|
unluckyjinxcrsd
Registered: Jan 2003
Posts: 5
|
Set Monthly Income - Full Analysis
Hi,
I don't usually post here, but I've got a friend who posts here, and he has been curious about certain strategies, so I decided to take a look and see what the deal is. I couldn't help but notice all of the hype about this "Set Monthly Income" strategy, so I downloaded a copy and decided to run a (Matlab) Monte Carlo simulation on it.
For those of you who don't know what this means, I'm basically simulating the strategy over a 30-game period - about 100,000 times, and the resulting effects on my simulated bankroll will be a distribution that will give you an extremely accurate representation of what to expect in real life, and with what probability. It's basically an accurate enough shortcut to doing complicated probability calculations.
My simulation was done on the example used in the write-up.
Parameters:
Starting bankroll = 1600
Starting series = Bet to win 40 - 20 - 40 with a -110 line.
Stop at 30 bets OR
Stop playing at +400 OR
Stop playing at bankruptcy (-1600)
I assume a 50% probability of winning each bet. I follow all rules explained in Chapter 1 of the write-up, including spreading losses over the next series, stopping after 4 successful series, etc. Don't feel this doesn't apply to you if you are continuing after winning 4 series, because your continuation is basically the start of the next "month."
Here are my results:
At the end of the 30-game period (or less), the bankroll was:
+400: 80179/100000 times, which is 80.139%
-1600: 19820/100000 times, which is 19.82%
In other words, within a 30-game period, if you follow the same rules, you will either hit all 4 series or go broke trying. There is virtually no middle ground. This is baffling, but understandable with some thought, and remember that my simulation quits before the 30 games is up, when either hitting +400 or going broke. It's just extremely rare that neither happens before the 30 games are up.
I'd personally recommend staying away from a system in which you have a 20% chance of going bankrupt, especially if the reward is only 25% of your bank. It sounds like it all evens out, but you must consider how much you are risking and how serious that risk is. The author seems to dance around the fact that we have limited funds, in which case it doesn't matter how much "discipline" you have, you won't have the ability to "stick to the system".
On the other hand, if you have a consistent method to give yourself a 55% chance of winning, my simulated chance of bankruptcy is 11.3% and the chance of hitting the ceiling (+400) is 88.7%. But hold on to that thought..
If you are using your entire bankroll as the starting bankroll, this is still a stupid idea. You will go completely bankrupt about every ten months (every 5 if you are the average gambler). If you are using maybe 10 percent of your total bankroll as the starting bankroll each month, this isn't too bad of an idea anymore. The problem is that you are now making 25% of 10% of your "real" total sum, which is a 2.5% interest rate overall. And too bad it's 2.5% on some months and -10% on others. So you really shouldn't even leave your newly grown sum out there on the line every month, because in any given month it can be taken all back. Ah... the dangers of chasing bets.
Of course, claiming a 55% chance of winning is a lot bigger of a claim than some of you imagine it to be. If you are one of those people who believes they can consistently win 60% of their bets for 30 bets a month, EVERY month, you need to take a deep breath, and admit to yourself that Santa isn't real.
Just to be completely clear on this, if you are consistent with the rules, and "dividing your bankroll by 4" as shown in the paper, it doesn't matter what your starting bankroll is, the percentages of occurrence are the same (by proportion) as shown here for a 1600 unit bankroll.
So my advice would be to, first of all, find a way to prove to yourself that you can CONSISTENTLY win 55% of your -110 bets. Secondly, only use 10% of your "real" bankroll. Thirdly, don't leave all of that original monthly starting bankroll on the line after it's grown for a month.. re-calculate the new 10% of the real bankroll after a winning month (but not losing month). Fourthly, don't use this system, it sucks. Fifthly, if you don't want to listen, I guess I don't mind since it's your loss anyway.
-Dave
|
02-13-04 05:33 AM |
|
|
| |
|
unluckyjinxcrsd
Registered: Jan 2003
Posts: 5
|
simulator
Unfortunately, I created this simulator as a Matlab program. It runs in Matlab, but not on its own. Matlab is used for engineering programmers and financial programmers, and requires some programming knowledge (C or Fortran) to use. I use it because it is very simple to run loops to do calculations over large amounts of real or simulated data, and I can write programs to organize data for more complicated theories and strategies. Monte Carlo simulation is a term from probability theory, not a software package or anything.
However, if you are familiar with Excel, and if you have Excel on your computer (usually comes with Microsoft Office), you have the opportunity to do your own simulations. It might take some initial work and careful thought, but will speed things up ridiculously once you've gotten over the hump, and it will give you much more accurate results than doing a few instances by hand.
