realshyt proof data pack: https://realshyt.com/proof/umar-punjabi-nasdaq-strategy-backtest

What this is: our backtest of the Nasdaq "initial balance" (first 30 minutes) breakout strategy exactly as
Umar Punjabi described it on Figuring Out with Raj Shamani, FO414 (27 Sep 2025, 26:48-29:12).
Our test, our method, our opinion. Not financial advice.

Files
  trades_*.csv     every trade, one row per trading day it fired (one file per version of his rules)
  yearly.csv       win rate, full-target hits and total R per year, per version
  rules.json       the exact rules and costs used (written before the first full run, never tuned)
  results.json     every number on the proof page, straight from the backtest
  METHOD.md        full method, all 18 versions, sanity checks and limitations
  code/            the Python used (ib_data.py downloads the data, ib_backtest.py runs the test)

Columns in trades_*.csv (times are New York time, prices are Nasdaq-100 index points)
  date             trading day
  direction        long (first 30-minute high broke first) or short (low broke first)
  first30_high/low high and low of 09:30-09:59, the "initial balance"
  open_0930        the 09:30 open
  entry_time_ny / entry_price   when and where the breakout order filled (spread + slippage included)
  stop / target    stop loss and the 1:2 target
  exit_time_ny / exit_price     when and where the trade closed
  exit_reason      target = full 1:2 hit, stop = stop loss hit, eod = closed at the end of the day
  risk_points      entry to stop distance
  result_R         profit or loss in units of risk after costs (+2 = full target, about -1 = stop)

Data: Dukascopy Bank historical datafeed, USATECHIDXUSD (NASDAQ-100 CFD), real 1-minute BID and ASK,
2 Jan 2015 - 30 Sep 2026, 2,953 trading days. Check any row against your own 1-minute Nasdaq chart.
Found a mistake? Tell us on Instagram @realshyt__ and we will correct it publicly.
