Cool moneyline, you should learn poker, you can get a much bigger edge and they won't kick you out.
When I said 50 losses out of 100 I didn't mean 50 straight losses. Just 13 more losses out of the 100 than expected (I don't think that's unreasonable if you win 63% of the time). Ofcourse like I said it depends on the variance of your trading results.
Another formula I read about today in "New Trading Systems and Methods" is to use the Student t-test (a statistics formula). It's a bit too complicated to explain right now but it might be worth looking up. It also compares your expectation per trade and the variance to give you an idea of how reliable your results are.
Money Management and the Kelly Formula
This is a list of MM links I posted on the MTEI Forum:
Money Management Links and Tidbits:
If you like math try:
If you like simple try the Kelly Formula at the bottom of the Seykota article, It is what I use.
There were a lot of trading systems at Wealth-Lab that used the Risk Of Ruin Code early on in the communities begininng.
"Risk Of Ruin" methodology.
Here is a sample from the Knowledge DataBase:
I would run my script on 10 different data samples (1000 bars - depends on the time frame) for each currency, and then average the payoff ratio for the Kelly formula.
If the average payoff ratio is less than 1.10, then I don't think you have an edge worth investing in.
If your EA is not profittable on at least 7 of the 10 data samples then I would go back to work on the EA until it is.
This is exactly what I need, thank you
Definitely a sticky in my book....excellent info to be had here.
Of important note: Bet Size (or Position Sizing)....
The optimum size is no more than 25% of your capital, free margin or equity (whichever is less).
I've always inituitively used this level - and as Ed Seykota's doc shows - it truly is the best way to maximize your profits in the long run.
Since I view Forex as a 50/50 game of chance, which can be improved upon vastly with quite a bit a time, effort, study and $$$, Seykota's charts, the Kelly formula, and the seminal Kelly paper from the Bell Labs have once again awoken the beast within.
Thanks for posting this info!
thanks for a great post.
work with me on this a little will ya?
let's say for sake of example that your system had a 100% expectancy, never missed a target profit. But let's also assume that it had to ride out two positions open at the same time thru it's worst case scenario which was to be each negative 20 pips at it's worst case limit. How many lots or what percentage of the account should each position be to maximize returns without going out in a blaze of glory? Assume a starting balance of $500.
Fine Tuning Your Money Management
just a free article from trader.com that could be used on MM on some EA's
Forward Test - Money management EA with tight stop loss
Hello all traders
There are so many EAs being developed in this forum. I am sure this thread would be one-stop-center for those who believe in EA to forward test. I am really looking for EA that has money management with tight stop loss. Kindly post any potential EA for testing
it is hard to find one that has low stop loss... the lower it is, the easier it is to lose a bigger number of trades...
Non-Hedge Grid Trading and Money Management
I've been doing more thinking about forex lately and how to profit from it. I've done a little reading as well and come up with the following forex ideas -
1. The market moves up or down. Plain and simple. There is no third option. You can't lose by the market going sideways. You can only lose if the market goes up when you're short, or it goes down when you're long. "But the market is going sideways when it's ranging," you would say. Well how big is the range? What if you were only looking for 10 pips when it ranged 40 pips each way. Could you not profit from the range? Or what if your TP/SL was large enough such that the market could range inside your TP/SL zone and not trigger either of them until it broke the range? The market goes up or down, simple as that.
2. Indicators, trendlines, and the like don't make for a better system. Sometimes they work, sometimes they don't. It is only in hindsight that we can curvefit indicators to see what we should have used. Can you tell me why price breaks through one resistance level, but not the next? What point is the real 2 point in a ABC123 method? When do we know when to follow MACD or when it is diverging? In the end these tools should only really make the decision to go long or short. Many times they can not, and I find them useless. BECAUSE.....
3. There are pips available anywhere on a chart. You don't need to wait for an MA cross or RSI crossing 0 to enter a trade. If you enter a trade randomly anywhere on a chart you will always have the possibility of winning or losing on that trade. Why do we need to wait for an indicator to tell us it's now time to trade? As I said, the indicator should only tell us to go long or short. Unfortunately most indicators lag terribly or have levels where they can not give us a buy/sell decision effectively.
4. Therefore, money management is the most effective tool we can use to produce a winning system. You can have a 95% successful system, but then lose it all on those 5% losers. You can have a 30% successful system and make huge profits.
Thus, I propose the following idea for a system. It's a grid type.
Enter a trade randomly anywhere, up to you. Place a TP and SL equidistant from your entry point. So say your spread is 3 pips. You would place a TP of 47 and SL of 50 since when you enter the trade you are already 3 pips closer to the SL. This way price needs to move 47 pips to hit the TP and 47 pips to hit the SL.
From this I will say at this point in time you have a 50/50 shot of winning the trade. In that aspect your system is 50% successful, but the spread would slowly eat away at your profits if you randomly went long or short. The next part to the system would be to base the buy/sell decision off of price action.
Say your original entry was a buy. The SL was hit as price moves down. Immediately you enter a sell trade from the exact point your long trade hit it's SL, thus placing you in the correct direction of the current movement. You now do the same for this trade of having a SL of 50 and TP of 47.
If the TP is hit for this sell order as price continues to drop then immediately enter another sell order with the same 47 TP and 50 SL. If that gets stopped out then immediately reverse the position to a buy with 50 SL and 47 TP. This makes the decision to go long or short easy in hopes that we have caught enough momentum that the current movement will continue to close out another winning trade.
On it's own I think the system could produce positive results. I can not backtest since I don't have EA programming knowledge, but I will say I believe that the system is more than 50% successful since we allow price to dictate the buy/sell decision. I expect that momentum carries itself through price levels.
