FICO · Fair Isaac Corp.
Stock Universe · global stock marketQuantitative Strategy Analysis: FICO 15-Minute Algorithmic Model Demonstrates High Win Rate and Multi-Year Structural Persistence
This analytical report evaluates a core 15-minute algorithmic trading strategy for FICO (Fair Isaac Corporation) covering 91 completed trades across a seven-year historical dataset from September 2019 to October 2026. Achieving a DevioLab score of 74.94 and ranking second for the asset, the model pairs a remarkably high win rate of 79.12 percent with a cumulative backtested return of +9,360.18 percent. Operating with a low trade frequency of 12.89 trades per year and extended position durations averaging 471.23 hours, the strategy demonstrates high consistency and low winner concentration, with the top three winning trades generating only 17.75 percent of gross profit. However, a profit factor of 1.25 alongside a worst trade of -20.48 percent reveals a structural asymmetry where infrequent losing trades offset a notable portion of accumulated gains. This study details the strategy's operational rhythm, risk parameters, and recent post-June 2024 performance density.
Analysis as of 31.10.2026. Strategy data is updated separately.Read full analysis
Strategy profile
The quantitative strategy under review, designated as FICO 0 +9360.18% 1TRAD-DYE6, represents a core algorithmic model engineered for FICO stock on a 15-minute execution interval. Spanning a simulated evaluation history from September 16, 2019, through October 6, 2026—a timeframe encompassing 7.06 years—the model recorded 91 completed trades. Over this sample, it generated a total historical return of +9,360.18 percent, corresponding to an annualized figure of 527.24 percent under the model's standardized compounding framework. The strategy achieved a DevioLab score of 74.94, placing it as the number two ranked model for FICO within the DevioLab library. Engineered as decision logic suitable for execution via automated trading bots, trading robots, or quantitative trading agents, the strategy utilizes the 15-minute timeframe not for high-frequency scalp entries, but rather for refined execution timing within broader trend cycles. Across the 91 closed positions, 72 concluded in profit while 19 resulted in losses, establishing a overall win rate of 79.12 percent.
Trading rhythm and position duration
Despite utilizing a 15-minute bar interval for entry and exit signals, the statistical cadence of the strategy reflects a patient, medium-term holding profile rather than intraday churning. The model completes an average of 12.89 trades per year, which translates to approximately 1.52 trades per active month. Position holding times reveal substantial duration: the average holding time across all completed trades is 471.23 hours (approximately 19.6 days), while the median holding duration sits at 259.25 hours (roughly 10.8 days). This positive skew between median and average holding times indicates that while many trades exit within eleven days, select multi-week trend holdings extend the overall statistical mean. Furthermore, the spacing between consecutive exits averages 27.08 days, with a median exit interval of 20.96 days. For automated trading agents or algorithmic execution frameworks, this low-frequency rhythm minimizes active market exposure and avoids the elevated transactional friction typical of lower-timeframe strategies.
Quality of historical results
The quality profile of the historical returns presents a noteworthy statistical structure characterized by high directional accuracy alongside moderate trade payoff asymmetry. The strategy achieves a mean trade return of +5.90 percent and a median trade return of +4.29 percent, reflecting a reliable positive baseline across closed positions. The single best trade produced a gain of +43.84 percent, whereas the worst trade incurred a loss of -20.48 percent. Notably, the profit factor stands at 1.25. The divergence between a 79.12 percent win rate and a modest 1.25 profit factor indicates that while nearly four out of five trades close in positive territory, the average magnitude of losing trades is considerably larger than that of winning trades. On the positive side, gross profits are exceptionally well-distributed: the top three winning trades account for only 17.75 percent of total gross profit. This low concentration ratio confirms that historical performance was driven by repeatable, broad-based wins rather than being dependent on a handful of anomalous outlier events.
