I was thinking about the random nature of roulette and the "random walk" theory of investing. Even if there is random price movement in the markets, we can still apply technical trading tools to exploit the random movements. I began toying around with some of the same tools I use to trade Forex to see if they apply to roulette.
My methodology was as follows (RNG - NZ BV):
1. I created an Excel spreadsheet where I would enter spins as they came up, left to right, on row 1.
2. I entered a formula to determine if each spin was low or high in row 2: =if(A1<19, "L", "H"). I copied this formula several hundred cells to the right down row 2.
3. Beginning at J3, I started counting how many high spins we had in the previous 10 spins using formula =countif(A1:J1,"H"). This formula I copied as far as I had copied the formula from step 2.
4. Beginning at S4, I took the average number of high spins from the previous 4 results, using formula: =average(P4:S4). I copied this formula as far down row 4 as I had the other formulas.
5. Beginning at S5, I took the average number of high spins from the previous 10 results, using formula: =average(J5:S5). I copied this formula as far down row 5 as I had the other formulas.
Here is where the investing tools came in. I like to use the cross-over of moving averages to give me market direction. The 4-result average is the same as the fast moving average and the 10-result average is the same as the slow moving average I use to time investments.
Any time my fast line (4-result average) was higher than my slow line (10-result average), I anticipated an upward move in the averages, and bet on High (19-36). When the fast line was lower than the slow line, I anticipated a downward move in the averages, and bet on Low (1-18). When the numbers were the same, I considered that a choppy market and stayed out altogether. My spreadsheet is set to zero decimal places. I don't see any real benefit to playing farther out than whole numbers.
I used a Grand Martingale progression, as I had not seen many 4-in-a-row losses. I played 1,3,7,15, using the indicator I just explained in the previous paragraph. While I did experience some occasional bankroll losses of 26 units, I easily surpassed this in gains in every session that I have played (admittedly by hand, as I haven't taken the time to learn to code yet). Obviously, this could be adjusted to a regular Marti, or any other progression that isn't too incredibly aggressive.
I was quickly impressed at how this gained hold of genuine streaks quickly and kept me out of a lot of junk. True chops won't hurt you here, as you would win every other bet for a total of +2 using the grand Marti. What would hurt is a strong up trend, immediately followed by a strong down trend, immediately followed again by a strong uptrend (or vice versa). Gradual trends are the friend of this method.
Also, this system seems to absolutely love the Labby progression [1-1-1-1-1-1-1-1-1-1] (Thanks GLC - I owe you one for introducing me to that!!!)
Would love some feedback with people familiar with this type of investment analysis especially. Do I need to change my time periods (4- or 10- numbers too fast or too slow?).
Currently playing with BR of 26 (1,3,7,15). Stop playing after full loss, restart tracking for next session. You will have to track the first 19 spins without a bet. Also, there are periods, though not long, where you are watching and not actively betting. Should be under 10 spins at a time without betting, although that may have exceptions. Since Betvoyager doesn't allow for free spins anymore, I play 2 units on Doz 1, 2 units on Doz 2, 1 unit on 25-30, 1 unit on 31-36. That is a true free spin that will never cost or gain you anything while you are tracking between bets. I have this arrangement saved as a personal bet for quick play.
Hi Colbster,
Nice topic, I'm surprised more people don't look into this, because there are a lot of sophisticated tools available in TA for those who like to play trends. Some might say it's nonsense to use moving averages etc on roulette because the game is random, but no more so than price movements in the markets.
Personally I like point 'n' figure charts, they give you a nice visual overview of both 'global' and 'local' trends.
There's some free software - Forex Strategy Builder (link:://forexsb.com/forex-software/) which might be fun to play around with. Although the data in the text files are price movements, you could easily replace it with some roulette related data. As long as the format is correct in the file, the program won't know the difference, then you can use all the built in tools to analyse roulette. :)
Bayes,
Point and figure charts could be brilliant for this, just substituting one side of an EC for an uptick and the other for the downtick. Wow - not my area of expertise (I am a candlesticks guy myself, which work great with MA and other TA indicators), but I may have to give it some consideration.
When I first started playing around with this idea, I put a very basic charting mechanism into my spreadsheet that showed the ups and downs of the rolling counts. I was struck immediately by the similarity to stock charts. Not only do we have the usual undulations of the price/count, we also experience very clear resistance/support levels that respond to all the usual TA methods. Additionally, we have the benefit of a fixed channel to work within (0 to 10 Highs (or reds or evens) possible in 10 spins. We don't have to worry about setting a stop limit or having prices gap beyond our expectations. If anything, this is less random than the stock market. Frankly, I cannot imagine anything more random than human nature, driven by fear and/or greed, seasoned with mob hysteria dynamics and directed by commentators that have their own agendas. This seems safe compared to equities, and most see that as semi-legitimate.
Hello Colbster,
This looks really good. Thank you for taking the time to explain it. I think you should get a lot of help here soon.
What are your stats at the moment? You said you lost your BR several times but this was easily regained. How many spins/sessions have you tested etc.?
I still have to fully grasp everything but it seems to me that there will not be too many greater fluctuations than what you have seen. On the stockmarket there will be far greater fluctuations. Roulette is in this sense "stable".
As with the very-slow progressive betting, the straight-up random-walk stuff alone will never beat even a vig of 0.1%.
I've done a lot of that sort of math. Not interested in posting the hints here as the mention of anything serious quickly finds its way to the trash-bin, on this message-board.
Come on, were the real markets random, on a "random walk"... no one would be playing those, which would have collapsed more than a few times since the big one.
