Blind Trust: With Great Algo Wheels Comes Great Responsibility
Having tools to fairly and intelligently distribute flow across electronic brokers is not just a luxury but an absolute necessity. For buy-side traders to offset regulatory responsibilities to the algo wheel logic, however, they should understand and validate the decision-making logic and all analytics feeding the wheel, as the ultimate responsibility and accountability lies with them. Here are 5 questions that should be answered before any algo wheel is implemented into a trader’s workflow.
Algo Wheels. The latest buzzword. The perfect response to MiFID II.
It’s not hard to recognize that having tools to fairly and intelligently distribute flow across electronic brokers is not just a luxury, but an absolute necessity, especially as more buy-side traders, are working towards consolidation of their trading technologies and trading desks. We now need more tools to help simplify the workflow and become more systematic. The ability to automate more of the ordinary orders effectively frees up execution traders to focus on the trades where they can and will add the most value.
Several factors accelerated the need for automation, but we also need to take greater responsibility for what automation means.
Adoption of algo normalization became more prominent once the MiFID II regulation placed responsibility for execution quality and best practice onto the buy-side community. Suddenly the buy-side had to prove ‘best execution,’ justifying how, where and with who they traded. The broker selection process became anonymized and unbiased and, in turn, gathering objective, and unbiased performance data fueled a feedback loop where brokers are selected solely based off of performance.
The future promises even more artificial intelligence, more data-driven analytics, and more automation. Technology can simplify the most complex, data-driven decisions and now allows the buy-side to become a lot more systematic and introduce (and benefit from!) automation into their workflow. It is no surprise that the current algo wheel has now become a prominent tool in the industry and used as a critical part of the buy-side execution.
But what insights does the trader have into all of the nuances at work by the algo wheel?
As the algo wheel became a more established practice, different “flavours” developed. Where some wheels not only select the broker but also defines the trading strategy, we now see how algo wheels offset the buy-side’s regulatory responsibility of “how” and “with who” to the algowheel logic.
And just as algo wheels have become a necessity, so is the need to demand transparency. Adoption will continue to rise as traders cannot afford the opportunity costs of executing increasingly higher volumes of trades in an increasingly complex market.
So, for a buy-side to offset the regulatory responsibility to the algowheel logic, one could argue that they should understand and sufficiently validate the decision-making logic and all analytics feeding the wheel. The buy-side should not blindly trust the machine; they should demand transparency into the decision-making logic as well as back-testing their vendors results. The buy-side must demand enough transparency from their algo wheel providers with regards to data integrity, sufficient data, and analytics testing. In turn, vendors must provide this transparency.
We have yet to reach a point where traders are no longer responsible for the decisions they are making; however, it is our responsibility to avoid a fate of walking blind (or rather, trading blind). With great and sufficient transparency and comprehensive reporting, comes great responsibility and the ultimate responsibility and accountability of the decisions, as well as execution quality, lies with the buy-side.
The focus has now shifted towards reporting. Given the reliance on data and performance analytics, are people asking the right questions to evaluate their algo wheel logic? At TradingScreen, we believe that these five questions should be answered before any algo wheel is implemented into a trader’s workflow to ensure that these algorithms are making the right decisions:
- Is there transparency into the algorithm logic or are vendors keeping it proprietary?
- If they keep it proprietary, are these algorithms sufficiently back-tested against historical data and manual trading behavior?
- Do vendors even have enough data to back-test against a firm’s manual trading behaviour and are the discrepancies sufficiently understood?
- Most importantly, are the vendors giving the buy-side transparency into these results?
- How might you be able to validate these results in-house?
Surika Vosloo is the European Product Manager at TradingScreen.
The Dollar Milkshake Theory – Stress Testing the Potential Impact of a Sovereign Debt and Currency Crisis
Stress Tests play an important role in improving financial stability by enhancing market discipline and transparency. Here we consider the third example of these stresses based on current market conditions and show how the risk factors can be generated along with their relevant shock examples.
Since the start of the war in Ukraine, Western nations have imposed sanctions on many Russian individuals, businesses and state-run enterprises. Stress Tests play an important role in improving financial stability by enhancing market discipline and transparency.
Stress Tests play an important role in improving financial stability by enhancing market discipline and transparency. Clients leverage TS Imagine’s flexible and powerful risk engine to stress test a variety of specific scenarios. With Americans set to commence voting early next week, markets around the globe are watching carefully. Will the midterms disrupt the Democratic legislative agenda or tip in their favor? We ran two scenarios through our stress-testing engine, gaming the potential market impacts. Please note that these are hypothetical shocks and are not a prediction of any forthcoming events.