Beyond Broker Scorecards: What’s Next for Algo Wheels?

“Everyone gets the experience, some get the lesson” -T.S. Elliot

Trading involves a bewildering number of individual choices. Each order could be traded in a huge number of different ways and every market participant gets to witness the outcomes. Everyone gets the experience, but do they also get the lesson? As an industry, do we structure our decision making so that we can best evaluate which strategies are working?

As you might suspect our answer to this question would be a firm no. Empirical methods for continuously improving processes and decision making have not been nearly as widely adopted by the trading community as they have in industries like healthcare, advertising, and communications.

To date, algo wheels have been the most visible evidence that the philosophy of continuous process improvement has been making inroads into the trading community. By submitting broker selection to randomized controlled trials (RCTs), buy-side firms can generate unbiased broker performance statistics. This leads both to improved trading performance in the short term, and in the long term, stronger incentives for brokers to invest in their execution capabilities. This article surveys the current algo wheel landscape; looks at what the trading community can learn from applications of RCTs in other industries; and anticipates future developments in this space.