is only around 3.5% of all
loosing streaks.


The basic progression is listed below:
5.0 base


Bank CS TB TW Profit ReInvestRoundLoss
(750) 150 750 900 150 50 1 (750)
100 10 50 60 10 5 2 50
105 11 55 66 11 5 3 50
111 12 60 72 12 5 4 51
118 13 65 78 13 5 5 53
126 14 70 84 14 5 6 56
135 15 75 90 15 5 7 60
145 16 80 96 16 5 8 65
156 17 85 102 17 5 9 71
168 18 90 108 18 5 10 78
181 19 95 114 19 5 11 86
195 20 100 120 20 5 12 95
210 21 105 126 21 5 13 105
226 22 110 132 22 5 14 116
243 23 115 138 23 5 15 128
261 24 120 144 24 5 16 141
280 25 125 150 25 5 17 155
300 26 130 156 26 5 18 170
321 27 135 162 27 5 19 186
343 28 140 168 28 5 20 203
366 29 145 174 29 5 21 221
390 30 150 180 30 5 22 240
415 31 155 186 31 5 23 260
441 32 160 192 32 5 24 281
468 33 165 198 33 5 25 303
496 34 170 204 34 5 26 326
525 35 175 210 35 5 27 350
555 36 180 216 36 5 28 375
586 37 185 222 37 5 29 401
618 38 190 228 38 5 30 428


Since the likelihood of getting all the way to level 30 is very small, this system is ideal
for inclusion of partial regressions as illustrated below:


Bank CS TB TWProfitReInvestRoundLoss
-375 75 375 450 75 25 1 -375
50 5 25 30 5 5 2 25
50 6 30 36 6 5 3 20
51 7 35 42 7 5 4 16
53 8 40 48 8 5 5 13
56 9 45 54 9 5 6 11
60 10 50 60 10 -20 7 10
90 6 30 36 6 5 8 60
91 7 35 42 7 5 9 56
93 8 40 48 8 5 10 53
96 9 45 54 9 5 11 51
100 10 50 60 10 -20 12 50
130 6 30 36 6 5 13 100
131 7 35 42 7 5 14 96
133 8 40 48 8 5 15 93
136 9 45 54 9 5 16 91
140 10 50 60 10 -20 17 90
170 6 30 36 6 5 18 140
171 7 35 42 7 5 19 136
173 8 40 48 8 5 20 133
176 9 45 54 9 5 21 131
180 10 50 60 10 5 22 130
185 11 55 66 11 5 23 130
191 12 60 72 12 5 24 131
198 13 65 78 13 5 25 133
206 14 70 84 14 5 26 136
215 15 75 90 15 5 27 140
225 16 80 96 16 5 28 145
236 17 85 102 17 5 29 151
248 18 90 108 18 5 30 158


The are two disadvantages of regressing. The first is when you do get a long winning
streak of 20 or more, the total win will be less, but for the thousands of streaks you will
get in the meantime, this will create a higher return from less wins. The second is that
you are essentially paying to regress if the loss comes after the regression as you can see
from the blue squares. If you loose, you still show a profit, but the profit is less than if
you had lost on the previous spin. I feel that the loss is nominal for the gain.


Finally, mathematically, the change of one double street hitting is 16.2162%. The chance
of that double street hitting twice in a row is only 2.6297%. Logically, if one were to
place bets where the double street that just hit is the street that is not bet upon for the next spin, then the chances of that double street hitting twice in a row is significantly less than just covering the same numbers every spin (say 1-30). However, my tests have shown
that the occurrence of losses and loosing streaks by fleeing the winner is only nominally
less than betting the same numbers every time (the Win/Loss ratio increases by about
19%). This is probably because the occurrence of a number hitting in the first double
street (1-6) twice is a loss if you flee the winner, but a win for the player who stands on
his/her numbers 1-30. Conclusion: why not flee the winner? It does help. It can be
annoying for online casino play, but in a real casino, you must replace the bets every spin
anyway, so what could it hurt?


