Back to all posts

Making Michael Lewis Proud: Algorithmic Investing through IEX

Journalist-story-teller Michael Lewis ignited a frenzy of debate and acrimony over high frequency trading with the release of his book, Flash Boys. We adore his writing, and his singular ability to shine a spotlight on esoteric fields. His book has made the arcane, complex, and sophisticated field of ‘high-frequency’ trading dinner table talk across America. While the level of discourse has ranged widely, from technical critiques of market microstructure, to downright theatrical showdowns on the cable news networks, we believe in the power of transparency.


 For those of us in the financial services or investment management fields, there was arguably not much ‘new news’ in Lewis’ Flash Boys – the financial incentives to find and exploit latency (or any type of) arbitrage in public markets have always been and will always be considerable. In the best of worlds this is a good thing, ruthless competition to exploit inefficiencies should minimize those same inefficiencies with all market participants benefiting from faster, cheaper order execution. The problem arises when, through regulatory oversight or incompetence, we allow conflicts of interest to form within our capital markets such that inefficiencies are not minimized through competition, but are allowed to balloon in a self-reinforcing manner.

One might argue that Flash Boys came late to the HFT party, and market forces had already begun to deflate the balloon of profits that could be made (at least in US markets) from latency arbitrage.  And yet, anyone who would dismiss the signs that the ‘investing public’ is becoming an increasingly well-educated, technologically savvy, and demanding customer base with a passionate interest in the fairness of their financial markets is not paying attention.

At Quantopian, we believe that broader interest in capital markets will lead to more educated investors, and that market forces will continually drive creative solutions to anomalies such as latency trading. Therefore, we are extremely pleased to be able to deliver new functionality to our customers whereby orders placed with Quantopian algorithms can be routed to the IEX trading venue for execution. IEX is a stock trading platform launched late last year with the express goal of negating latency arbitrage by introducing a delay of 350 microseconds in the execution of trades. If you aren't already managing your algorithmic trading with Quantopian you can test out a basic investment strategy for free*, or to learn how to route the trades from your existing account to IEX check out a sample order.

We fully expect the conversation on fairness and transparency in the US stock market to continue to evolve, both at the regulatory level and in the court of public opinion.  From our vantage point the inexplicable lack of automation and algorithms in the investment process is an even bigger target for reform than market microstructure. While trading has become essentially automated (albeit with bumps along the way), asset allocation and portfolio management remain stubbornly manual and by extension artificially expensive. The premium investors today are paying for even the most formulaic portfolio management services can be on the order of 1-2% per year, a rate which frankly dwarfs the per-investor costs of something like latency arbitrage. At Quantopian our mission is to continue to put the power in the hands of investors to help them drive down costs while getting superior performance from their ideas and we think that mission is well served by routing trades to IEX and doing our part to promote increased transparency and fairness in our capital markets.

Learn more:


*Live trade with Quantopian by May 19th and we'll give you two years of free access for accounts up to $100,000.

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian.

In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.

Grant Kiehne

Hello Jess,

What do you see as the path to get the relative cost of trading down? As a back-of-the-envelope example, if I take your example of 9 ETFs and wanted to re-allocate daily, the annual cost would be:

$1/trade * 9 trades/day * 252 trading days/year = $2268

So, to get down to even the 2% level, the capital required is:

$2268 / 0.02 = $113400

So, it seems like there is a pretty big gap. Or am I missing something? Any idea what drives the minimum trade cost at Interactive Brokers and elsewhere? How would Quantopian push the cost down?


Jess Stauth

Hi Grant,

I think there are probably at least two different categories of strategies to consider transactions costs for. The first would be considered 'low frequency' strategies, that make fewer than say 4-5 round trip trades per week. That category of strategy is a good candidate for smaller capital base accounts since the transaction costs will be less frequent and Reg T account limits don't apply -- but on the flip side the 'per transaction' relative cost might be a bit higher since you'll likely be triggering minimum trade costs (like you said at IB it's around a dollar).

The 9 Sector ETF example I posted falls into this first category - I wouldn't suggest rebalancing that strategy more than monthly in live trading since I don't think there would be much benefit in it, the example we've been sharing for people to deploy out of the box rebalances every 21 days for example. Even at that rebalance frequency you'll find that you won't end up needing to place trades in all 9 stocks to reach your target allocation. In the first 4 months of trading this strategy we've seen it trigger on average 3 trades every 21 days, which annualizes to a cost of around $51 per year. Of course if you use a retail broker with higher minimum trade costs like an E-trade or a TD Ameritrade then you can multiply that by 5 - 10x and I agree it gets less attractive.

The second broad category of strategy someone might deploy on Quantopian would be a faster turnover, perhaps even classical 'intraday' strategy. If the strategy is placing many round trip trades per week or even day then a few things change, first you need a higher account balance to satisfy Reg T 'day trader' limits. Second you have a much higher bar for the returns your strategy must produce to clear your transaction costs. Now if you are trading enough shares with IB that you are not hitting the minimum trade costs, and if your strategy is well suited to intraday turnover, you can still make money this way. Or, of course you can lose money this way quickly as well 🙂

Overall since I think you're more interested in the former category I'd say that the costs to an individual of managing a portfolio via IB are about the lowest of any I've seen in the market. It all comes down to tailoring the strategy to make sense given the account size and cost structure.

Best wishes,

Grant Kiehne

Thanks Jess,

Regarding the 9 ETF example, I picked it because it has broad market coverage and basically mirrors the default SPY benchmark for an equal-weight allocation. So, to get a feel for trading cost, the hypothetical is suppose I found an active algorithm that provided an edge over a fixed allocation (maintained with relatively infrequent re-balancing, as you point out). I understand that IB has the lowest cost for an online brokerage (and Quantopian has put some very high-end trading capabilities into the hands of the mass public), but there is still a high minimum capital requirement due to trading cost.

On a separate note, my hunch is that at the small-account retail level (<$1M), most "portfolio management services" are highly automated and formulaic and still charging 1%-2%. Back-of-the-envelope, I figure access to someone with any expertise at all would run $150-$300 per hour ($1200-$2400 per day). Assuming even one day per month, it would cost $14,400-$28,800 per year. The implication is that it is impractical to have "active" management--in the end, it's gonna be some form of an automated system, with 1%-2% charged for data entry and "advice" from a sales rep. In other words, it's a rip-off, as you imply.

To me, for infrequent investment portfolio adjustment (weekly/monthly/quarterly), the question is what is the potential benefit of Quantopian/IB over what could be done at Vanguard with index funds, for example?


IEX solves a problem that retail investors do not have. Latency arbitrage is simply HFT eliciting smart order routers' demand, and they will move the markets away from informed traders regardless. I strongly suspect that retail investors that choose to tie their hands behind their back by routing only to IEX will find they have higher slippage costs, if the bother to do an analysis at all, but I look forward to being proven wrong.

Comments are closed.