For example, Excel has a random number generator. You can fill 10,000 or 100,000 cells (whatever very large number you want) in one column with uniform random numbers between 0 and 1 very easily (using "=RAND()" and then copying and pasting the formula down the column), and then use the "IF" function to do the comparisons you need in the next column in order to simulate betting results. This is useful to test money management strategies, but obviously not for testing trends and theories. The biggest problem is when strategies get complicated (like the Set Monthly Income strategy), your spreadsheet gets somewhat large and difficult to organize, which is why I use Matlab, but I wouldn't discourage using Excel.
Alternately, if you have found some way to acquire actual game data history with the lines, you can replace the random number generator part with this real data, sort it by date or by team name (or whatever), and enter the simple math formulas you need to calculate what the result was using a certain theory or following a certain trend, copy and paste the formulas down each column and if you've acquired enough data (big "if"), you've got a valid long term test. It will probably take a few separate calculations (a few columns), and you'll have to change things for each theory you want to test, but it inconceivably speeds everything up once you've got it working.
Then you can easily make charts, histograms, etc. with the results you've found. Excel has a list of all the available functions and extensive help documentation under the help menu that you can search through to figure out how to do what you want to do. Sorry if it sounds boring, it's actually not boring at all once you get a little bit of confidence with it.
-Dave
|
02-13-04 09:57 PM |
|
|
| |
|
DevilDog
FoxDen Hall of Famer
Registered: Mar 2003
Posts: 796
|
That is the dumbest post
I have ever seen. Serious.
Are you trying to tell me that you can't hit 8 out of 30 winners!? You have got to be kidding.
Absolutely absurd.
Apparently myself and others are just "special". While I don't fully understand your simulation, apparently you are not in the real world of handicapping.
The Set Monthly Income System is FANTASTIC and when you can hit .266 and still hit your set monthly goal I call that GREAT.
Run all the simulations you want. It is nothing but B.S., in the highest degree!!
Some of you spend waaayyy to much time with numbers and not enough in the real world of money management.
I simply cannot believe how damn dumb that post was.
By the way, you don't neccessarily need to be setting your monthly goal as a 1/4 of your bankroll.
The Set Monthly System is a money management system that works wonders for guys that actually want to win money and NOT have to hit even 50%!!
Dog
|
02-14-04 04:13 AM |
|
|
| |
|
unluckyjinxcrsd
Registered: Jan 2003
Posts: 5
|
system
DevilDog, I have no response for you, except that I admire your ability to believe almost everything you hear.
Dork99, upon further analysis, the system can be used responsibly, by only using a low percentage (10%) of your current real bankroll as the "starting bankroll" each month. Also, there is DEFINITELY a necessity to have a consistent winning average of, at the very least, 53%, to make money, and to lower the risk of bankruptcy when using this system. Keep in mind that it is always a bad idea to use your entire real bankroll as the starting bankroll for this system. Perhaps there are other creative ways to prevent the disaster of losing everything, but I haven't thought about them. I believe some of you when you say you've been using the system for a while and have been doing well. It works, when you don't lose everything trying. You might even go a year without losing. But there is no way to predict which month will be your last month.
|
02-14-04 09:55 PM |
|
|
| |
|
Ragtop
StatFox Hall of Famer
Registered: Jan 2003
Posts: 5517
|
Systems, Momey Mang. and Simulations?
You can do any or all this together, but unless you can find out why you wager went into the tank, what to hell good is it? My point. I lost tonight on my wager with the Canucks. I than went back to my sheets, stats and situational numbers to see where I may have made a mistake. I missed a couple of important stats on the home team and knew right away that I had made a bad bet tonight and stored this info way for another day. True you don't always find the answer, but it is usually there. My point, there is no substitute for good old fashioned handicapping, both before and after the fact. If you pay attention to your capping, you won't have to worry about running a Startrek like analsis or your money management, because it won't mean Jackshi$t, if you keep making the same mistakes and lose. Where is parley Perry when you need him? The only thing I like to chase after a big dumper is a nice shot of Canadian wiskey.
|
02-15-04 08:04 AM |
|
|
| |
|
unluckyjinxcrsd
Registered: Jan 2003
Posts: 5
|
handicapping
There are two kinds of mindsets - that of the gambler and that of the investor. We have all at least tried to be one or the other at some point. I have been a part of both worlds. I don't live in a cave.
Gambling is entertainment, primarily for the purpose of getting thrills out of testing our instincts. Investing is not entertainment. It is similar to a job. The purposes of investing include making a stable income off of a sum of money and lowering the risk of loss of that sum. If you are a gambler, so be it, but at least admit it to yourself as enjoyment, and not a dependable source of income. Whenever someone tries to make the transition from gambler to investor, they are either serious about it, and put a considerable amount of long-term mathematical research into what they are doing, or they are just greedy and they look for short cuts into a dream world. There is no substitute for significant mathematical research - both money managing and trend searching - if you want to make a stable, consistent return.