Risk, drawdown and losing behavior
Risk metrics across the 7.06-year test period highlight key boundary parameters for capital management. The strategy experienced a maximum historical peak-to-trough drawdown of 21.61 percent. When evaluated alongside the cumulative total return of +9,360.18 percent, the depth of this drawdown demonstrates reasonable resilience against catastrophic structural loss. Risk distribution across trade sequences shows strong historical stability: the strategy recorded a maximum winning streak of 15 consecutive trades, whereas the longest losing streak was contained at 3 consecutive trades. However, individual loss severity remains the primary risk factor. With the worst single trade registering at -20.48 percent, a rare loss can erase the net gains of multiple typical winning trades (given the median win profile). For deployment in automated trading systems, managing tail risk through calibrated position sizing or stop mechanisms is crucial to maintaining equity curve smoothness during adverse regime shifts.
Behavior through time and yearly stability
Analyzing annual performance reveals distinct operational phases across the strategy's backtest history. In 2020, the strategy generated a sum return of +138.69 percent across 8 trades with an 87.50 percent win rate. Activity was lower in 2021, yielding +3.64 percent over 4 trades with a 50.00 percent win rate. The year 2022 marked a rebound, generating +82.44 percent over 9 trades (66.67 percent win rate). Performance expanded significantly in 2023, delivering +94.59 percent across 16 trades with a 93.75 percent win rate (15 wins, 1 loss). High performance and trade density continued in subsequent years: 2024 produced +71.32 percent across 17 trades (76.47 percent win rate), 2025 yielded +63.50 percent over 20 trades (75.00 percent win rate), and the partial year 2026 added +82.73 percent across 17 trades (82.35 percent win rate). The year-by-year evidence shows that trade frequency effectively doubled from 2023 onward while maintaining win rates consistently above 75 percent.
Recent period since 2024-06-01 versus full history
The performance window starting June 1, 2024, offers critical insights into the strategy's recent historical behavior. Of the 91 completed trades in the entire 7.06-year dataset, 47 trades—over 51 percent of all historical activity—occurred on or after June 1, 2024. During this recent window, the strategy accumulated a sum of trade returns equal to +209.16 percent. This concentration of trade completions indicates a pronounced shift toward higher signal generation in recent market conditions compared to the sparser trade volume observed between 2019 and 2022. Rather than degrading under recent market dynamics, the strategy maintained strong profitability and high signal density, suggesting that the underlying quantitative rules remained well-aligned with FICO stock price behavior throughout 2024, 2025, and into late 2026.
Strengths and limitations
The primary strength of this FICO model lies in its exceptional directional reliability, evidenced by a 79.12 percent win rate, a 15-trade winning streak, and a robust +9,360.18 percent cumulative historical return. Its low reliance on top winners (17.75 percent profit concentration) demonstrates organic profitability spread across dozens of trades. Furthermore, its extended holding periods reduces active trading noise and transaction overhead. Conversely, the model's main limitation is its asymmetric loss structure, reflected in a modest profit factor of 1.25 and a worst trade of -20.48 percent. Because losing trades carry disproportionate magnitude, even a short cluster of losses can temporarily impact equity growth. Additionally, with 91 total trades spread over seven years, the overall statistical sample remains relatively compact, requiring careful validation before live execution.
DevioLab analytical conclusion
With a DevioLab score of 74.94 and a rank of #2 for FICO, this algorithmic strategy presents a compelling case study in trend-following efficiency on a 15-minute execution frame. The quantitative evidence demonstrates that low-frequency signal generation paired with multi-day position holding can capture substantial multi-year appreciation while keeping drawdown bounded to 21.61 percent. The high density of recent profitable trades since mid-2024 reinforces its ongoing statistical relevance. Quantitative developers seeking to integrate this logic into automated trading bots or execution agents should account for the payload asymmetry between average wins and maximum losses by applying prudent leverage bounds and risk management protocols.
Data scope and methodology
All statistics in this report are derived directly from backtested simulated trade histories for FICO stock covering the period from September 16, 2019, to October 6, 2026. The dataset evaluates 91 completed trades executed on a 15-minute price interval. Standardized calculations assume fixed-percentage return summing and compound performance metrics without accounting for variable slippage, exchange fees, or corporate actions. The recent evaluation window consists of all positions closed on or after June 1, 2024. Historical backtested results are presented strictly for quantitative research purposes and do not constitute investment advice or guarantee future trading performance.