The public track record

Our accuracy ledger

The monthly ledger builds day by day — every call is graded after a full trading month, and the final rate will appear automatically once the sample is large enough. No invented numbers here, ever — and the day's actual accountability is right below.

51.6% — our directional hit rate across 2,569 graded calls on 279 distinct markets from 5 asset classes. Every call is graded 21 trading bars after it was issued.

Read it against its own baseline. Buying and holding the same markets over the same windows scored 62.7% on this very record, so our 51.6% sits 11.1 points below doing nothing at all. We print the gap because a hit rate without its baseline is not a result.

Measured 2026-09-11. Past performance does not guarantee future results.

Confidence tiers on the full record

How to read that fourth column. "Confirmed call" holds 1,009 bullish calls and 0 bearish ones; "Lean" holds 0 bullish and 0 bearish. When every call in a tier points up, its hit rate must be compared with how often those same markets went up — not with zero, and not with a coin flip. That is what the fourth column is, and the number in brackets is the whole edge: 0.0 points for Confirmed and — for Lean.

Each card carries its tier on the day it is issued and is graded on that tier, never on a later rule. 2,569 calls older than the rule were tiered from the record itself; the independent test of the tiers starts with the first card sealed with its tier.

Rows that dropped out of grading: 352. They are not inside the counts above. Why they dropped: 270 because the price source changed between the call day and the grading day, 55 with no matching bar at the grading horizon, and 27 ambiguous symbols. Every one of those names is listed in the scoreboard file we publish.

The MM V-Shape indicator on MetaTrader — a separate product from the daily report

These trades do not come from the daily directional calls you subscribe to. They come from the signal log of our MM V-Shape indicator running on MetaTrader on shorter timeframes, each with a stop and a target. We publish it here because it answers one question the tier table cannot: a hit rate under 50% can still pay when the average win is larger than the average loss.

446 closed trades: 44% winners · average reward-to-risk 2.41 · net +93R · +0.21R per trade, after deducting spread and slippage. The profit comes from the average win being larger than the average loss, not from the hit rate.

Today: we said — and this is what happened

(Arabic report sample) Updated every day: the day's calls next to what the market actually did — misses are published exactly like hits.

How do we work?

We analyse every market with several quantitative tools (time cycles, historical analogues, the seasonal pattern, and multi-horizon momentum) — but only the tools that pass our strict test get a vote on the statistical lean; the rest is written context in the card. Every reading is logged the day it is issued and graded after a trading month — nothing deleted, nothing dressed up.