EXPE · Expedia Group Inc.
Stock Universe · global stock marketQuantitative Analysis of EXPE Algorithmic Strategy: High Win Rate Positional Dynamics and Asymmetric Tail Risk
This analytical report evaluates a core algorithmic trading strategy developed for EXPE stock, evaluated on a 15-minute price timeframe over a 7.04-year historical dataset spanning from September 2019 to September 2026. Demonstrating a historical win rate of 82.88% across 111 completed trades and achieving a cumulative backtested profit of +18,283.82%, the strategy exhibits exceptional profit factor characteristics (5.00) alongside low winner concentration, where the top three winning trades account for only 10.08% of gross gains. However, quantitative evaluation reveals an underlying structural asymmetry: the strategy's maximum drawdown of 51.11% is directly shaped by its single worst trade loss of -51.11%. Operating with an average position holding time of nearly 350 hours (14.58 days) and executing approximately 15.77 trades per year, the model functions as a slow-frequency positional system suitable as automated decision logic for algorithmic trading bots and trading agents.
Analysis as of 28.10.2026. Strategy data is updated separately.Read full analysis
Strategy profile
The quantitative strategy identified internally as EXPE 0 +18283.82% 1TRAD-RHF4 represents a core systematic trading algorithm engineered for EXPE equity on a 15-minute intraday candle horizon. Across a backtest spanning 7.04 years between September 16, 2019, and September 29, 2026, the strategy executed 111 closed trades, generating an aggregate historical return of +18,283.82%. DevioLab rates this strategy with a quantitative score of 69.11, placing it at rank 2 among analyzed strategies for the EXPE ticker. Designed as a foundational rule set for automated trading bots, trading robots, and execution agents, the system prioritizes high-probability trade setups over high execution frequency. Over its full evaluation sample, the algorithm achieved 92 winning positions against 19 losing positions, resulting in a win rate of 82.88% and a profit factor of 5.00. The mathematical interplay between its profit factor and hit rate indicates robust historical edge, though its long holding periods and deep single-trade drawdown profile require rigorous risk analysis.
Trading rhythm and position duration
Despite utilizing a 15-minute candlestick chart for signal generation, the strategy operates with a multi-day to multi-week holding structure rather than an intraday scalping rhythm. Over the 7.04-year evaluation window, the system generated an average of 15.77 trades per year, which translates to approximately 1.54 trades per active month. Position duration metrics reinforce this patient holding profile: the mean trade holding time stands at 349.99 hours (approximately 14.58 calendar days), while the median holding duration is 239.50 hours (roughly 9.98 calendar days). The divergence between mean and median holding times indicates that while half of all positions close within 10 days, a subset of trades remains open for substantially longer durations. Furthermore, the temporal spacing between trade exits averages 23.13 days, with a median spacing of 19.41 days. This low exit frequency confirms that automated trading agents deploying this decision logic EXPErience extended flat or passive phases between trade completions, avoiding overtrading while subjecting individual positions to multi-day market exposure.
Quality of historical results
An examination of payoff distribution reveals remarkable balance across winning trades rather than reliance on extreme statistical outliers. The strategy achieved an average trade return of +5.64% and a median trade return of +6.10%. The proximity of the median to the average trade return demonstrates consistency in trade EXPEctations, suggesting that historical returns were not artificially inflated by a few massive wins. This broad-based profit distribution is explicitly supported by the concentration metric: the top three winning trades generated just 10.08% of aggregate gross profits, proving that the model's historical performance relies on repeated statistical success across its 92 winning trades rather than extreme luck in isolated events. The single best trade reached +30.39%, which is moderately proportional to the strategy's median win. Combined with a overall profit factor of 5.00, the payoff structure highlights a strategy that historically captured reliable directional swings while maintaining consistent performance across individual trade cycles.
