LearnOctober 22, 2026 · 2 min read

How Mangrove matches survey answers to trading strategies

Tim Darrah

Mangrove’s onboarding survey asks about the assets you follow, your goals, your investing familiarity, the growth you seek, and the decline you would tolerate. Survey-Based Strategy Selection explains how those answers become explicit requirements for recorded strategy results.

Your asset choice filters the candidate pool. The remaining answers determine separate risk and reward levels. The risk level sets a maximum historical drawdown and minimum Sharpe and gain-to-pain ratios. The reward level sets the minimum return difference against buying and holding the same asset over the same period.

A strategy qualifies only when all four conditions pass. A larger historical return difference cannot compensate for exceeding the drawdown ceiling. Missing or invalid measurements also fail selection.

A worked BTC example

The paper follows a profile with a balanced goal, some investing experience, modest growth ambition, and a 15% acceptable decline. That profile requires a recorded Sharpe ratio of at least 1.50, drawdown no more than 15%, gain-to-pain of at least 1.50, and a return at least 10 percentage points above BTC buy-and-hold over the same evaluation window.

Risk level 3 and reward level 3 select historical strategy requirements: Sharpe at least 1.50, drawdown at most 15%, gain-to-pain at least 1.50, and same-asset outperformance of at least 10 percentage points. — Figure 2. A risk level 3, reward level 3 profile maps to compatible historical strategy placements. Each candidate must pass all four measurement checks.

Three historical BTC evaluations meet those conditions. Change the decline answer to unsure, and the method assigns risk level 2 while retaining the answer as unsure. The drawdown ceiling becomes 10%, and the Sharpe and gain-to-pain floors become 1.75. The reward requirement stays the same. Two examples still qualify; the third fails because its recorded drawdown is 12.88%.

This default is a starting policy. It does not establish that an unsure user is comfortable with a 10% loss, and a historical drawdown ceiling does not cap future losses.

What was checked

All 108 combinations of the non-asset answers produce requirements. The 81 combinations with a numerical decline answer reach all 25 risk/reward cells. Comparing 2,812 unambiguous historical records with those 25 cells produces 70,300 comparisons, with complete agreement between direct matching and strategy placement.

These results establish computational consistency. They do not establish preference validity, future investment performance, or present-market eligibility. The paper’s BTC examples use data from February 17 through August 17, 2020. User studies and out-of-sample performance evaluation remain necessary.

Read the methodology

The six-page paper includes the survey scoring rules, the complete threshold policy, profile diagrams, the historical BTC example, and limitations.

Download Survey-Based Strategy Selection (PDF), by Timothy Darrah, PhD, Mangrove Technologies, October 20, 2026.

For a longer plain-language explanation, read Timothy Darrah’s Substack article. You can also explore Mangrove.

Mangrove provides trading tools and infrastructure, not financial advice. Digital asset trading involves significant risk, including potential loss of principal. Past strategy performance, whether backtested or live, does not guarantee future results. Users should consult a qualified financial advisor before making investment decisions.