Hello GARNabby,
Please share with us. Your insights would help greatly. I am sure they will not enter the trash bin.
There are many great threads and posts. Look at Gizmotron for example. There should be a lot more interest in his ways but it seems that people have a preferred approach to the game and if it does not meet there vision will leave it alone.
Only by contributing will we all get somewhere......
@AMK, moving to the Labouchere was perfect. I'm playing with a line of 10 1s and a BR of 40. Highest bet yet was 8, win every time since changing progressions. I am up to 20 sessions with the new progression, averaging 30 spins after the 20 tracking spins. I haven't even sweated yet with this method. This seems super-solid
" There are manygreatthreadsand posts. LookatGizmotronfor example. There should be a lot more interest in his ways but it seems that people have a preferredapproach to the game and if it does not meet there vision will leave it alone."
Thanks. I happen to agree with the common sense offered by John Patrick.
He's an advocate of trends but you won't find much information from him explaining it. That's what I'm trying to do now.
Until just recently, I did not pay enough attention to Giz. In the past few weeks, I have been noticing some strong underlying thoughts, especially concerning trends. Giz asked how we know when a trend goes into effect and when one ends in one of his other posts. I think that this is a perfect means of quantifying trend beginnings and endings. A lot of my thought process for this idea came from Giz. This gives us a concrete (objective vs. subjective) entry and exit, rather than trying to "get a feel" for the game.
@GARNabby
I am undecided on the randomness of the markets. Some claim that roulette isn't even all that random, although please don't count my in that group. Regardless, there are great tools that can be applied to investments that can be very profitable, just by looking at the charts. I initially started this concept for myself a few weeks ago by trying to use traditional charting methods (specifically trend lines and support/resistance lines) to lock roulette into some sense of order. My moving averages method is the end result, and it looks as solid as any other method I have ever created or tested. In the near future, I will be moving this from the notepad area to full systems with a copy of my personal spreadsheet so I don't have to explain the formulas to people who don't know Excel too well. Right now, I am extremely excited by how accurately this grabs trends and avoids a bunch of the junk chops in the middle.
Colbster - " This gives us a concrete (objective vs. subjective) entry and exit, rather than trying to "get a feel" for the game."
Two very powerful things that you are doing. You use charts to see the trends easily and you deliberately avoid the chop. That makes you a proactive and pragmatic expert. Randomness requires these techniques in order to take advantage of it.
Colbster, great post using moving averages to determine possible trends.
Both Gizmotron in his latest post on trends and XXVV on his moving averages are pointing the way to possible success at this game.
I created an Excel Spreadsheet per your specs and added my testing data as well, many thanks to Reddwarf, Madmax and Superman for their assistance in creating and testing in Excel.
First I tested your system using just a flat bet on the first spin after the moving averages cross. I like to see what a flat bet does first. My results was a -64 loss based on 50 sessions each of 100 spins.
I am not in favor of a negative progression so I tested to see how long were successive losses.
I tested this for 70 sessions of 100 spins each:
Under 3 in a row losses: 25
4 in a row losses 20
5 in a row losses 9
6 in a row losses 12
7 in a row losses 0
8 in a row losses 1
9 in a row losses 3
Nothing much to get excited about, but this is the first try.
What appears to be more favorable, would be to try using the 10 spin average only. When the 10 spin average goes from 4 to 3 bet on LOW and when it goes from 6 to 7 bet on HIGH. Or any other combination that look favorable.
I will test out this tweak and see what happens.
Again, thanks for the post. It is most enlightening.
Make sure that you are stopping your betting when the trend ends. If you had a 7 on the 4-spin and a 6 on the 10-spin, indicating High, you only bet until the 4-spin drops down to 6. If that happens, you stop betting immediately and wait for the next trend to emerge. I wasn't sure if that was how you are playing.
I did just have a loss of a 40-unit bankroll. Still up around 200 total using the Labouchere on BV NZ. Despite the inevitable loss, I love this method!
And just as quickly, I have recovered the entire session BR that I lost. I am playing the 10x1 Labouchere to completion, 4 in a row to comprise a session. I just won 4 straight, with 3 of them exhibiting extremely strong trends in both directions. Will continue to test this and keep you all in the loop. Looking great so far!
Thanks for the update on stopping the betting when the trend ends, I will re-do my successive loss figures.
In the meantime, I did a quick test of betting Low when the 10 spin ave went from 4 to 3 and betting High when it went from 6 to 7.
27 Sessions of 100 spins each made a flat bet profit of 15. However, only 2.67 bets per 100 spins.
Nick
I am not surprised at the lack of bets. The trend would need to be severe to get up to the extremities. That is why I like the gradual direction. 6 out of 10 vs 5 out of 10 can make us a LOT of money if we identify it.
:)
As long as 10% goes to a charity........
Previously reported: Stop Bet when trend Ends
Under 3 in a row losses: 25 38
4 in a row losses 20 16
5 in a row losses 9 6
6 in a row losses 12 0
7 in a row losses 0 1
8 in a row losses 1 1
9 in a row losses 3 0
Here is the revised figures based on stop betting when the trend ends. Much less severe.
Thanks again for posting this moving average method.
I think there is a lot more here than people realize.
Quote from: GARNabby on Jan 16, 09:05 AM 2012
Come on, were the real markets random, on a "random walk"... no one would be playing those, which would have collapsed more than a few times since the big one.
Well, the "Efficient Market Hypothesis" says that markets are essentially random, although not everyone agrees with it. It's certainly the case that if you take small time frames, outcomes ARE random, but over longer periods there may be underlying trends which reflect the economic and political climate.