Risk, drawdown and losing behavior
While the strategy boasts a high win rate of 82.88% and a maximum winning streak of 14 consecutive trades, its risk profile features a distinct structural vulnerability. The maximum backtested drawdown reached 51.11%, a magnitude identical to its worst single historical trade loss of -51.11%. This exact numerical convergence indicates that the strategy's historical peak-to-trough decline was driven directly by a single severe trade failure rather than an extended sequence of losses. The strategy's maximum consecutive losing streak was limited to just 3 trades, confirming that losing clusters are infrequent. However, the presence of an unhedged -51.11% loss illustrates that when market conditions breach the algorithm's invalidation parameters, holding duration can expose capital to severe adverse price movements. For automated trading robots implementing this algorithm, managing tail-risk scenarios and single-trade position sizing represents a critical risk control requirement.
Behavior through time and yearly stability
The yearly performance breakdown demonstrates consistent profitability across different market environments throughout the 7.04-year dataset. Following an initial baseline period in late 2019 (1 trade, +4.93% return), the strategy delivered strong positive annual returns in every full calendar year evaluated. In 2020, it logged 16 trades with an 87.50% win rate (+107.21% sum of trades). Performance remained solid in 2021 with 15 trades and an 80.00% win rate (+80.38%), followed by 2022 with 13 trades and a 69.23% win rate (+53.63%). Activity rebounded in 2023 with 18 trades and an 83.33% win rate (+123.12%). In 2024, the system generated 14 trades with an 85.71% win rate (+53.80%), followed by 2025 with 14 trades and an 85.71% win rate (+103.90%), and 2026 with 20 trades and an 85.00% win rate (+99.11%). Trade frequency remained remarkably stable across full years, fluctuating within a narrow range of 13 to 20 trades annually, demonstrating structural behavioral stability across shifting equity regimes.
Recent period since 2024-06-01 versus full history
Analyzing the recent operational window beginning June 1, 2024, provides vital context regarding the strategy's recent historical relevance. Within this period, the strategy recorded 45 completed trades, generating a cumulative trade return sum of +246.46%. Accounting for over 40% of the total 111 historical completed trades, this high density of recent trade completions confirms that the strategy's quantitative patterns remained actively engaged in recent EXPE market structures. The performance during this window mirrors the overall historical metrics, demonstrating sustained win rate consistency and strong gross gain aggregation. Rather than relying on outdated historical anomalies from earlier years, the strategy demonstrates ongoing alignment with EXPE price dynamics up through the end of the evaluated sample in late September 2026.
Strengths and limitations
The chief quantitative strength of this EXPE algorithmic strategy lies in its outstanding win frequency (82.88%), robust profit factor (5.00), and exceptionally low concentration of profits (top 3 trades account for only 10.08% of gross profit). The alignment between the average trade (+5.64%) and median trade (+6.10%) confirms uniform trade performance without dependency on statistical outliers. Conversely, the strategy's primary limitation is its exposure to severe single-trade drawdown. The presence of a -51.11% worst trade matching the maximum historical drawdown highlights an asymmetric tail-risk profile. Additionally, with an average holding time of nearly 15 days and an average of only 15.77 trades per year, the strategy demands significant operational patience and may not suit trading environments that require frequent turnover or tight stop-loss thresholds.
DevioLab analytical conclusion
Ranked 2nd for EXPE with a DevioLab Score of 69.11, this quantitative strategy presents a compelling case study in high-win-rate positional trading. The empirical evidence demonstrates that an automated trading bot or trading robot operating on a 15-minute timeframe can effectively execute multi-day swing holding rules, achieving substantial historical outperformance (+18,283.82% total simulated profit) with a balanced profit contribution across trades. However, quantitative analysts must weigh the system's 5.00 profit factor against its -51.11% maximum drawdown potential. When considering deployment within automated trading agents, incorporating overlay risk controls or dynamic position-sizing constraints is essential to mitigate single-trade tail risk while preserving the strategy's high natural win rate.
Data scope and methodology
The statistical findings in this report are based exclusively on closed historical backtested trades for EXPE stock between September 16, 2019, and September 29, 2026. All metrics, including holding durations, win rates, drawdown levels, and yearly breakdowns, reflect simulated performance generated on 15-minute bar intervals. The recent window metrics evaluate completed trades with exit timestamps on or after June 1, 2024 UTC. This analysis does not include transaction slippage, variable broker commission schedules, order execution latency, or live market impact parameters. Simulated historical performance does not guarantee future results, and this research is published strictly for analytical and educational purposes rather than direct financial advice.