Do not try this system off of a 47 TP and 50 SL. I was only giving those numbers as example. We would need to find a TP/SL zone that is small enough to take profits easily and frequently, but be large enough to not whipsaw many trades in a ranging market. In that case, as I mentioned earlier, the ranging market would fit inside the tp/sl zone and not trigger anything until a trend ensues. I think something close to a 20-30 TP/SL combination would prove effective. This would allow us to have a TP/SL zone of 40-60 pips and allow for plenty of intraday ranging, while still only requiring 20-30 pips in one direction to bank profits. So on the larger ranging periods it could still take profits.
STILL! The idea is that the system is successful based on the money management. What are strategies we can use to increase gains and cut our losers?
Some ideas of mine are:
1. If you have consecutive winners, add to your winners. Meaning if you would win a trade, don't close it, just open another trade on top of it to add to your position. Keep adding to your position every 20-30 pips. Then as soon as one of those positions gets stopped out, close all of the open positions. This way you let the first opened trade run for a long time. The problem is you risk not being able to lock in profits on the first trade.
2. Or if you want to close out trades to lock in profit you could still institute a system of increasing your lot sizes as you win more consecutive trades. The problem here is if you win 2 trades and increase lot size on the third, but lose, then you're effectively eliminating all the profits from the 2nd trade and perhaps erasing the first winning trade as well.
3. People will hate me for saying it, but you could try martingaling. Since it isn't so random of a system, it would depend on our TP/SL levels to determine how many consecutive losing trades in a row we can expect to encounter. The larger our TP/SL zone is the more unlikely it is we'll be caught in a ranging market to get whipsawed. I'm not so sure I like the idea though since you can only put so little towards a trade in case you do get a long losing streak. This means profits are coming in very very slow.
4. What if for consecutive winners you let the first trade ride, but the positions added to it afterwards close out regularly? Then when the first of the additional positions gets stopped out, close the original trade as well. Again, this risk is losing the profits and only breaking even on the first trade.
5. For consecutive trades look for increasing or decreasing TP's. What if we assume that as we enter more trades after consecutive winners the trend will slow. So perhaps after 2 consecutive winners of hitting 47 TP we only look for 37 on the 3rd and 4th trades, then 27 on the 5th, 17 on the 6th, etc.. Or we could reverse that and look for 17 on the original, 27 on the second, 37 on the 3rd, etc.. The problem of increasing the TP levels is to do so we might have to increase our SL as well to keep our system at 50% success. So as we increase our TP/SL and get stopped out on the bigger trades then it erases profits from previous winning trades. I'm not sure how leaving the SL constant would do.
6. We could combine the increasing decreasing methods from the 5th strategy. So we could look for 17 pips on the first trade, 27 on the second, 37 on the 3rd, and 47 on the fourth. Then on the 5th trade we would decrease back to 37, on the 6th trade back to 27, 7th back to 17. This would follow the idea that in the beginning a trend will be picking up steam and then losing steam at the end. Again, we risk losing pips from our winning trades as we also increase the SL when we increase our TP levels.
I think a system as simple as this could be very consistently successful if we just know how to manage our money. How do we capitalize on the 50% winning trades? I leave it up to the rest of you to offer suggestions, advice, tips, an EA, backtest results, etc.. I am a member of the KISS method club. I don't want to look at 20 indicators, memorize chart patterns and have to wonder if it will hold up. What if the pattern is showing to sell on a 4 hour chart, but to buy on a 30 minute chart?
I don't care for all that. I am accepting price action to dictate the decision. All I'm trying to do with these ideas is to get a ride on it's back and take pips when I can.
I offered the possible exit strategies and money management ideas just to get the ball rolling. I also mentioned them to show that there are probably hundreds of possible ways to trade a system like this just off of the exit strategies itself. Please offer up any strategies I have no mentioned. This is an open discussion as I need your help to build this system. Thanks. Matt
Ive thought of another method to play it.
What if, for example, we entered one trade with a TP/SL of 30/30. Then say our trade moves towards our TP 20 pips. We then add to our current position by adding another lot in the same direction, but with the same TP/SL of 30/30. This way we stagger our profits. When the first trade takes profit the second trade will be 10 pips towards it's goal of 30 TP. If it goes another 10 pips we enter another trade in the same direction just like how we added the 2nd trade on top of the 1st.
What's nice about this idea is we never add to the losing trades. We aren't hedging, so don't add to a trade that might get stopped out. What could be nice about this though, is that when you enter the second trade, it will get stopped out before the original trade gets stopped out if the market were to turn on us. So we could have a system where once the second trade gets stopped out we also close the first even though it didn't drop to that level yet. This way our losses would be -30 and -10 for the two trades as we enter a new trade in the opposite direction.
We only enter one position and again wait for it to move 20 pips in our favor to add the 2nd position. This way if it again turns on us we are only losing the -30 on that one trade rather than -30 and -10. It's also nice because it offers areas where we allow the market to turn on us without erasing all of our profits. What if we have two trades triggered, the first profits, and the second loses. We would be even out of the two trades while allowing ourselves to recognize the turn in momentum early and capitalize on the new reverse trend. The problem with this idea is we may have been better off just by entering a new trade when one trade hits its TP or SL rather than staggering the trades since we close out the original trade with the second instead of allowing it to possibly close in success. We will assume it wouldn't though since price just dropped 30 pips against us and momentum will carry it further.
The benefit to staggering, is again, being able to recognize the new trend early on and being able to ride it for longer than waiting for one trade to close out on us. Matt