As the market leader on risk, we can’t stay silent.

The proliferation of fintech firms claiming to be able to guess how investor portfolios will behave compels us to highlight the stark philosophical differences between the “Historical Data Model” Riskalyze has always leveraged, versus the “Predictive Guesswork Model” used by a few others.

It’s time to bring sunlight to a rare, but dangerous risk modeling framework.

These differences have profound implications for the liabilities shouldered by financial advisors and wealth management enterprises as they serve investors, and we believe risk analytics should control and reduce those liabilities, not increase them.

More importantly, deeply flawed predictive methodology puts the mission we share with these advisors — empowering the world to invest fearlessly — at risk.

That’s why we have consistently won plaudits from compliance teams, enterprise executives, and regulators. They understand that Riskalyze isn’t a prediction engine. Instead, we use objective data to calculate historical ranges of risk for portfolios, and help investors react to risk appropriately, so they can invest without fear.

Why speak up about the small number of providers who use dangerous methodology? Isn’t Riskalyze’s approach winning in the market?

It’s true: over 4x as many advisors have chosen Riskalyze as all of the other risk solutions combined.1 We serve tens of thousands of financial stewards who have a duty to act in the best interests of their clients. We can’t allow flawed methodology to hold these advisors back from being their best.

A tale of two completely different approaches.

Advisors should beware: there are two very different approaches to risk analysis and stress testing. Riskalyze leverages a Historical Data Model and calculates a historical range to illustrate risk and support client behavior.

Some tools use a Predictive Guesswork Model that attempts to forecast what complex markets and portfolios will do.

These are two fundamentally different views of how to approach risk.

Predictive Guesswork Model

Historical Data Model

Make guesses about what will happen to a variety of factors in a complex market, and try to model the impact on a portfolio if those guesses are correct. Use objective data to calculate a range of historical probabilities for a portfolio, based on the actual risk in the underlying securities. Strenuously avoid subjective assumptions that can lead to guesswork.
Approach to Portfolio Stress Testing:
Here’s what will happen to your portfolio as long as we’re right about inflation, currency, the stock market, interest rates, oil prices, commodity prices, the impact of certain news events, a host of other assumptions, and the validity of the calculation under the hood.
Approach to Portfolio Stress Testing:
Let’s use historical data to look at your current portfolio through the lens of the 2008 bear market and see how the math says it would have behaved then.Let’s use a previous interest rate hike to see how the math says your portfolio would have behaved in that kind of environment.
Data Inputs:
Onus is on the advisor to guess a multitude of inputs, or trust the provider’s guesses for how the factors might react.
Data Inputs:
Market data feeds historical scenarios to objectively paint a more stable and accurate picture of markets over the long term.
Methodology Outputs:
Somewhat of a black box.
Methodology Outputs:
Discussed extensively at and in a decade worth of white papers and knowledge base articles.
What happened in 2020:
One tool predicts the S&P 500 would fall by 36%.The S&P 500 actually ended 2020 up 22%.
What happened in 2020:
Riskalyze’s 95% Historical Range illustrated the historical likelihood of upside and downside within six months, and a reminder that there are 5% probability events that can create additional downside risk.Still, there was no measurable increase in client portfolios falling outside of the 95% Historical Range in any six month period, even if you cherry-pick the high point before the 2020 market crash.
Forecasting market downturns is akin to flashing a red, neon SELL sign in front of clients, and then asking them to remain calm.Clients who sold at the bottom of the pandemic crash missed the incredible snap-back that allowed markets to recover and generate substantial return during the rest of 2020.
The 95% Historical Range puts the best principles of behavioral finance to work by reminding clients that loss and gain are normal behaviors for their portfolio, and reminding them of the long-term wisdom of their advisor’s recommendation.The clients supported by solid risk analytics who stayed fearlessly invested enjoyed substantial risk-adjusted returns in 2020.
Value to advisors:
Fearful clients focused on speculation.Incentivizing the wrong behaviors.Flawed software discounted to “free.”
Value to advisors:
Fearless clients invested for the long term.Harnessing behavioral science to protect the advisor/client relationship.Incredible return on investment.

Not only is the Predictive Guesswork Model fundamentally flawed, the providers who use it are wildly inaccurate.

If you’re going to make predictions, you’d better be accurate when your clients’ future is on the line.

In a March 2020 webinar, one tool shared a risk model that said the S&P 500 should be expected to drop 24.8% in the coming year, when in fact the S&P 500 returned a positive 22%, a difference of 47%. Another webinar predicted the S&P 500 would drop 36% in the coming year, a difference of 58%!

Here’s a table that illustrates the danger of the Predictive Guesswork Model, with the eleven holdings this company tried to predict in March 2020.

In another example, when one tool’s crash test modeled oil dropping 50%, they projected that three different funds would therefore be down between 22% and 38%.

Soon after, it just so happens that we got to see the scenario play out in real life. Oil actually did drop 50%, and all three funds went up.

If you had been lucky enough to guess the price of oil correctly, you would’ve still would have gotten it completely wrong — all because of bad methodology.

Now imagine having to correctly guess on ten different factors.

Which leads us to the critical question: do you want your risk solution to create fear, or fight it?

The approach championed by others — modeling doomsday scenarios based on guesswork about how they will impact market factors — does nothing but incite fear in your clients. It’s like a neon red “SELL” sign flashing in their faces. It completely undermines your message as an advisor, and exacerbates a client’s natural tendency to make fear-driven decisions that can blow up their financial futures.

Riskalyze was founded on the belief that you solve the behavior problem, and fight fear, by helping clients see what is normal behavior for their portfolio using the 95% Historical Range, and educate them about 5% downside events using historical stress tests. When you do that effectively, you empower that client to make fearless investing decisions that lead to great long-term financial outcomes.

There’s no discount high enough to offset the downside of predictive guesswork.

White Paper: Differentiating the Historical Data Model

Download our white paper to dive deeper into the distinction between historical objectivity and predictive guesswork, and see the results of each approach.


Want to talk to a specialist about putting the Historical Data Model into practice?

Get A Riskalyze Demo