I’ll conclude with the note that there is one distinct advantage to this system: the first loss doesn’t cost any money (unless it comes at the very first bet), it just ends the sequence. Therefore it effectively extends the loss streak you can withstand by 1. In other words, a series of 3 losses will only cost you money on the second and third loss. And, at the 5:1  position, around 80% of the loss streaks consist only of 1 consecutive loss, and only 20%  consist of 2 or more consecutive losses. On the other hand, the major disadvantage is recouping the initial outlay if the first spin looses. With no controls for limiting loosing streaks, simulations have shown this to be a consistently loosing system.


5:1 Half Progression


This is one of the systems I have developed myself. As mentioned above, the Win to
Loss ratio for a full hedge play on the 5:1 position is around 4:1. The tests that I am
currently running examines series of 65,500 random spins. With some of these tests, the
W/L ratio has fallen below 4:1, but never below 3:1. Also, tests have shown that around
99% of all loosing streaks are have a length of only 1, 2, or 3 consecutive losses. So this
system assumes that the likelihood of two consecutive wins is significant.
Mathematically, the chance is 65.7414%. Essentially, this is a Martingale progression
that reduces the doubling.
This is really hard to chart out like the above systems so:
Let’s assume the player does not flee the winner, and places a bet on the first 5 double
streets covering the numbers 1-30, with no-value chips.
Bet per Square Total Bet Net Win Gross Loss
1. $1 5 1 5
If the player wins, the same bet is placed again. If the player looses the next bet is ½ the
gross loss per position.
2. 5/2=2.5 12.5 -5+2.5=-2.5 5+12.5=17.5
If the player wins, the bet is replayed. If the player wins a second time, go back to level 1
with a Net win of 0. **Note, an alternate progression can be played as followed
2. 5/2+0.5=3 15 -5+3=-2 5+15=20 NetWx2= +1
2. 5/2+1=3.5 17.5 -5+3.5=-1.5 5+17.5=22.5 NetWx2= +2
but let’s stick with the straight progression:
If the player looses the first time at level 2 then:
Bet per Square Total Bet Net Win Gross Loss
3. 17.5/2=8.75 43.75 -17.5+8.75=-8.75 17.5+43.75=61.25
However, if the player wins the first time, and looses the second time then:
Bet per Square Total Bet Net Win Gross Loss
3a. 17.5/2=8.75 43.75 -8.75+8.75=0 8.75+43.75=52.5
And a loss on the first win:
4. 52.5/2=26.25 131.25 -52.5+26.25=-26.25 52.5+131.25=183.75
And a loss on the second win:
4a. 52.5/2=26.25 131.25 -26.25+26.25=0 26.25+131.25=157.5
This system is adaptive to whatever your current situation is, and is best done at the
moment with paper or calculator.
The above is fine if you’re playing on a table with no-value chips that you can designate
at whatever value you want, however some tables and particularly online casinos don’t
allow this. You have to deal with whole increments, and this can be done easily enough
-you just round up to the next whole amount.


Bet per Square Total Bet Net Win Gross Loss
1. $1 5 1 5
2. 5/2=2.5+0.5=3 15 -5+3=-2 5+15=20
2a. 5/2=2.5+0.5=3 15 -2+3=1 2+15=17
from level 2:
3. 20/2=10 50 -20+10=-10 20+50=70
3a. 20/2=10 50 -10+10=0 10+50=60
from level 2a.
3. 17/2=8.5+0.5=9 45 -17+9=-8 17+45=62
3a. 17/2=8.5+0.5=9 45 -8+9=1 8+45=53


The biggest danger to this system is the same as any other progressive system. After a
while it just gets absurd to chase high losses for minimal returns, and you may hit the
table limit, however it would take a lot longer to reach the table limit than a standard
progressive system. Good money management should help you decide what your cutoff
limits are to the progression. I have used this system many times on a $1-500 table with
a $1000 bank, and I have never busted my bank or hit the table limit. Most times it never
makes it past level 3. But, of course, it can happen.


Finally, in closing, I have been playing around with combining the last two systems. This
would be to have an initial bet of $25 per position, and on a loss, follow the 5:1 Half
Progression, and on a win, follow the 5:1 Up & Pull