I'm far from an expert trader, but I've learned a few things along the way . . .
1] Set aside your trading system and begin with money management. First, set a realistic goal (20 pips a day for example).
2] Next, determine the number of contracts to trade that provide the optimum risk/reward.
3] Lastly, determine your exit plan.
Once you have completed these step you are in a position to develop a strategy to meet your goal. Here's what I like to use . . .
GOAL: 20 pips a day
EXIT: Take off 1 a +5, 1 at +10 and leave the 3rd as a runner.
I use different entry strategies based on market conditions. The strategy I select determines the stop I use at entry. Regardless of the strategy I keep the plan as outline above since it is far more important than any given strategy.
The reason I always enter with 3 contracts is that it reduces my risk dramatically. For example, I enter a trade with a 10 pip stop. If I get stopped out I am -30. I know my risk going in.
As the trade develops, I take off the 1st contract at +5. When the market moves to +8 in my favor I move my stop to break even +1. I am in a free trade at this point. Should it continue on in my favor the 2nd contract is out at +10, which only leave the runner. I exit the runner based upon my strategy.
In summary, my goal is to make 20 pips a day, but I only need to reach the +8 point in the trade to be in a free trade. Knowing this, I can work on developing strategies that give me good odds of reaching the +8 target. You'll find over the long run that the +5 contract pays the utilities, the +10 makes the house payment and the runner is Porsche money.
The following 14 users say Thank You to hondo69 for this post:
Technically, trading more contracts does not lower the inherent risk, however it allows you to control your risk exposure and thus lower it by using good money management. With one contract you are very limited in what you can do, by using multiple contracts you can scale in/scale out and dynamically manage position size.. Good MM is what will make or break your strategy, it is *vastly* more important than what indicators you use.
My MM technique is somewhat similar to yours. I break up the entire position into 3 different classes: cost covering, scalp, swing. Each class of orders has its own settings for initial stop limit, breakeven handling, trail stop handling, target, etc..
The amount of contracts assigned to each is dynamic. For instance with cost covering, there is a TicksToCoverCosts parameter, if I set it at 2 then based on the total position size, it will adjust the amount of contracts dedicated to cost covering so that commissions for the entire order are covered after 2 ticks. Any additional contracts after that are either scalp or swing orders based on current estimated risk (ie recent trend history, bid/ask spread if applicable)
The reason for differentiating between scalp and swing orders rather than just using 2 classes is because swing orders generally have significantly looser breakeven and trail stop handling than scalp orders. So during ranging or consolidating periods or if making a contrarian move I will use scalp orders primarily. If trending heavily or in a divergence situation at the end of a trend where there may be another small stop hunt but the chance of extended movement is low then it will primarily use swing orders.
This technique is heavily influenced by Joe Ross books, the main difference being that he only uses 2 classes, no 'scalp' orders. He puts a huge amount of importance on trading multiple contracts, in fact he goes so far as to recommend that you not trade until you have enough capital (and enough faith in your strategy) to trade multiple contracts. I would tend to agree with him.
The following 8 users say Thank You to sefstrat for this post:
I wasn't familiar with Joe Ross so I looked up his website and found it very interesting. Looks like there is some real value there and I'll investigate further. Thanks for the tip.
I'm also wondering about your settings for position size, cost covering, etc.? You must have a formula that calculates the proper balance based upon trade class. This has always interested me but I never read much about it. If you can tell me more without divulging any secrets I'd really like to learn more.
- - - - - - - - - - - - - - - - - - - - -
There's a guy I know that likes to bet 4-5 college football games a week. He explained his "system" to me one day and it more or less made sense. But he also mentioned there was typically one game that stood out as a slam dunk bet while the others were less so.
I asked if he bet a larger amount on the slam dunk game and he replied "no", since the Vegas odds makers were much smarter than he. Then he told me the slam dunk games averaged about 80% win rate and the others closer to 60%. Seemed to me the only slam dunk that existed was to increase the bet on the 80% games and lower it on the others.
The following user says Thank You to hondo69 for this post:
adding contracts just to cover order fees sounds risky to me. presumably you need 1 to 2 ticks to cover fees. also presumably if you are in a liquid market you would need to have a stop wider than 1 to 2 ticks on same contracts. so now you are risking more, possibly a lot more, to cover commission costs. if you are certain you can easily get 2 ticks then why not just load up a few dozen contracts and call it a day.
don't confuse: i am in complete agreement with money management being more or less the only important thing in trading but this particular technique raises question for me.
additional contracts does not lower risk. i cant wrap my head around that. but i recognize what you are saying. additional contracts empowers you to have much larger and more efficient gains. so we know what you are saying that the end result of trading more contracts is much more favorable because you are in a position to make much smarter decisions and capture a lot more of what the market is willing to give you.
"Let us be thankful for the fools. But for them the rest of us could not succeed." - Mark Twain
The following 4 users say Thank You to caprica for this post:
I'm in agreement with you Caprica and I should have worded my argument better. What I should have said is that 2 contracts can be less risky than 1, if managed properly. And 3 can be even better than 2.
This is based upon the goal of 20 pips a day and only making 2-3 trades a day. Trying to make 20 pips a day trading 1 contract is higher risk because it does not allow you to pocket small gains.