However, crashes actually point to the market being NON-random (a bit like getting 100 reds in a row in roulette).
And, random or not, there are certainly regular patterns that we can use to exploit whatever underlying trends and influences there are. By using moving averages, we can witness the exact same trends of peaks and valleys as a market. I have attached a very brief example of the visual dynamic I am speaking of below, based on the opening spins of Bayes' BV NZ spins file. I have plotted the average number of "High" spins over the past 10 spins. Notice the rhythmic ups and downs, and especially the support/resistance lines. Anyone who has ever looked at stock charts will find these very familiar, and therefore, able to be useful with any of the charting tools we might use in other endeavors.
Quote from: Bayes on Jan 17, 01:43 PM 2012
It's certainly the case that if you take small time frames, outcomes ARE random...
Ever hear of "micro-trading"? No name dropping, or hypotheses involved. Only the certainty that if you have the fastest hardware and software, then you are able to make your trades before the existing trends can reverse.
Quote from: Bayes on Jan 17, 01:43 PM 2012
However, crashes actually point to the market being NON-random (a bit like getting 100 reds in a row in roulette).
And more crashes would point to it being a "random walk", as i noted. And only the largest players can effectively ride out the crashes. No randomness there either. Unlike the casino-games, though those aren't really random either, where there are NO such "market forces".
If there were no corruption in the markets and nobody was allowed to use software which was faster then 99% of the competition then the markets would be more "random". I say this because if this was the case the markets would fluctuate differently.
Definitely benefiting everyone
Quote from: amk on Jan 16, 09:15 AM 2012
Please share with us. Your insights would help greatly. I am sure they will not enter the trash bin.
Flattery will sometimes get you everywhere.
Straight-up "random walking" is about the fact that if you wait long enough playing an even game, then one side will be up a given number of wins. That will yield a very-small percentage, generally less than the HE, when it does happen for the player betting that side. But eventually, it will swing the other way to the point that all such gains will be lost, and you'll be starting over again but less the full-HE amount. (For any given number of wins, one could review all of this backwards from such a starting point to the same number of overall losses which were "balanced out".)
Regardless of the HE being slightly- positive, or negative by the advantage strategies, those sorts of casino-games revolve primarily about the remainder, the 50-50. After you've figured out how to beat an even game, by how far to extend your BR, by which slight progressions to nest when, egs, then you can start to get fancy with the straight-up RW stuff.
Of course, other types of 50-50 approaches must be considered. In baccarat, the game on which i'm focused, the shoes almost-always end up in a solid mixture of P's as well as B's. If you bet only B, then you will will be right close to half the time; if you bet both ways, on average, then you've likely created the small chance that you will have a lot right, or a lot wrong. There are some other things you can do here, but the objective is to avoid the back-and-forth outcomes, where you're only paying the commission, to crowd your strategies to beat the game head-on against a lot of wins, or a lot of losses... where those strategies can do the most good. (In baccarat, the L/W's will exactly even out over-and-over-again over the long run if an even combination of P's and B's have been played.)
Quote from: GARNabby on Jan 17, 06:15 PM 2012
Ever hear of "micro-trading"? No name dropping, or hypotheses involved. Only the certainty that if you have the fastest hardware and software, then you are able to make your trades before the existing trends can reverse.And more crashes would point to it being a "random walk", as i noted. And only the largest players can effectively ride out the crashes. No randomness there either. Unlike the casino-games, though those aren't really random either, where there are NO such "market forces".
Even if your software is fast, it still doesn't give you a crystal ball. Price movements are essentially probabilistic, just like roulette outcomes. True, there's a psychological element which is missing in casino games, which means that movements in the market can drive activity in a certain direction (past prices can affect future prices), but the results are nothing more than the sum total of thousands of trader's opinions, which tend to cancel out (if all traders had the same opinion, there would be much less randomness).
Why would more crashes point to randomness? surely the opposite is the case. Roulette outcomes conform to the normal distribution, but events in the world and opinions aren't necessarily constrained in the same way - the "wheel" of opinion can sometimes become heavily biased!
QuoteUnlike the casino-games, though those aren't really random either, where there are NO such "market forces".
So if casino games aren't random either, what IS?
These debates seem to depend on the definition of what "random" actually is, but it's a very tricky concept. How do you know if something is "truly" random? apparently random outcomes sometimes appear to be non-random, and paradoxically, if you don't sometimes get non-random behaviour in a random game, then you don't have "true" random! :o
The fact is that short-term trends do exist, and can be exploited, whether you're talking about casino games or trading.
Quote from: Colbster on Jan 15, 04:39 PM 2012
3. Beginning at J3, I started counting how many high spins we had in the previous 10 spins using formula =countif(A1:J1,"H"). This formula I copied as far as I had copied the formula from step 2. it should be =countif(A2:J2,"H"), right?!
4. Beginning at S4, I took the average number of high spins from the previous 4 results, using formula: =average(P4:S4). I copied this formula as far down row 4 as I had the other formulas. it should be =average(P3:S3), right?!
5. Beginning at S5, I took the average number of high spins from the previous 10 results, using formula: =average(J5:S5). I copied this formula as far down row 5 as I had the other formulas. it should be =average(J3:S3), right?!
Or am I wrong?
@Marivo No, I was trying to visualize on a computer that didn't have Excel installed. You are absolutely correct. When I get a chance, I will update the original post and delete your post to reduce confusion for people who are just getting started reading this thread.
Despite my error, this method has only lost once with the session BR of 40 using the 10x1 Labby. My wife is sick and needy, so I am having to sneak mini-sessions in right now, but it is performing great! You should definitely run this yourself and give me a note if you have any questions.
