Suppose you are long the Sept Euro contract at 1.3222, and it is getting to the time in mid September when you must rollover to the December contract. When you have to perform this rollover in an algorithmic strategy, how do you actually accomplish it? I usually do it one of three different ways. Each one has its advantages and disadvantages.






Method 1: Quick Roll (most expensive typically)

In this approach, you enter a sell order at the market in September Euro, and enter a buy order at the market in December Euro. Since both orders are market orders, they will be immediately executed. Using the prices above, here is the math:

Sell September at 1.3299 (bid price)

Buy December at 1.3304 (ask price)

Closed profit = $962.50 - $5.00 commission = $957.50
Open Profit = Long From 1.3304

Advantages: guaranteed fill, quick, easy

Disadvantages: you pay 2 bid/ask spreads, and if you are not quick with second order, market could run away from you, costing you money.


Method 2: Leg In Roll (cheapest method typically, IF done correctly)

In this approach, you enter a sell order at the market in September Euro, and try to work a buy order at the bid price in December Euro. The sell is immediately executed, and the buy is a limit order, which hopefully will get filled at the price you want. Using the prices above, here is the math:

Sell September at 1.3299 (bid price)

Buy December at 1.3303 (bid price)

Closed profit = $962.50 - $5.00 commission = $957.50
Open Profit = Long from 1.3303

Advantages: you save yourself $12.50 by getting in the December contract 1 tick better than you did with method 1.

Disadvantages: more complicated, plus you may have to chase the market up to get the December fill. You can easily lose more than 1 tick by trying to be too greedy with your limit order.


Method 3: Exchange Supported Spread Roll (cost usually between methods 1 and 2)

The exchanges have a great tool to help spreaders - a dedicated quote feed and tradable symbol for executing spread orders, which is what a rollover is. In this case, you are buying or selling the spread, not the individual legs. You get simultaneously filled on both legs at the same time. Using the prices above, here is the math:

Buy Spread at 4.5 (ask price). Note that this doesn't tell you the actual execution prices. You'll see these on your statement, and they are really irrelevant in the marked to market accounting of futures (the spread price is all that really matters). So, we will just assume some prices, keeping the spread fill price correct:

Sell Sept at 1.3329

Buy Dec at 1.3329+.00045 = 1.33035


Closed profit = $962.50 - $5.00 commission = $957.50
Open Profit = Long from 1.33035

Advantages: simple, no chance of only one leg executing, cost usually in between method 1 and 2.

Disadvantages: Some brokers don't support this. Tradestation's main platform, for example, does not allow this. You can do it manually in Tradestation 4.0 platform, but that platform is not made for algorithmic strategies. It is also easy to screw up the order, and buy the spread instead of selling it. You have to be careful. Also, don't think that just because you are trading the spread symbol, you'll only pay 1 commission. Some people think this is true, and I guess those folks don't read their statements very closely. When the brokerage does its accounting, it splits the spread into a separate buy and sell fill for each leg. When this happens, commission is charged for each. There is no free lunch, commission wise, with this method!


You can see from the above example there is $12.50 cost difference between all 3 methods, with Method #1 being the most expensive, Method #2 being the cheapest, and Method #3 in between the other two. This is not always the case, but is true in general.



So, that is how a rollover is actually accomplished. Whenever I can, I use the exchange supported spreads to do my rollovers. I can do this with systems where I enter orders manually. For my fully automated systems, I generally use method 1, even though it is the most expensive. When I use Method 2, I sometimes find myself chasing the market with my order, or worse yet, I forget about the rollover for a while, leaving me temporarily doubly exposed, until I fix it.

Thanks a lot for the in depth answer. and all the other answers you provide to my questions. I really appreciate it.

Hi @kevinkdog

I really appreciate you answering all my questions. But I have another maybe more personal than most of my others. I have been in a big rut lately, for the last few months. Trying to turn out a system into incubation each month but without success. My development process goes like this.

1) Randomly Select 20% of total data to use for Insample
2) beat the junk out of the in sample until i get something that works. Always 5 rules or less.
3)Walk forward bench test 50% or less of the remaining data. So i can test different WFO IS OOS settings or equity curve vs net profit etc.
4) WFO against the entire data set (minus 3 months)
5)String together the WFO windows and put it into Monte Carlo
6) Use the reserved 3 months for instant incubation
7) move to normal real time incubation
8) go live.

I have tried many instruments (mostly focusing on emini dow, euro and five year notes) but from daily all the way down to 5 minute. and nothing can pass the WFO or MC. I have a lot of pre coded stuff so i will try say 20-30 ideas right away. From simple break outs to more complicated time of day stuff. Whatever Idea I had before i try again on different time frames/instruments. Maybe that's making me go in circles.

Anyway what do you do when you get stuck? How do you deal with not having any systems pass your workflow? Is my workflow to rigid and hard to beat? I dont really have any idea how to progress and am feeling like I am running around in circles. Thanks again for all your help.

Here is a very long winded answer to your question. I wrote it for another purpose, but I think it applies here. I hope it addresses most of your concerns.

Back in the old days, before I developed the process I now use, I did what most other people do: I picked a market, selected a bar interval and time period, put a few rules in a strategy, and then optimized. Whatever turned out to be best is what I started to trade with live. The magic of the computer, with its ability to quickly do millions of iterations, uncovered what I was sure would be riches. Boy was I ever disappointed!

