Three layers of disclosure
Every number on Puravida Edge falls into one of three layers. We disclose which layer each metric comes from to be transparent about what is verified empirical data versus what is modeled projection.
Layer 1 · Verifiable
Trade counts, win rates, profit factors, gross profit/loss, directional P&L breakdown, average win/loss, max drawdown values come directly from raw TradingView Strategy Tester exports. These match TradingView output 1:1.
How to verify: after purchase, run the strategy on your own TradingView account, set the date range to May 1, 2025 – April 30, 2026, and check the Performance Summary. The numbers should match the published stats exactly.
Layer 2 · Computed
Sharpe ratio, Sortino ratio, Calmar ratio, Profit-to-DD, DD%, SL%, Pass:Blow ratio, and joint daily P&L distributions are computed from the same trade lists using industry-standard methodology (annualized from daily P&L).
How to verify: raw trade lists for every strategy × instrument are available on request. Anyone with the trade data can reproduce these ratios in Excel, Python, or R.
Layer 3 · Modeled
Time-to-payout, payouts/year, modeled Net $/year, blow rate, and viability come from a 1,500-path Monte Carlo simulation over a 3-year horizon. These are probabilistic projections, not historical facts.
How to verify: the Monte Carlo model source data and a sample of simulation paths are available on request for audit.
Monte Carlo simulation specifics
- Paths: 1,500 alternate trading sequences generated per (strategy × instrument × account × sizing) combination, and per joint portfolio.
- Horizon: 3 years per path (approximately 756 trading days).
- Resampling: block bootstrap with 5-day blocks — preserves serial autocorrelation of trade streaks (a real strategy doesn't shuffle trades randomly; winning days cluster, losing days cluster).
- Dropout: 25% random masking of trade days (20% when sample n ≤ 30 trades) — simulates regime stretches with no signal and worse day-mixes than the empirical sample.
- State machine: each path simulates the full prop firm account lifecycle: challenge phase → funded state → payouts → potential blow-out at any point.
- Profit splits applied: 90% Futures Prop (Apex, MyFundedFutures, Topstep, Tradeify, Lucid). 85% Forex Prop, sized for a $5,000 DD (FTMO, FundingPips, The5%ers).
Three percentiles · not just an average
For every modeled metric we publish three percentile outcomes from the 1,500 paths:
- P10 (Worst): the 10th percentile path. 90% of simulated paths perform better than this; 10% perform worse. Disclosed in the Risk Disclosure table on the Portfolios page.
- P50 (Median): the typical outcome. 50% of paths perform better, 50% worse. This is the "most likely" single number.
- P90 (Best): the 90th percentile path. Only 10% of paths perform better. Headline numbers ("Fastest time to first payout") use P90.
We disclose all three rather than only the median because the spread between P10 and P90 is itself important risk information. A strategy with P10 $7k / P90 $80k is fundamentally different from one with P10 $25k / P90 $35k, even if both have the same median.
Joint portfolio simulation
Single-strategy stats and joint portfolio stats are simulated differently. For portfolios (multiple strategies on one prop firm account):
- Joint daily P&L is constructed by summing per-strategy daily P&L for each calendar day (preserving real-day correlation between strategies).
- The 1,500-path MC operates on this joint distribution against a shared hard DD floor — the account's single drawdown limit applies to the combined position, not per-strategy.
- This captures both the diversification benefit (lower joint variance when losing days don't overlap) and the constraint (overlapping setup days can compound exposure on a shared floor).
- Result: every joint portfolio number on the Portfolios page comes from this joint MC, not from naive sum-of-singles.
SAFE filter · blow rate threshold
No published portfolio variant has a modeled annualized blow rate above 15%. This is our SAFE threshold — any sizing combination that exceeds it is excluded from the published roster. The threshold is conservative; institutional risk tolerance is typically much lower, but the 15% line draws a clean boundary between "deployable" and "too aggressive".
Three Futures portfolios sit between 10% and 15% (the 10% AT-EDGE yellow line). These remain inside SAFE but trade higher absolute return for higher blow probability. We flag this clearly with amber-highlighted Blow/y values in the sizing tables.
What the model does not capture
Monte Carlo is built on the empirical sample. It can't simulate regimes outside that distribution:
- Regime shifts — we assume the future P&L distribution resembles the past 12 months. Major structural changes in futures market microstructure (CME rule changes, dramatic volatility regime shifts) are not modeled.
- Evaluation fees and reset costs — prop firm fees vary by firm and product. We do not subtract these from modeled Net/y; you need to factor them in based on your specific challenge fees.
- Minimum trading-day requirements — some firms require X trading days before funded status. The model doesn't simulate this constraint explicitly.
- Payout caps — some firms cap per-cycle payouts (~$2.5–3k). The model assumes uncapped payouts up to the simulated profit.
- Extreme tail events — only visible over 5+ year horizons in the empirical data. The 3-year MC horizon doesn't fully capture these.
- Small sample sizes — some strategies (Hook on MNQ, Open on MNQ) have ~20–30 trades over the 12-month sample (Jun 2025 — May 2026). Smaller samples carry wider confidence intervals. We adjust dropout to 20% for these to partially compensate.
Reproducibility
Raw trade lists for every strategy × instrument, the Monte Carlo model source data, and a sample of simulation paths are available on request for audit. Email support@puravidaedge.com with your specific request.