If this system requires you to put down more money than you feel comfortable putting down (which it probably does), then don't use it, plain and simple. There's no long-term reward for following it, so don't expect one. If you don't admit this to yourself, you might lose a lot of hard-earned money by believing a fraud.
I don't like to make generalizations, but when someone writes a book that claims it can turn you into a long-term "professional" gambler by just using one simple system, it is almost always written to make a quick buck, and doesn't hold the key to life-long riches. In this case, the system he proposes is a threat to your entire bankroll, but its drawbacks, similar to the drawbacks of Martingale betting, are disguised by partitioning schemes and possible short term gains, so many people will consider trying it. Don't believe the b.s. about "discipline" and "sticking to the system" though, because betting a grand on the Nets to cover against the Celtics doesn't require discipline. It requires a spare grand in your bank account.
-Dave
p.s.
caldrummer00: I'll send you what I've written after I put some comments in. Version 1 sets the bankroll back to zero and ends the month if it goes negative. Version 2 is more realistic (or just more sensible) and holds on to the remaining bankroll if the system's suggested betting amount is higher than the available bankroll, and then it ends the month. I'll send both.
|
02-15-04 10:27 PM |
|
|
| |
|
GeorgePDog
Registered: Dec 2003
Posts: 98
|
I was seriously thinking about trying out this system on my bankroll which is small.
With a small $400 bankroll, you are looking at a 10-5-10 series, not much money, right?
10-5-10 bet 22-20, loss, 378
17-12-17 bet 37.4-34, loss, ~340
22-17-22 bet 48.4-44, loss, ~291
26-21-26 bet 57.2-52, loss, ~234
29-24-29 bet 63.8-58, loss, ~170
31-26-31 bet 68.2-62, loss, ~102
33-28-33 bet 72.6-66, loss, ~30
Now, I don't have any money left to chase. It would take 7 consecutive losses to where I'm trying to figure out where my money went.
The thing that scares me and why I am trying out Q's system is that I can't reload at that point. I'm not made of money.
I have been horrible with money management in the past, but one of the gems that I have heard but not necessarily followed is that there is a small percentage of your bankroll you should risk daily. I've heard 5% from the conservative to 25% from the bigger risk-takers. That ensures that you live to see another day. You may still lose your money, but it will take longer and give you more exposure to those possible wins. It's not glorious, but I've read here and elsewhere that sports gambling (or investing if you will) is a marathon, not a sprint.
My problem is that I almost need someone to tell me how much to play because I get all giddy about some sure thing only to end up scrambling to cover my debts.
Keep the discussions going about money management. There are those of us like me who need to learn.
George
|
02-15-04 11:13 PM |
|
|
| |
|
nntrader
Registered: Jan 2004
Posts: 105
|
George,
Your numbers are off. Check your series and rule. Specifically, if you lose, you divide previous wager amount and add it to previous series.
E.g., if you bet $22 for 10-5-10 series (two outside numbers=20, bet 22 to-win 20), if you lose, you divide 22 by 3 (7) and add that giving you new series of: 17-12-17. Now you wager $37 to-win 34 (17x2=34).
You did this correctly for your 2nd series, but 3rd series is off. When you lost $37 in step #2, you should add 37/3=12 to the "17-12-17" *NOT* the original 10-5-10. So your #3 series should be 29-24-29.
In other words, your bet increases a lot quicker than in your example! Here is what the series looks like when you lose 7-in-a-row (to follow your example).
Starting bankroll=400
1. Series=10-5-10, bet 22, if lose bankroll=378
2. Series=17-12-17, bet 37, if lose bankroll=341
3. Series=29-24-29, bet 64, if lose bankroll=277
4. Series=50-45-50, bet 110, if lose bankroll=167
5. Series=87-82-87, bet 191, if lose bankroll=-25
6. Series=151-146-151, bet 332, if lose bankroll=-357
7. Series=262-257-262, bet 576, if lose bankroll=-933
Notice that you've blown your bankroll after the 5th loss.
I've posted on other Set Monthly Income threads that you're better off just doing a straight martingale and quit after you've lost 5-in-a-row. For example, a straight martingale can handle this series: LLLLWLLLLW. Whereas the Set Monthly Income cannot. I know that is extreme example, but all the alternating W/L results used with that system in the original book as examples were just as arbitrary.
BTW-- one way to get passed this simulation vs. real results debate is to just apply this progression or ANY progression on a series of results. For example, take the ATS results for every NBA team. Or O/U results for every NBA team. Just start with team #1 and assume you always bet on that team, or always bet on the UNDER. Now instead of randomly generating W/L results, just use every single team's ATS results and O/U results. Now you'll see for yourself how often 5-in-a-row or other streaky results will doom your progression system.
You need to test everything. I trust math, but if you do not, instead of taking it on blind faith, try the exercise I just mentioned. See how often you blow your bank account.
|
02-15-04 11:58 PM |
|
|
| |
|
| |
|