Those early failures led me to a more involved, robust and ideally trouble free development process. I can't take credit for any of the individual concepts - certainly many traders before me long ago developed most of the steps of my process - but I have tailored it to something that feels right to me, and by and large it has yielded good results in real time.

That is not to say it is an easy process for strategies to pass. When I first started using the basic process I use today (I have enhanced and refined it over the years, and if anything it is tougher today than it was a few years ago), it probably took me about 100-200 trading ideas for entries and exits before I found something worth trading. With time and experience, that number has dropped significantly, but coming up with tradable strategies is not an easy task.

Traders that are new to using walkforward analysis, Monte Carlo analysis, etc. frequently are frustrated by the difficulty of discovering a strategy. My answer to that dilemma usually is "but that is how it is supposed to be!" Think about it for a second - if it was easy to find a strategy, don't you think others would have already found it and exploited it? There are thousands of traders and researchers out there every day looking for edges, mining data and running tests. I guarantee you that all the easy strategies either no longer work or have been long ago discovered.

A friend of mine who is a Commodity Trading Advisor (CTA) once told me that if he finds one new strategy a year to trade, he is a happy camper. He should be, since even one strategy, with proper money management, can make you rich. But to get that one strategy per year takes a lot of work. I frequently think of the strategy development process as a factory. At the receiving door of the factory are your trading ideas and strategies, the raw material you constantly need. Your analysis tools, backtest software, walkforward algorithms are the machines in the factory. You, of course, are the skilled labor running the machines, monitoring the quality of the product. The output usually fills a big garbage bin right outside your factory, unfortunately. But, what isn't thrown away as garbage is pure gold - your tradable strategy.

The factory metaphor is appropriate because strategy development is tough work. Factory workers are among the toughest people I know of, and that is how you need to be to develop strategies. I am always amazed by educators out there who gloss over strategy development. Instead, they focus on such nonsense as getting in touch with your feelings, or writing everything down in a journal. Don't get me wrong, those items have their time and place in trading, but they are no substitute for having a strategy with an edge. If you don't have a good strategy, all the journaling in the world will not save you. As an aside, it is ironic, though, that many times "soft" skills such as psychology or journaling will be indispensable to a trader with an edge.

Since strategy development is a factory, you need to keep the factory running at all times. Here are some tips that I use or have used to keep things humming along:

· Any time you see a trading idea that intrigues you, write it down. Keep a list of ideas you want to test.

· Look for ideas anywhere. Trading books, magazines, and internet forums are all good sources of raw ideas. I would not recommend taking an idea as presented and trade it exactly as is, though. I'd look to modify it first, and put your own unique spin on it.

· No idea is too silly, too stupid or too dumb. The only dumb ideas are the ones you never test.

· If things are going bad, try the opposite. Buy when you think you should sell, and vice versa. Maybe something interesting will develop from your opposite idea.

· If you are a goal setter, try to test 1-5 strategies per week, at a minimum. It may take 6 months to a year of rigorous testing, but eventually you'll find something.

· If you run out of ideas, pull up random charts, and just stare at them. You can even add an indicator or two. After a while staring, but not thinking, walk away from those charts, and revisit them a few days or a week later. Then start to think: do you see anything in the relationship of the indicator to the chart, or in the chart itself? Write down what you see, program it, and test it.

· Find other traders roughly the same skill level as you. Offer to swap ideas or strategies. Take what these traders have, and then build strategies around their idea. I do this frequently with some of my fellow World Cup winners.

· Change your criteria. Maybe you are being too restrictive in what you consider acceptable. Open the door a bit to strategies that meet most, but not all, of your criteria. You can always tighten the criteria once your factory starts producing.

Awesome answer. The factory is exactly what I believe my "trading business" is. I am supposed to develop product "systems" and put them out the door. You really struck a cord there with me on that. Its the same metaphor i been using with myself. So basically as for me the factory worker, I just need to grind out more hours and more systems.

I have ideas and concepts that i understand coming in and always looking for more. I have the factory line, my process listed above. I also have the capital to put things live. I just need to continue to do that process until i strike something. I may continue to loosen my restrictions as well just to see if anything turns up. All good stuff.

As always thanks for your honest and complete answer. i always appreciate it.

Hi Kevin @kevinkdog,

I have a general question. When you are adding rules, filters, stops or targets your baseline system. Are you then optimizing those parameters over your in sample to test if its generally a good idea?

For example: Say my in sample is 25% of my total data set.
I found a basic rule(s) for my trading system that over an optimized parameter set, say 1-200 by 1. is profitable 75% of the time.

I want to add a new rule, be it a filter, a target, a stop. How do you go about doing that and testing it?

Would you add that new rule. then optimize again 1-200 over the in sample. Then see if it increased the % profitable of the entire test? Or would you just add it in and look at a particular setting? I really have no idea how to modify my basic trade idea in order to get to walk forward.


I went through your webinars and looked at your futures.io (formerly BMT) CL system. you have tested a fair amount of rules. Close Close[x] and RSI. with 3 types of exits. how did you add, test or accept those rules over others?

If I read your question right: I would probably test the rule over the in sample dataset, then optimize again 1-200. I normally consider the in sample period as a "sandbox" of sorts, where I can try some different things before testing over the whole dataset.

Is there anything your looking for in the re optimization? Like more parameter sets are profitable? or higher average net profit or something?