Curriculum·R404 Execution and Trade Craft·about 33 min
Exits
By the end of this lesson you can
- →Name the three reasons to exit and match each to the entry model that produced the trade
- →Compute whether scaling out beats a single target, from your own record
- →Explain why an exit that requires your presence is conditional rather than planned
- →Estimate the probability that a venue is unavailable when you most need it
Senior · enrolled learners
This lesson opens with The Robinhood outages, 2 and 3 March 2020.
- What happened
- Robinhood's website and mobile applications went down on 2 March 2020 and were unavailable for the entire trading session, with further disruption on 3 March, during some of the highest volume days of the fastest bear market on record. Customers could not access their accounts, place orders or close positions. FINRA found that from January 2018 to February 2021 the firm had failed to reasonably supervise the technology it relied on to accept and execute customer orders, in violation of FINRA Rules 3110 and 2010, and noted that although the firm had a business continuity plan it did not apply it, because the plan covered only events affecting the company's physical location. On 30 June 2021 Robinhood agreed to pay $57M in fines and $12.6M in restitution, the largest financial penalty FINRA had ordered.
- The decision point
- Every customer with a plan to exit that day held a plan conditional on being able to reach the venue. That condition had never been written down, because it does not look like part of the strategy. An exit that requires you to be present, connected and able to place an order is not an exit, it is an intention with a dependency, and the dependency fails on exactly the days the exit was designed for.
- Recorded loss
- $70,000,000
What you will be able to answer
- →What are the three reasons to exit?
- →When does scaling out beat a single target?
- →Why is a manual exit conditional?
- →How likely is that dependency to fail?
Orientation and Year One are open: anyone can read them without an account. From Year Two onward the lessons are for enrolled learners, because progress through the later years only means anything if it is tracked against a record.
It is free. We do not sell the list and there is nothing to buy at the end of it.
Sources and review
- https://www.cnbc.com/2021/06/30/robinhood-to-pay-70-million-for-misleading-customers-and-outages-the-largest-finra-penalty-ever.html
- https://www.investmentnews.com/fintech/robinhood-to-pay-record-70-million-in-finra-settlement/208325
- https://lowey.com/blog/finra-hits-robinhood-with-70m-fine-and-restitution-penalty/
- https://www.sec.gov/newsroom/press-releases/2020-321
Confidence high·Volatility low·Reviewed 2026-08-07·Owner unassigned
Contested
The FINRA action covered several distinct failures including misleading communications and options approval, not only the outages, and the $70M figure is the total across all of them. This lesson uses the outage findings and does not attribute the whole penalty to them.
The conditional outage model in part three is an illustration with stated assumptions, being 99.9 percent headline uptime and outages ten times more likely on high-volatility days. Venues do not publish conditional availability and the multiplier is not measurable from outside. The conclusion, that headline uptime overstates availability when you need it, holds across any multiplier above one.
R404-05 uses the December 2020 SEC action against the same firm, which concerns execution quality and payment for order flow rather than availability. Two different failures at one firm, and the split is deliberate. Keep it.