PMTS Weekly Performance Review: May 14–21, 2026 — 1,258 Trades, 65.74% Win Rate, USD 1.19M Net Profit on XAUUSD

The week of May 14–21, 2026 closed with PMTS delivering institutional-grade results across our XAUUSD-focused AI trading book. The system executed 1,258 trades with a 65.74% win rate, generating USD 1,189,671.15 in aggregate net profit across our multi-broker, multi-account architecture. This review documents the week, contextualizes it inside the rolling 30-day performance window, and explains what the underlying execution data tells us about the robustness of the strategy stack heading into the second half of Q2 2026.

Headline Numbers: Week of May 14–21, 2026

Performance figures below are aggregated across all PMTS production accounts and synced directly from MetaTrader 5 server logs. They are not backtest results, and they are not adjusted for any prospective allocator's projected fee schedule.

• Reporting period: 2026-05-14 to 2026-05-21 (7 calendar days)
• Total closed trades: 1,258
• Winning trades: 827
• Losing trades: 166
• Win rate: 65.74%
• Aggregate net profit: USD 1,189,671.15
• Primary instrument: XAUUSD (Spot Gold) on MT5
• Execution venues: 19 live accounts across 7 brokers

The week was driven by a pronounced rotation between range-bound conditions in the first half and directional momentum once the post-FOMC volatility profile compressed. The PMTS multi-regime architecture — which routes signals between mean-reversion, breakout and trend-following modules — captured both phases without manual intervention.

Rolling 30-Day Context: April 21 – May 21, 2026

A single week, no matter how strong, is statistically thin. To frame the May 14–21 result honestly, we benchmark it against the rolling 30-day window — a sample size that begins to suppress single-event noise without yet being large enough to obscure regime shifts.

• Reporting period: 2026-04-21 to 2026-05-21
• Total closed trades: 5,400
• Winning trades: 3,150
• Losing trades: 1,039
• Win rate: 58.33%
• Aggregate net profit: USD 3,048,736.92

The 30-day win rate of 58.33% against a 7-day rate of 65.74% is consistent with what we expect from a multi-regime gold trading system: the headline weekly figure outperformed the monthly average, but stayed inside the statistical envelope rather than collapsing into a single outlier print. That is what disciplined allocators look for — not the highest week, but the absence of regime-induced breakdown.

What the Profit Factor Tells Us

For the May reporting block on the master allocation account, the system recorded a profit factor of 2.5793 — meaning every USD 1.00 of gross loss was paired with USD 2.58 of gross profit. Institutional allocators tracking algorithmic gold systems typically classify profit factors above 1.75 as production-grade. PMTS continues to trade above that threshold on its primary book, and that consistency — not the peak figure — is the metric that drives allocation decisions at the desk level.

Trade Distribution and Behavior

Beyond the headline win rate, the texture of the trade distribution determines whether a system is repeatable. Three points stand out from the May 14–21 deal log:

1. Concentration in XAUUSD

The vast majority of executed volume continued to route through XAUUSD on MT5. Gold remains the cleanest beneficiary of the current macro mix: a Fed in extended pause, sticky core inflation, sustained central bank reserve accumulation, and elevated geopolitical risk premium. PMTS does not chase symbol diversification for its own sake — it concentrates capital where the AI signal-to-noise ratio is highest. Diversification is delivered at the broker and account layer, not by spreading the strategy across instruments where the model has weaker edge.

2. Asymmetric Loss Profile

The week recorded 166 losing trades against 827 winners, but more importantly, the average loss size remained tightly controlled. Position-level stop-loss logic, capped at pre-trade risk budgets, prevented any single losing position from materially impairing the equity curve. This is the institutional risk discipline that separates production-grade automated trading from retail bot behavior: small, frequent, bounded losses paid in exchange for the right to keep capturing the asymmetric winners that anchor the profit factor.

3. Multi-Broker Execution Quality

Trades were distributed across 19 accounts on 7 brokers — including MultiBank Group, FTMO, DarwinexZero, MEX Atlantic Corporation and MetaQuotes Ltd. — with the MAM (Multi-Account Manager) layer handling proportional allocation. This multi-broker topology is not cosmetic: it materially reduces single-counterparty execution risk and smooths the realized fill profile when liquidity tightens around macro releases. When one broker's bid-ask widens, the system is not forced into adverse execution because the same signal can be routed through deeper venues.

Risk Metrics in Context

Win rate and net profit alone do not describe a strategy. The risk-adjusted ratios institutional allocators actually care about — Sharpe, Sortino and Calmar — require longer realized track records to stabilize. The PMTS dashboard publishes these ratios as they mature, and we explicitly avoid quoting unstable intermediate values that would mislead allocators evaluating the system on a single reporting block.

What we can say with confidence about the May 14–21 block:

• Realized maximum drawdown on the master allocation account remained well inside the system's hard risk budget.
• No margin events, position blow-ups or counterparty incidents were recorded across the 19 live accounts.
• Equity curve continuity was preserved through the week — no overnight gaps and no forced de-leveraging.
• Execution slippage stayed inside expected bands across all 7 brokers, including during the New York close.

Why This Week Matters for the Strategy Stack

One of the most common misconceptions about algorithmic trading is that a strong week is evidence of a strong system. It is not — strong weeks happen for weak systems too, and weak weeks happen for strong systems. What this week does tell us is something narrower but more useful: the PMTS multi-regime router continued to operate inside its design envelope while macro conditions shifted intra-week.

The week tested three specific failure modes that historically have broken retail-grade gold bots:

• Whipsaw conditions around mid-week consolidation — handled by the mean-reversion module without forcing the trend module into adverse selection.
• Headline-driven volatility expansion on policy commentary — absorbed without breaching per-trade risk caps.
• Liquidity thinning at the New York / Asia handover — execution slippage stayed inside expected bands across all 7 brokers.

None of those failure modes triggered an intervention. The system handled the week the way it was designed to, which is the actual product allocators are paying for.

How to Track These Numbers Live

All performance figures cited above are produced by the same data pipeline that feeds the investor dashboard. Investors with active PMTS allocations can monitor their share of these results in real time, with per-trade drill-down and intraday equity curve, from the PMTS investor dashboard. The dashboard refreshes directly from MT5 server logs — there is no intermediate manual reporting layer, and no figure is editable from the back end.

Prospective allocators who want to evaluate the system before committing capital can open a read-only investor account to observe the live track record, broker spread, and per-account performance attribution for the next reporting block.

Looking Ahead to the Week of May 21–28, 2026

The macro calendar into the next reporting block is moderately heavy, with mid-tier inflation prints and Fed speakers scheduled. The PMTS volatility-regime detector will route capital weight between modules accordingly — no manual rebalancing required. We will publish the next weekly performance review at the close of trading on May 28, 2026, using the same methodology described in this report.

For institutional readers requesting deeper attribution — by module, by broker, by intraday session — the data is available on request through the platform's allocator reporting suite.

Past performance does not guarantee future results. Trading involves substantial risk of loss and is not suitable for every investor. The figures published in this report are realized results from PMTS production accounts as of May 21, 2026, and may be revised as broker settlement data is finalized.