Thanks for the clarification!
Colby
@colbster hi mate sorry being pain but is it possible to get excel tracking tool as i am not able to do one myself just my old brain has big resistance area to such a programs... :thumbsup: thank you
Quote from: Bayes on Jan 17, 01:43 PM 2012
Well, the "Efficient Market Hypothesis" says that markets are essentially random, although not everyone agrees with it. It's certainly the case that if you take small time frames, outcomes ARE random, but over longer periods there may be underlying trends which reflect the economic and political climate.
However, crashes actually point to the market being NON-random (a bit like getting 100 reds in a row in roulette).
If you investigate and go deeper in this issue, you will discover the markets are not random, mainly because they are manipulated by big players. And this is not conspiracy theory, a BIG player can and does influence things on many levels.
@Colbester Just one question mate, why 4 and 10 ? What logic you used to come up to this values?
The trends in roulette don't last for too terribly long, so I didn't want the high (slow) number to be too high that it would lose it's effectiveness. 10 seems like as good of an arbitrary number as any other. I originally was thinking of 3 for my fast number, but felt that it would be to susceptible to whipsaws and doubles/triples to be very good. There is probably someone who is able to optimize these numbers and tell me what the "best" numbers are, but I do not have the first clue how we would do that.
Also, one of the Forex trading methods I have played around with lately used 4 and 10 sessions for the SMA cross-over. That is what actually got me thinking in this direction. Don't know that they are the best, but they are certainly very effective from what I have seen so far.
Hoping to start giving details of winnings/sessions now that I am solidifying my play guidelines.
Session 1: 88 spins, 48 non-tracking spins, +20 units, largest drawdown 5 units
+0.227 units per total spin, +0.417 units per non-tracking spin
Total time: 12 minutes
BV NZ - Session bankroll 40 units
Session #2: 65 spins, 25 non-tracking spins, +20 units, largest drawdown 2 units
+0.308 units per total spin, +0.800 units per non-tracking spin
Total time: 7 minutes
BV NZ - Session bankroll 40 units
Session #3: 117 spins, 77 non-tracking spins, +20 units, largest drawdown 22 units
+0.171 units per total spin, +0.260 units per non-tracking spin
Total time: 17 minutes
BV NZ - Session bankroll 40 units
Obviously, this was a little harder session that still ended in our favor. As I was recovering from the drawdown, I was amazed at how this system called the end of the trend and kept me out of jumping into the next trend before it was time. The accuracy was great after the rough start to the session.
Spreadsheet with playing rules now in Full Systems here: link:://rouletteforum.cc/index.php?topic=8531 (link:://rouletteforum.cc/index.php?topic=8531.msg74882#new)
"First I tested your system using just a flat bet on the first spin after the moving averages cross. I like to see what a flat bet does first. My results was a -64 loss based on 50 sessions each of 100 spins.
I am not in favor of a negative progression so I tested to see how long were successive losses.
I tested this for 70 sessions of 100 spins each:"
Nickmsi, any chance you could rerun this data with a flat-bet variation? Instead of just betting once after the crossover, how about flat-betting for as long as the trend lasts? So basically every viable bet. I think this could give us crucial information.
Birdhands . . here's what I did . . .
Any time the fast line (4-result average) was higher than the slow line (10-result average), I flat bet 1 unit bet on High (19-36). I kept betting 1 unit until the averages were the same, or lower then no bet.
When the fast line was lower than the slow line, I flat bet 1 unit on Low (1-18). I kept betting 1 unit until the averages were the same, or higher then no bet.
I did two test of 100 sessions of 100 spins each session(10,000 Spins Total)
Test #1: 11 Units Profits. 46 Losing Sessions and 54 Winning Sessions
Test #2: -25 Units Loss: 50 Losing Sessions and 50 Winning Sessions
Hope this helps . . . Nick
Birdhands sent me a PM yesterday concerning this idea I am working with, and it made me think about one aspect of this that I don't have a firm grasp on. Playing how I laid this method out, if we have a 3 in the fast number and a 2 in the slow number, I have suggested playing high. However, while we may be seeing more highs over the last four spins than over the past 10, that still averages only 3 out of 10. This may indicate that our low streak is coming to an end, but 3 stills sees us losing 7 of 10. I have been having success with this, but I wonder how much of my success has been the progression instead of my indicator.
Birdhands mentioned the MACD oscillator. It is well beyond the scope of this forum to discuss something that technical in the investing world, but it is a strength of signal indicator. A lot of traders use it in addition to whatever tool they happen to be using to generate their buy and sell signals. I am going to look at adding a MACD to my averages method to see if it can give us a stronger signal than mere edges that might be working against the market (ie. buy signals when the market is just giving slightly less aggressive sell indications).
Other suggestions along these lines are always welcome.
I have incorporated the MACD(4 Spin Average minus 10 Spin Average) indicator in my excel spreadsheet.
Please clarify:
When the MACD indicator shows a +1 or +2 what would you bet?
When the MACD indicator shows a -1 or -2 what would you bet?
Thanks
Nick
What I played around with earlier today was taking the averages out to 2 decimal places and calculating the MACD with the decimals to give us more nuance in the movement, a la 0.45 to 0.85 to 1.10 instead of 0-1-1. Frankly, I don't know what to do with it right now. I have thought about taking a square of the MACD and multiplying it by some amount to determine betting stakes. I have thought about only betting high when the 10-spin-count is over 5 or betting low only when when count is under 5. There is something to be had here, but I'm at a loss where to go. I'm sorry I don't have specific rules for +2 or -1 yet. I'm consumed with the idea, so hopefully I can give better direction soon.
