Hello, I am interested in hearing from anyone who has done self-implemented, fully automated options trading. In other words, you write a program that runs on your computer and uses a broker's API to acquire real-time market data and submit option orders, unattended. My question is not about the strategy aspect but about technical and regulatory feasibility. It seems this is difficult because of CBOE Rule 6.8A prohibiting generation of option orders without manual intervention. On the other hand, CBOE already proposed years ago to eliminate this rule (www.sec.gov/rules/sro/cboe/34-51030.pdf), but I can't find whether the SEC approved, so I assume the rule is still on the books. If CBOE is out, is anyone successfully using the Philadelphia Stock Exchange or even foreign markets for fully automated options trading? If you have actually been able to trade equity/index options on any market with no manual intervention: What broker and API do you use? What instruments do you typically trade (equity versus index options)? Are you satisfied overall with the process and are there pitfalls (speaking only of getting things to work technically)? If you trade on CBOE, how does your broker handle Rule 6.8A? Thanks very much!
As "Professional Customer", you can make two sided markets on AMEX, ARCA, BOX , CBOE and NASDAQ. ISE has a filing to stop it but I'm not sure when the new rule starts. AMEX and ARCA have some limitation, but the others don't. They charge the professional customer rate not the customer rate. We have access to two programs that with a co-location site and a server allow for lo-latency trading, and making two sided markets with auto hedging. You still have to watch and monitor volatility, but they will run your strategy. Let me know if you need more information..... Bob rmorse@victorsecurities.com
It's easy enough to do with the Interactive Brokers API. It's been a while since I did it, but you used to have to slow down consecutive orders or you would get a "pacing violation".
An API would be too slow. It would require a much faster connection and much faster data feeds. You need to be co-located with a very fast data feed or your options valuations will be delayed. It can get very expensive being left out in the market with even the smallest delay. Automated trading in options is much different than stocks.
No it actually works fine using the IB API. But then I was collecting the underlying quotes, not the option quotes, and running a valuation calc before sending in the limit orders to buy/sell the ops. It was really an effort to swing trade the options based on the underlying overbought/oversold. It worked well enough, but I decided that just trading the stocks was less hassle.
Yeah, that caught my eye, too. I used to run a program which got me in, but I had to get out manually. Now it's fully auto, in and out, but I would never leave it alone. I know some guys do, though. Just seems nuts to me.
Thanks to those who've commented. Latency is not a major concern since my automated trading would be on a time scale of minutes or longer. I am interested in an ordinary retail brokerage solution. Good to hear of experience with a particular broker. But, Wayne, you have pointed out in other threads that Rule 6.8A throws a monkey wrench into automated options trading. Who enforces this rule -- only CBOE? Does anyone know the motivation or current status of the CBOE proposal I linked that showed them wanting to repeal it? Does this mean that CBOE will not enforce the rule, even if the repeal is not yet approved? I notice that the rule is mentioned in IB's Canadian customer agreement but not the U.S. one -- any significance to that? Alternatively, is there anyone who has had positive guidance on an approach to fully automated options trading that is compliant -- whether because someone at the brokerage is manually approving your orders, or you are placing a type of order that is exempt from Rule 6.8A, or you are trading on an option exchange other than CBOE? I am trying to make a choice for automated trading: * option-based strategies ** good: higher leverage and defined risk ** bad: wider spreads and the aforementioned regulatory obstacle * stock-based strategies ** good: narrower spreads and no known technical or regulatory obstacle ** bad: at most 4x leverage, requires margin and high worst-case risk If it weren't for Rule 6.8A, I'd strongly favor options. I'd like to know if there's a way to solve that problem. Maybe I should think outside the box and look at options on futures? Fair enough -- I know it has to be done very carefully, but I want a setup that can function while I am at my job. The alternative is only having orders entered overnight, which isn't flexible enough to respond to intraday conditions. A major advantage of options for automated trading is that positions can be entered that will incur a manageable worst-case loss in the event of subsequent market panic, lost connectivity, etc.
I doubt that IB can easily see, through their API, that a button push (or a popup window close) is by a finger/keyboard, or by an app running. And since "OPT" is a valid type for API Order methods, it seems that the rule is not enforced. Probably someone attempting to send in many API option orders much faster than a human can do, would cause some regulatory action. That being said, the "speed bump" limitation that IB builds in is likely to make a fast automated option trading system not worth the effort. What I did (just as a paper test, I'd never knowing violate any trading rules...) was not much more than a "finger substitute" that ran when I was at work, and certainly didn't run very fast. But it did trade options automatically.