KingdomEdge Algo · Backtest Validation Document
This document expands on the headline backtest numbers (61.3% / 2.75 / +47.1R / 6.3R) with full transparency on the test design, assumptions, limitations, and the forward-validation work currently in flight. It is intended as a companion to the slide deck and the kingdomedgealgo.com /methodology page — not a replacement.
A 62-trade backtest of the KingdomEdge Algo system on SOXL (3x Semiconductor ETF), 1-hour bars, January 2024 through April 2026 (28 months), produced:
These results are indicative, not definitive. A single-ticker, single-timeframe backtest is the floor of the evidence required to validate a system, not the ceiling. We treat this test as one data point in a multi-source validation effort — supplemented by live forward tracking, expanded universe testing, and methodology transparency. This document walks each number through what it means, what it assumes, and what it doesn’t tell you.
| Parameter | Value | Notes |
|---|---|---|
| Symbol | SOXL | Direxion 3x Semiconductor ETF — leveraged, high-volatility |
| Timeframe | 1 hour | Single timeframe; no multi-timeframe confirmation in this test |
| Period | Jan 2024 – Apr 2026 | 28 months; covers a mix of trending and choppy regimes |
| Trades | 62 closed positions | Both directions per Decision Matrix output |
| Configuration | Default preset + Zone Proximity Filter | Standard SurgeU Surge Strategy parameters |
| Risk Model | 1R per trade, fixed-R sizing | Position size adjusted so each trade risks 1R |
| Slippage | Not modeled in this version | Real fills will deviate; see “Limitations” |
| Commissions | Not modeled in this version | See “Limitations” |
| Fills | Assumed at exact entry/stop/target | See “Limitations” |
SOXL was selected because it represents a high-volatility instrument where setup quality is most testable. A choppy, low-volatility instrument would generate fewer signals; a steady mover would produce easier wins. SOXL is a stress-test environment.
Each trade is the product of the SurgeU 6-step Surge Strategy, applied automatically by the KE Algo indicator stack:
Critical: the backtest only counts trades where Steps 1-4 all aligned and the score crossed the threshold. The 62 trades are not the universe of signals on the chart — they are the universe of high-confidence setups the methodology said to take. Lower-confidence setups (LOW grade, score <7) are excluded by design.
38 winning trades, 24 losing trades.
What it means: Above the 50% break-even line, leaving room for the strategy to be profitable even if win rate softens by a few percentage points. Win rate alone is not a measure of profitability — a 90% win rate with tiny wins and huge losses is a losing strategy. Win rate matters in combination with the average win-to-loss ratio.
What it doesn’t mean: It does not mean future trades will win 61.3% of the time. Win rates fluctuate. A 95% confidence interval on 62 trials with a true 60% rate runs roughly 47%-72%. Future student results will be inside that band, not on the point.
Total winnings ÷ total losses = $2.75 / $1.
What it means: For every dollar lost across all losing trades, the system earned $2.75 across winning trades. This is the most important headline number because it captures the win/loss asymmetry that a flat win rate misses. A profit factor above 1.5 is generally considered “good”; above 2.0 is “strong”; above 2.5 means the strategy has substantial cushion.
What it doesn’t mean: Profit factor is path-dependent. A series of small wins followed by one large loss can drag profit factor below 1.0 quickly. The current 2.75 is a snapshot of this 28-month window — a different window could yield a different number. Forward live tracking will tell us the durable value.
Cumulative R-multiple result over 62 closed positions.
What it means: With 1R risk per trade, after 62 trades, the system was up the equivalent of 47.1 risk-units. If a student risked $100 per trade, that’s roughly $4,700 net over the 28 months — before commissions, slippage, taxes. If a student risked $500 per trade, roughly $23,500 net. The metric is denomination-independent — that’s the value of R-multiples.
What it doesn’t mean: Real-world R is degraded by slippage, commissions, missed fills, and behavioral execution gaps. Conservative live results would be expected to be 70-85% of backtest R, not 100%. We expect students to see roughly +33-40R over a similar period if they execute disciplined.
Largest peak-to-trough decline from any high-water mark during the test.
What it means: The worst losing streak cost 6.3R. If a student risked $100/trade, that’s a $630 drawdown at the worst point. A student needs to have psychological and capital tolerance for 6.3R drawdowns — and the live experience will likely produce drawdowns at least 1.5-2x worse than backtest. So plan for ~10-12R drawdowns in real trading.
What it doesn’t mean: It is not the largest drawdown students will ever see. It is the largest drawdown observed in this specific window. Future drawdowns can and will be larger. The Core/Pro/Ultimate tier ladder (Core = learn, Pro = paper-practice, Ultimate = automate) is structured precisely so students experience drawdowns on paper before they put real capital at risk.
The backtest as currently configured makes several simplifying assumptions worth surfacing:
To address the limitations above, the validation roadmap is:
If you’re considering whether to recommend KE Algo to your students, here’s the honest read:
This document represents one student-facing read of the backtest evidence. Numbers are accurate to the underlying TradingView strategy tester output as of 2026-04-23. Past performance does not guarantee future results. Backtests reflect historical price behavior under specified configurations. KingdomEdge Algo welcomes detailed scrutiny of the methodology and the validation work in flight.
Compiled on 2026-05-08.