I wonder if it would work to plug spin data into a program that would give us a graph. Successful traders use dual momentum reversal strategies to decide when to consider a trade, but then use Elliot wave analysis and other objective strategies to determine whether or not to actually enter the trade. Even just looking for a basic ABC wave pattern in our spin graph would possibly make the difference. Maybe there's an equation that could do this.
Another consideration is that most successful traders lose the majority of their trades; but they lose a lot less than they win on their winning trades. This points to the fact that exit strategies are pivotal in assuring profit. I think most of this is done with fibonacci resistance levels. Again, we would probably need software to make this possible. Still, why mess with Forex when we can play BV NZ for pennies if we want.
I still think there's a lot of possibility here.
Sam
I never thought that I would even be discussing the thought of Elliott Wave Analysis applied to roulette. You have no idea how turned on that makes me to even type in the same sentence! :thumbsup: I'm thinking Forex for retirement and roulette for income. Either way, I think that a lot of the tools available to me as a trader can be applied here as well in the interest of taming the beast. I don't think that my averaging idea is the only place, either. Early in this thread, point and figure charts were mentioned. I am starting to do some research into this idea, and think that we might be able to use this for B/R tracking, or other EC bets. Looking at the chart that I generated from my averages, I still see resistance/support lines that might be useful at some time. I can see this branching off into an entirely separate school of roulette study that some people might naturally be drawn too, and hopefully bring their experience to, as well.
Money management will be inherently key here. Bird, you know my position that the Eggleston Betting Method gives us a mathematical edge over roulette. Identifying trends through charting doesn't really give us an edge from a math standpoint. We will always have the same 50/50 spins, just with the hope that our timing can identify the tops and bottoms of moves. Like I mentioned at the end of my Moving Averages thread over on the main board, I think that it may have been the progression that helped me early in my testing of this first method. Testing lots of ideas and seeing what sticks will be the only way to find our winner, which I strongly believe we will eventually find. I'm not at all bored or let down by our failures to this point, so I am looking forward to the days, weeks, months and years of honing our craft.
I'm already testing all 3 EC at the same time with their own individual betting, hoping to smooth out the results.
The point of all this is to identify a trend. Whether we use the MACD as the indicator or the moving averages or something else, we need to keep trying.
Currently I am testing the simplest of all trend indicators. KISS, Keep It Simple Stupid, is still applicable.
If a 15 Spin average goes from 6 - 5 - 4 then it is in a downward trend. It if goes from 9 - 10 - 11 then it is in an upward trend. All 3 EC at same time.
Will keep you posted.
Nick
It might be worth trying an exponential moving average, which gives more weight to recent data points, or maybe a combination of SMA and EMA -
link:://:.investopedia.com/video/play/SMA-versus-EMA#axzz1kBEgfH1s (link:://:.investopedia.com/video/play/SMA-versus-EMA#axzz1kBEgfH1s)
Regarding TA applied to roulette, it's certainly worth investigating, but just a cautionary note: TA may well work in trading precisely to the extent that traders use it; patterns, movements and signals tend to repeat in the expected way because they are identified as being significant by large numbers of traders, who then act on them (a kind of self-fulfilling prophecy). Obviously this doesn't apply in roulette, because outcomes are totally independent of the decisions that players make.
However, It's not my intention to pour cold water on the idea, and if you're going to use and play around with systems, why not make use of the sophisticated techniques which traders use to determine trends, rather than reinventing the wheel? A lot of roulette systems are just crude approximations of TA charts and signals. As a way of presenting and summarizing data in a visual way, TA has a lot to recommend it.
Quote from: Colbster on Jan 21, 08:46 PM 2012
the Eggleston Betting Method gives us a mathematical edge over roulette.
Colbster, so why aren't you using it? Is it posted on this forum?, I did a search but it didn't turn up anything, except this post.
Hello Bayes,
It is over on the VLS roulette forum.
Type in 'The eggleston betting system'
cheers.
Thanks John. :thumbsup:
It can also be found on this forum here: link:://rouletteforum.cc/index.php?topic=3608.0 (link:://rouletteforum.cc/index.php?topic=3608.0)
Bayes, as a math guy, I am interested to hear your input. Pay more attention to the mathematical element at the beginning of the post and less on the bet selection that it moved to later. I introduced a different betting method later, Eggleston Betting Method 2.0, also available on this forum. The problem with both is that, like the Parrando Paradox that is getting so much attention currently, finding a way to apply this bet correctly seems to be undoing it.
I never got into roulette to make money. I dream numbers at night, and doing roulette spins in my mind is helpful in falling asleep. I did make some decent money playing this method until an errant click blew my entire bankroll. With no money to play, and confident that I had a solid solution, I got bored by playing my method. I still play around with it sometimes, but I don't get the thrill I do from trying new angles that I think up or read about here. This is more a mental exercise for me, although I don't rule out making a little money if I ever find something more "holy grail" caliber.
I think that there is a method, or methods, out there that can beat roulette. I don't buy the conventional wisdom that it can't be beat. The only time the HA is absolute is when our bet is static. If you put 1 unit on 3rd dozen and hit auto spin for 1000 spins, you deserve to lose your money.
Regarding the earlier post about how charting may not be applicable to roulette the same way it is to investing because people use charting in investing and it becomes a self-fulfilling principal, I would state that charting was invented after investing and had merit before investors were using it. I understand your thought, but not all investors use the same charting, or even use the same charts in the same way to the same conclusion. Different time frames and different conclusions stop charts from really driving the markets in a specific manner.