It's all about managing your stops and exit points. Some people I know trade 3 contracts and their exit points are +3, +6 and a runner. Makes perfect sense to me imho. The odds are better than the +5, +10 and a runner as I mentioned above. Just depends upon your trading strategy.
The following user says Thank You to hondo69 for this post:
My MM library has various strategies built into it, Kelly ratio, Fixed fraction, anti-martingale etc.. the one that I actually use is fixed fractional.
And yea, most trading books are rubbish IMO but Joe Ross is a notable exception. I highly recommend his stuff, even if you don't use his methods exactly understanding how he views the market, how he identifies trends, plans entry/exit, etc.. is very valuable. Traders trick entry is pure gold if you know how to use it right. (Having good money management similar to what I discussed above is the key to using it right, by the way)
If you read the beginning of my reply earlier I mentioned that adding contracts does not inherently lower the risk. What it does allow you to do is manage your level of risk exposure in a way that you cannot when trading single contracts.
You make a good point though, yes it would be very risky to add extra contracts if you aren't sure that your strategy will hit the cost covering target. That ties in with the bit I mentioned at the end about having faith in your strategy. If your strategy is opening positions that don't go at least 2-5 ticks in your favor almost every time, it is not a good strategy, period.
However, even with a good strategy there will still be many times you get 'faked out' and you find yourself with a few ticks of profit but things are turning on you suddenly. The reason cost covering orders are so important is that slippage can be a real bitch when you are trading many contracts, especially on fast moving markets like CL or spot forex.
It is not rare that I have 4-5 tick slippage on scalp orders when I get in a bad trade. If you are trading 12 contracts thats $500 or so right there. By having the cost covering orders you are covering the costs of all the orders right away and usually making a few ticks of profit also, that way if you have to close all the other orders in a hurry with your breakeven stop you stand a chance of actually breaking even or coming out slightly ahead rather than having your capital eaten away by slippage and commission from bad trades.
Why not just open 12 contracts and close them all at 2 ticks? That is fine if 2 ticks is all you are after, but why settle for 2 when you could have many more.. You will be glad that you collected those extra ticks on the rare times when you do get fully stopped out (which I still find painful even if I am way on top for the day) or the times when you are on the wrong side of a news release and get hit with $4000 in slippage (that one still stings)
Really it depends on your strategy though, the money management is the most important part of the equation no doubt, but your overall strategy has to be designed to work together.. ie the choice of instrument, bar type and sampling frequency, entries, etc.. all of those must consider the MM strategy and vice versa.
That is another thing Joe Ross and other experienced traders continually warn against, choosing your MM strategy based on fear or pain rather than designing a strategy based on your own insight into the characteristics of the market. Most traders fall victim to this because they are undercapitalized, ie they are using too tight of stops.. it is exactly this kind of psychological factor that operators exploit with their stop hunts and fake outs.
The following 7 users say Thank You to sefstrat for this post:
Well said Sefstrat. I hope many people read this topic as it is truly the key to trading.
My biggest problem is stop management in the middle of a trade. Although I've tried many combinations there really isn't one or two I can settle on. Since Ninja backtesting doesn't allow testing of stop strategies it's hard to prove one method over another statistically. Therefore, I rely on my gut and trading journal. I'll spend some time this week looking over the MM methods you've mentioned as they really interest me.
- - - - - - - - - - - - - - - - - - - - - - - -
I have a hunch I'm trying to prove based on the following assumptions: a high percentage strategy is in place, goal of 20 pips a day, trading 3 contracts and entry point is easily established. The "secret sauce" is therefore stop management and exit points.
Exit Points: the trader must decide where to take off the 1st and 2nd contacts. It seems like it should be possible to develop a spreadsheet that over time would produce a success ratio for this strategy. Something like:
Assume the numbers are based on a hard stop at 10 pips. Then the trader could change the stop to 11 or 12 pips and do the exercise all over again and compare the numbers. At some point the trader would arrive at the best combination of stop setting and exit points (I'm not concerned with the 3rd contract at this point). It could end up something like: exit at +4 and +9, but when price gets to +7 change the stop to break even +2.
I've tried to write some formulas in Excel but it never comes out quite right. Maybe some of the MM methods you mentioned above will nudge me along.
The following 4 users say Thank You to hondo69 for this post:
You can backtest any type of MM strategy in ninja if you use the custom IOrders interface for handling everything. I thought you could do it using the standard NT order management also but maybe not..
Custom order management is a bit of work to set up initially but very useful as it can be made much more flexible than the standard NT order management. There are samples on NT forum reference section of how to do the basic stop loss and trail stop.
The best MM parameters are highly dependent on the specifics of your strategy and often they will be quite different than you initially expected (in my experience, at least) so it is very beneficial to be able to use the optimizer on them.
The following user says Thank You to sefstrat for this post:
I just wanted to say thank you for giving birth to such threads. In my opinion this is the kind of stuff that people should learn first, before they go on looking for a system that's going to make them rich!
We can never get enough of MM. And people should spend more time on this kind of topic instead of finding the ultimate system. For what is the ultimate system without the proper MM!
The following 2 users say Thank You to George for this post: