Skip to content
Prosper Osaigbovo

The Trader's Hindsight

A trading journal that shows forex and prop-firm traders, in money, what their habits cost them. I designed, built and run it on my own1. It is a live subscription product2.

Role
Founder and only engineer
Built
Dec 2025 to now
Price
From $12 a month
Stack
Next.js 16, React 19, TypeScript, Supabase (PostgreSQL), Node.js

The question it answers

A journal shows what happened. It does not say what each habit cost, or whether the month would have been positive without it. The Trader's Hindsight answers that.

What it does

The Trader's Hindsight dashboard for June 2026: net P&L of minus $568.99, equity of $70,601.31, a 33% win rate, trade counts and prop-firm challenge status.(opens the screenshot at full size)
A. The dashboard. Monthly P&L, equity and challenge status per firm. Fixture data, not a real trader.
Hindsight
Replays a month as if each costly habit never happened, then ranks the habits by the dollars they cost.3
Foresight
A read-only co-pilot. It sends a risk check to Telegram as each trade opens, then grades its own warnings A to F against how the trades closed.4
Firm Fit
Replays a trader's real trades against each prop firm's rule book to estimate the chance of passing.5
Rule tracking
Tracks the rules of any prop firm, with ready-made presets for FTMO, FundingPips, FundedNext, The5ers and Alpha Capital. It disconnects an account automatically when it breaches one.
Insight
An AI coach on the Anthropic Claude API that reads a trader's full history.
Trade sync and import
Read-only sync from cTrader (Open API) and MetaTrader, and CSV or Excel import from five platforms.
Billing
Cards through Flutterwave and crypto through NOWPayments, both confirmed by webhooks.
Foresight reads: a scorecard of each warning's record and net result, then a trade read at entry with the warning it raised.(opens the screenshot at full size)
B. Foresight reads. Each warning is scored against how the trade it warned about closed. Test account.
Daily equity curve from a $10,000 starting balance, 3 to 16 August, ending near $12,400.(opens the screenshot at full size)
C. Equity curve. Daily balance from a $10,000 start. Test account.

The core idea, working

Hindsight replays a trader's month without each costly habit and ranks the habits by the dollars they cost3. The panel below shows the idea on made-up trades that come labelled with their habits. Take a habit out to see the month without it.

What did these habits cost this month?

Example data: 15 made-up trades.

June net P&L

−$610

Take a habit out to replay the month without it.Habits are ranked by what each one cost.

Trades by day of June. Letters mark the habits behind a trade; dashed outlines are trades taken out.

  • Revenge trades after a losscost $670
  • Sizing up after a losscost $500
  • Tiltcost $440

Each habit's cost is worked out on its own. With two taken out, a trade that matches both is taken out once.

See the 15 trades
DayPairHabitsP&LReplayed
2EURUSDNone+$180+$180
3GBPJPYNone−$150−$150
3GBPJPYRevenge trades after a loss−$220−$220
3EURUSDRevenge trades after a loss, Sizing up after a loss, Tilt−$260−$260
5XAUUSDNone+$240+$240
8EURUSDNone−$130−$130
8EURUSDNone+$90+$90
10USDJPYNone+$210+$210
11GBPUSDNone−$160−$160
11GBPUSDRevenge trades after a loss−$190−$190
11XAUUSDTilt−$180−$180
15EURUSDNone+$150+$150
17GBPJPYNone−$140−$140
17EURUSDSizing up after a loss−$240−$240
22USDJPYNone+$190+$190

Hindsight on the public landing page: the costliest habits of the last 30 days, ranked by the dollars they cost, led by revenge trades after a loss at minus $1,687.(opens the screenshot at full size)
D. Hindsight. Habits ranked by what they cost. From the public landing page.

Firm Fit: the odds before paying for a challenge

Firm Fit estimates a trader's chance of passing each prop firm's challenge from their own trading history, before they pay for one. On the public page, the statement is read in the browser and never uploaded5.

Security decisions

Traders connect real broker accounts, so I designed the app RLS-first: the database decides who can read a row, not the API route. My background is in security: an M.Sc. in Cyber Security and over a year as a security specialist.

Each user sees only their own rows
PostgreSQL row-level security decides what every user can read and write, so the API is not the only line of defence.
A password is not enough
Multi-factor sign-in is enforced in the database itself, with recovery codes, so a lost phone does not lock a trader out.
Broker access is read-only
Connections cannot place, change or close a trade, and cannot move money.6
Broker tokens are sealed
Tokens are encrypted before they are stored, so a copy of the database does not hand over broker access.
Statements stay on the trader's device
Firm Fit reads statements in the browser. They are never uploaded.
Checks run without me
Secret scanning on every CI run, a scheduled dependency audit, and error monitoring in production.

Tested, then shipped by the pipeline

The repository has 279 automated test files, about 50,500 lines7. A push to main puts the web app live only after these steps pass, in this order.

  1. Secret scanEvery tracked file
  2. LintIncluding rules that keep server code out of the browser
  3. Type check
  4. TestsEvery test file, on every push
  5. Production build
  6. DeployThe web app, from main only

A weekly audit reports dependencies with known vulnerabilities.8

The numbers, and where they come from

Every figure about the product, with its source. The repository figures are counted by a script from the private code.

Repository figures counted on 27 Sep 2026, at commit 4f4a08e.
WhatFigureSource
Lines of TypeScript, not counting tests116,641Private repository
Pages38Private repository
API routes97Private repository
Automated test files279Private repository
Lines of tests50,465Private repository
Commits since December 20251,387Private repository
Price$12 a monthtradershindsight.com

See how it is built, on a call

The code is private, and so is the product's logic. On a call I can show you how it is built: the architecture, the tests and the delivery pipeline.