That doesn't mean that they are applicable to roulette, but I don't think it can be ruled out just yet.
I just did 100 spins to test how the charting would actually be, and I was struck by the quality of the trends. Look at the attached and tell me that we can't use investment strategies here!
The only problem being: Is the moving average giving an advantage in trading ? The tests i have seen isn`t giving any.
Anyway, if you wanna extend the idea, how about making 2 charts with different timeframes ? Like you would do it in real trading. To save a long explanation read here about multiple timeframes:
link:://storage.saxosoft.net/sg/pdf/fx/2-4-multiple-time-frames.pdf (link:://storage.saxosoft.net/sg/pdf/fx/2-4-multiple-time-frames.pdf)
Here is an example of Bollinger Bands applied to the same data as the previous chart
Personally i think Bollinger bands and the MA is lagging too much. How about inserting an Oscillator like the Slow Stochastich (or RSI) and a Parabolich Sar ? Difffycult ?
When the stochastich moves above 80, you exit your position when the Sar is trigged. Otherwise the Sar is ignored. When the Sar is trigged, you bet opposit until you are either stopped out or the stochastich goes below 20 and the Sar is trigging you to reverse your betting.
Kelly,
You are my hero!! What a great bit of fun this thread is turning into for me. I agree that the indicators are lagging behind the movements. I have never figured this stuff out on my own - all the trading software just takes prices and generates these charts for me. I will see if I can find a formula for a stochastic that I can incorporate into Excel. I love the idea of overbought and oversold as betting indicators. I am also a big fan of the CCI (Channel Commodity Indicator for people not familiar with it) as a leading indicator. I will hunt around and see what I can find. Thanks for the ideas!
Here is the slow stochastic. It was actually pretty easy to generate.
When it broke below 20, I went H until a down-tick in the Stochastic. When it was over 80, I went L until an up-tick. With this bet selection, I won 29 and lost 23, a win rate of 55.77% and a z-score of just ever so slightly over 1.
Personally i always use the parabolich sar combined with the slow stochastich because a stock can stay overbought/sold for a pretty long time where the parabolich sar executes the reversing only when the trend is actually changing. The stochastich gives you an alert signal that a trend might be reaching an end, and the sar gives you the actual "go ahead".
Yeah, CCI i use as one of the components for daytrading. But also, multiple timeframes and the sar is helping tools.
Here is a trade setup for Vestas
Stoch. indicate overbought/short and the parabolich sar gave a short signal at the same time.
(link:://img834.imageshack.us/img834/9789/vestas28.png)
This exact trade was shorted at 112 and closed in 91.
Yikes . . . you guys making my head spin with all these MACD, Stochastich and Bollinger Bands.
I need a “Band†Aid for my brain.
Your Chart of Roulette spins shows a fluctuating graph with high and low points which is exactly what we want. From a high point to a low point is a downward trend and from a low point to a high point is an upward trend.
When I used to trade stocks, I would find a solid company with an upward trend and purchase it with a 10% sell order. If the stock went up 10%, it was sold and I took a profit. The same can be said for a downward trend and sell short.
Gizmotron hit it on the head when he said you have a 50/50 chance of getting on a trend at anytime. However, we have some indicators to show a trend may be starting or stopping and we can increase our chances of jumping on or off at a good time.
When you see your Chart of Roulette trend going down, bet it for 1 or 2 spins and then stop and wait for another trend.
I have tested this method for 2 test of 10,000 spins each(100 sessions of 100 spins each). I used all 3 EC simultaneously. I used a 15 spin average.
If the EC showed a downward trend by going from 6-5-4 or lower or if it showed an upward trend by going from 10-11-12 or higher I would bet.
Both sessions showed a flat bet profit of 10 and 50 units respectively. Attached is a chart showing the 50 unit profit.
When we break down the 50 unit profit session we get:
1st Spin Win-----344
2nd Spin Win-----164
3rd Spin Win-----73
4th Spin Win-----30
5th Spin Win-----12
6th Spin Win-----7
7th & higher-----8
As you can see, you can easily play a 2 or 3 step martingale progressions to double or triple your wins.
The downside is that you only get 638 opportunities to bet per 10,000 spins. You would need a tracker and patience and/or a bot. This will probably pass the Van Keelen test if I carried out the placed bets to 1000.
And the fun continues . . .good grief Reddwarf, you've created a testing monster in me (just joking)
Hand testing it, so it is slow going, but I am having remarkable results using MACD and the slow stochastic together. Any time the stochastic is overbought/oversold, or within 1 spin of it leaving that condition, and the MACD is at 0 or favorable in the direction to correct the stochastic extreme, I am betting flat until the stochastic reaches/crosses 50%. There are numerous betting opportunities in a session so far, but this is a fair amount of tracking before the conditions concur. Several stretches of 4 straight wins and no real bad streaks.
Have also a few investment analysts taken up roulette?
aka, the Streak Fighter
Quote from: Colbster on Jan 22, 08:35 AM 2012
Regarding the earlier post about how charting may not be applicable to roulette the same way it is to investing because people use charting in investing and it becomes a self-fulfilling principal, I would state that charting was invented after investing and had merit before investors were using it. I understand your thought, but not all investors use the same charting, or even use the same charts in the same way to the same conclusion. Different time frames and different conclusions stop charts from really driving the markets in a specific manner.
That doesn't mean that they are applicable to roulette, but I don't think it can be ruled out just yet.
Good point about charting having merit before investors were using it (the point and figure method is a good example - it's the oldest form of TA and was in use when there where very few traders). And in any case, there are different explanations as to why TA seems to work (although there's no conclusive evidence for that either - it's an area of controversy). Some say it's to do with the psychology of traders, other's say it's just a way of recording supply and demand.
I'm glad others can see some potential here, it's a vast field and deserves a forum to itself. :)
I'll take a look at the Eggleston method, thanks for the link.
No worries about infrequent betting opportunities; we can always increase our unit size if we have an edge. This ii really great; you guys should all give yourselves a big pat on the back.
No big pats due yet. Most recent test had 17 betting opportunities in 116 spins. 12 losses, 5 wins. Pretty clear signals using the standards I was mentioning earlier, they just did not hit as we would like. Still plugging away.
Here is another idea to throw out, as this is the notebook section and it isn't solid enough to put in full systems. I think it is easier to quantify for a certainty when something is in a trend than to say that it isn't it a trend. Attached is another spreadsheet (sorry, I know it is a lot, but maybe something will trigger an idea or an improvement that might break this open). It is another tracker that tells you when not to bet (any time that any of the cells below our spins are red). When they are all clear, I suggest we bet opposite-of-last. It does catch some nice streaks of choppiness.
Here are the rules I have put formulas for into the spreadsheet:
1. I have Excel determine if last spin was high or low. The next row determines if this was the same or different from the previous spin. It keeps track of how many in a row of the same we have spun. Obviously, the higher this number, the more "in a trend we are" and the worse it would be for us to bet OOL (opposite-of-last)
2. We do not spin if the average streak over the past 6 spins is greater than 2. This means that we have hit 3 or more twice in the past 6 spins or have had a big streak recently.
3. Our average is not growing compared to the last spin. Again, this means that we are not moving into a new trend.
4. Our standard deviation is not growing. When we have a shrinking standard deviation, this means that the spins are consolidating. With the rule that excludes averages over 2, this means that we are consolidating with chops, or no more than 2 in a row.
When all of these indicators show clear, we bet opposite-of-last. I am trying to bet against random changing it's mind. I am aiming to bet against a trend when we are not in a trend, or for a trend when we have been in a trend. I like the though of random continuing it's own trends and not being able to regularly fight against itself.
Ideas are always welcome!
I am continuing to play around with applying technical analysis methods, despite the mixed bag results. I agree with those who have suggested that the moving averages methods tend to lag the movements, so I am toying around with oscilators, which are generally considered to be leading indicators. I have done some testing today with RSI (Relative Strength Indicator) and found it to be lacking for our purposes. Not only do we have infrequent overbought/oversold indications (>80 or <20), they don't seem particularly good at calling direction. When we get a hellacious streak of one side of an EC or another, there is no investor sentiment that causes it to come down as a correction. We are still left largely at the mercy of the market, aka random.
Next up: CCI (Commodity Channel Index), which is one of my personal favorite tools. Anything that you guys would like me to test, if you let me know, especially if you can point me in the direction of the formula that is used to calculate it, I will be glad to give it my attention.
Thanks for the continued interest and ideas along this discussion line.
Just tested CCI on a few hundred spins. It did much better than RSI, giving more trigger readings and performing better. It does, however, take more tracking spins before the readings are of any value. I computed the CCI using a 15-period mean deviation instead of the more typical 20-period reading that is commonly used. Also, for typical price, instead of using (High+Low+Close)/3 for the typical price, I used the SMA for the past 3 spins. This small test did give me a positive result flat betting, although it did have a losing streak of 5 straight losses. That said, it also had a winning streak of 9 straight wins. This seems like it could be of some use, but I think we are getting out on the periphery of what could really be useful to our endeavors. I would like to crawl back to the mainstream a little bit, which I think is where the moving average crosses are. Thinking maybe throwing an EMA/SMA cross into the mix to see where we end up. I will report back.
I got really promising results using a 10-spin SMA/EMA cross. As before, I am attaching a spreadsheet for you to test this with your own numbers. Just finished a 250-spin session that won 11 units flat-betting. Very solid candidate for a progression, maybe even a martingale of some sort and duration. The longest string of losses was 5. Flat-betting, the max drawdown was 7 units and the lowest actual bankroll was +/-0.
Let me know how it works for you.
Another 193 spins. +7 at the end flatbetting. Biggest drawdown was 9 units, lowest session bankroll was -2 units. There was a single series of 5 consecutive losses. Of the 136 bets, we would have been +79 with a grand martingale. I like the short series of losses that I am having so far here. Will continue to test.
Oh, and for those who have downloaded the spreadsheet, enter the spins across the top row. The bet is the H or L in the yellow highlighted row directly below your latest spin.
Short flatbetting session - 39 spins: 27 wins, 12 losses = +15 units. This was a total of 66 spins, including the tracking spins at the beginning. I was also tracking the d'Alembert progression to see how it fared against this bet selection. It resulted in a plus 27, as we would expect.
256 spins, +12, flat. Very stable.
Just spun 290 and ended with a +10 betting flat. However, this was a more volatile session than most of the others. I reached a high of +7 at 58 spins, before experiencing a huge drawdown of 14 to the session low of -7 at 99 spins, which recovered back to around zero very quickly and orderly. Gradually, the system came back, ending my session at +10 on a streak of 4 consecutive wins.
Still like it, just want everyone interested in this to know that it does have some deviation. Still, it hasn't lost and has always came back plus some.
Just had the system breaker series. I played close to 500 spins. The high of the session was +7, around 100 spins in. It dropped down to a low of -18, a draw down of 25. It then languished between -9 and -18 for an eternity and I gave up.
