The following article originally appeared on TT's Trade Talk blog on August 11, 2015. Click to view more posts on Trade Talk. Develop DIY Algorithms—Safely by Andrew Renalds, Product Manager Yesterday, The Wall Street Journal published an article titled "Algorithmic Trading: The Play-at-Home Version" highlighting the growth of a new crop of DIY tools that allow retail traders to easily automate their trading strategies. The users quoted in the article expressed excitement about having the ability to quickly build and deploy their own strategies, but they lamented that unforeseen issues in their algorithms led to sizable losses. Since 1994, TT has been building tools to allow professional derivatives traders to automate their strategies. It's encouraging to see the DIY algo programming trend start to migrate to retail traders, but the potential for loss with some of these systems is a detriment. To that end, allow me to point out a few differences between our approach and the others. ADL® (Algo Design Lab) Our ADL visual programming platform represented a major breakthrough in algorithmic trading when it was first brought to market in 2009. Using drag-and-drop actions to assemble building blocks, traders and programmers alike can rapidly design, test and deploy automated trading strategies without writing a single line of code. With ADL, users can generate executable strategies in hours to seize and act on fleeting market opportunities in timeframes that were difficult or even impossible to achieve previously. With ADL, users drag and drop blocks containing pre-tested code onto a canvas to create automated trading programs. Instead of providing pre-made strategies, ADL allows the user to create custom algorithms by stitching together building blocks that convert to well-tested code, allowing for extreme flexibility. This approach eliminates syntax errors, while visual feedback and warnings provide built-in safety by guarding against undesirable algorithm behavior. Moreover, algorithms can be tested with live market data using TT’s robust simulated matching engine for a high degree of realism. Algorithms are deployed to proximity-based execution servers to achieve ultra-low-latency performance no matter where the trader is located. Users can easily launch algorithms, update parameters and monitor each algo’s status from their workspaces. Mitigating risk and breaking down barriers We also provide risk management tools to prevent the scenario described in the WSJ article: "If things went wrong, he could lose his savings." ADL is a visual platform written in a domain-specific language, which means that it can catch many common mistakes before the user turns the strategy on for the first time. This is possible because domain-specific languages are developed with a precise context in mind. As such, ADL can detect issues with blocks connected in problematic arrangements. ADL warns against potentially problematic logic. Such intelligent assistance could have minimized, if not mitigated, the programming glitches discovered by these users during live trading. So, too, could the use of the ADL "Risk block" that allows the algorithm designer to create self-imposed limits to guard against the consequences of undesirable algorithm behavior. And then there is the variety of limits that can be set in the risk and administration system by administrators. ADL, which is an integral part of both our X_TRADER® and TT platforms, has been embraced by professional traders because of the flexibility and speed with which they can build, test and deploy custom trading strategies. The pace of technology advancement is breaking down barriers, allowing tools like ADL to be used by traders of all levels. As such, we will continue to focus on making ADL both an empowering and safe experience. Stay tuned as we continue to iterate on ADL and roll out new enhancements. Feel free to reach out to us to learn more or to schedule a product demo. Or you can try the new TT platform yourself by going to trade.tt—it only takes a few seconds to create a demo account.
Way too limited design structure. Newbies, stay away and rather go with a full fledged programming platform. If it has to be off the shelf then I recommend Rightedge or Amibroker which is by the way broker independent. You can use existing plug-ins to connect with your existing broker. This offering is not worth the time because very soon you will notice that you reach limitations on important order or fill issues.
Yeah it's limited, but I think there's a place for this for discrectionary traders, who can mock up some simple rules to handle trades without having to be glued to the screen....
You're right and volpunter is right as well. There is no way in hell that ADL can compete with complex algos due to its inherently limited design. However, for the punters taking a swing, this product has some utility.
I think that you're missing the point of ADL. ADL isn't designed to allow you to do everything a low level programming language does, but it does allow you to rapidly develop automated trading strategies by stitching together building blocks that convert to well-tested code. This allows you to focus on building your trading strategy instead of dealing with the details of price subscriptions and order routing logic. And it accomplishes this by providing building blocks that represent a specific context / task in trading. This is useful for those who don't write code as well as seasoned developers who are looking to quickly prototype trading ideas. TT also offers a number APIs to which developers can code directly when more control and flexibility is required. This allows users to choose a tool to match both their technical skill level as well as their need for fine grained control. Andrew Renalds, Product Manager of Algorithmic Tools and APIs
Then you miss your target audience completely. Beginners and those who are not well versed in programming will find much less pricy tools to get their feet wet and to come up with systematic trading building blocks. Even those with low level coding skills will miss some essential aspects your product suite cannot and does not provide. (We can discuss in detail which they are If desired). Experienced programmers will not gain anything from your product as they can buy other off the shelf products for proof of concept projects. Then you lock users of this product into your incredibly overpriced data and trading platform. After all anyone who has worked in this industry for more than a year will be able to attest that a tools of all trades sort of product just does not cut it. You try to serve a broad audience but do not excel in any single area. I can only repeat that anyone should feel strongly discouraged to waste their time on the concept to trading your product embarks on.
Volpunter, you seem exceptionally well informed on this topic. You've already suggested some alternative software packages. What do you think about using a free platform such as Rithmic API instead of TT once a new user has learned C#? If you could further elaborate on the problems with using ADL, I'm sure the community would benefit from it.
My point was that there is no ideal solution that covers all problems. I am happy to share my thoughts if you can describe the specific problem you try to solve. I am a strong proponent of using the most ideal tool for each problem. I can also elaborate on the core deficiencies of a trading and strategy design scripting language such as ADL but it makes more sense to do so in a specific context. Bring up a specific example and I am happy to chime in.
I'll start with something simple. I'm not a developer so this may be a little sloppy. Suppose I want to bid and offer on an exchange traded spread in an attempt to avoid paying the spread and perhaps capture a tick. Based on past experience I know that I will only get filled if the market runs past me by a tick or two. So I fire up ADL and set up a simple algo to: Bid/offer two ticks under/above the market. If filled on bid, offer two ticks above market. If filled on ask, bid two ticks below market. Some issues I can think of: Waiting for confirmation of fill before sending of order to exit position. Order to fill ratio. Each CME product has a different order to fill ratio and ADL has to somehow account for the variation.