I tested this system for 7,500 spins of 25 sets of 300 spins each. Results:
Totals at 1st 100 Spins: -1 Totals at 200 Spins: -10 Totals at 300 Spins: -4
Pretty much break even. This was all flat betting.
I was betting every time the SMA and EMA crossed and that was a lot of betting. I was thinking maybe if we reduce the number of bets by narrowing the bet selection, it might help.
Instead of betting every time the SMA and EMA crossed, I bet when they crossed only if their differential was .25. I tried other differentials, like .01, .50 etc. but most were not any better.
The .25 was area appears the best so far.
Results with .25 differential for 25 sets of 300 each.
Totals at 1st 100 spins: 38 Totals at 200 spins: 112 Totals at 300 spins: 151
It is just a suggestion but perhaps we should look at not only the moving averages crossing each other but that a secondary criteria might enhance the bet selection and improve the results.
Good hunting and keep the ideas flowing . . . Nick
I concur that adding a second trigger, where both conditions have to be met before we bet, would be an improvement. I am glad you threw the size of the differential into the mix. That is not something that I had seen.
Can you look at your data and tell how frequent certain draw downs were? If we get flat betting results of even, that means we have 1 win for every 1 loss. That is optimum for the d'Alembert progression, where each chop ultimately earns us +1 unit. If the draw downs are not typically large (say, less than 10 on average), we could easily weather the downs that would result with a modest bankroll (10 steps in the d'Alembert only costs 55 units).
I think my greatest interest would be if we could quantify whether we should be betting only when the EMA is on our side (ie H when it is above or below 5) or trending on our side (ie H when EMA is growing or shrinking compared to last spin).
While I was running some of the differentials I did a couple quick EMA and SMA only tests and did not see any positive results, however, I did not use any secondary triggers, ie. differentials just on one side. Will take another quick look and see if that helps. Will see if above 5 or whatever makes any difference.
Sorry, I don't keep my testing runs, only the results so cannot tell maximum drawdowns. Will keep an eye out in future runs.
Ran a couple more quick 25 sets of 300 spins each using only the SMA or the EMA numbers. Flat betting only.
For SMA numbers 6 or higher:
1st 100 spins: 2 Totals at 2nd 100 spins: -5 Totals at 300 spins: 8
For EMA numbers of 6 or higher:
1st 100 spins: 18 Totals at 2nd 100 spins: 40 Totals at 300 spins: 47
Looks like the EMA area worth pursuing further . . . Nick
Quote from: kelly on Jan 22, 10:10 AM 2012
Personally i think Bollinger bands and the MA is lagging too much. How about inserting an Oscillator like the Slow Stochastich (or RSI) and a Parabolich Sar ? Difffycult ?
When the stochastich moves above 80, you exit your position when the Sar is trigged. Otherwise the Sar is ignored. When the Sar is trigged, you bet opposit until you are either stopped out or the stochastich goes below 20 and the Sar is trigging you to reverse your betting.
Hi,
although this is an old post, I have some thoughts about this topic.
I did parameterized different scenarios where I'd bet following the dominant color, then basing bets on moving averages 4, 6, 10, 20,30,40,50,100, using different triggers, then using two different avg crossing to decide if red or black.
Strangely enough, the best results were when tracking only the absolute difference between red and black. Then avg 4, 6 and 100 were second best. I get and average and median of -1 win-loss out of 200 spins simulations. There always is this one game where all goes well and end up 30 more wins than losses. Other times -30: these games are when the trends is not so steep.
I thought averages were quite slow at triggering when going higher than avg 10. I then realized I liked referring to Parabolic SAR when trading a stock. Why not use it along with EC tracking? Depending on parameters, it'll NOT trigger a betting change on small trend variations, but trigger properly when the trend is flattening and a threshold gets crossed.
Just googled PSAR and there is a tutorial on how to create the function on Excel ( looking fwd to work on it this week).
So, I'll try to insert PSAR and will get back with results ( I am still a newbie with Excel sheet programming).
Other indicators such as stockastics, RSI might have their place.
Thoughts?
Ok,
PSAR is not enough on itself to enhance the hit rate. The net difference between red and black is still more accurate. PSAR works fine when the trend is steep but fails if the trend remains neutral. see hesitation and switching between spins 79 and 93
I'll see if combined with another tech indicator is better. Was suggested in the post above to go PSAR and stochastic.
Will see...
In example, PSAR triggered when crossing the moving average 4. Win-lose total of +5 on 200 spins.
bump :)
bump :)
A short-term concept taken from forex.
We treat 3 EC pairs as our currency pairs.
We make 20 spins, then we have a leader, one EC which is the strongest and we go with the trend.
We choose one EC and play Target profit +5 units, stoploss -5 units
R: R is 1: 1
Rather, we are doomed to failure by zero.
But this is the basic concept of the game with the trend.
I am guessing that some people use advanced technical analysis concepts, but rather no one will share if it brings profit
Quote from: ozon on Mar 23, 05:35 AM 2018
A short-term concept taken from forex.
We treat 3 EC pairs as our currency pairs.
We make 20 spins, then we have a leader, one EC which is the strongest and we go with the trend.
We choose one EC and play Target profit +5 units, stoploss -5 units
R: R is 1: 1
Rather, we are doomed to failure by zero.
But this is the basic concept of the game with the trend.
I am guessing that some people use advanced technical analysis concepts, but rather no one will share if it brings profit
You have to use excel charts on cellphone - by far the most detail and accurate method.
But no accomplished fx trader will reveal their secret - no visual ever seen online.