5-year backtest

Pattern Lab: how the index pattern changes, and where the traps are

Five years of daily candles and four and a half years of 5-minute candles, run through the same signal and exit rules as the trade151 auto-trader.

01 Β· Market regime

The pattern changes about every 2–3 weeks

Each trading day is labelled from its own close: Uptrend or Downtrend (ADX 23+ with price, 20-day and 50-day averages lined up), Range (ADX under 18), Volatile (daily range in the top 15% of the past year, or India VIX 22+), otherwise Transition. A new regime counts only after it holds for 3 days in a row.

Each box is one month: the regime that ruled most days, the month's % change, and a corner count of how many times the regime switched inside that month. Hover a box for the details.

How long each regime lasts

What usually comes next

When a regime ends, this is where the market went next (share of all switches out of that regime).

02 Β· Month by month

Does last month's or last year's pattern repeat?

Calendar-month returns by year

Teal months closed higher, red months closed lower; darker means a bigger move. The right-hand columns show how many of the years were positive, and the average.

Latest regime switches

03 Β· Time of day

When the day's pattern changes

When the day's high is made

When the day's low is made

When the direction turns

Days that made one extreme and then travelled at least 60% of the day's range the other way. Bar = time of that first extreme.

Whipsaw by time

How often the 5-minute Supertrend flips in each 15-minute slot, per day. Higher = more fake signals.

Is the turn time shifting? Share of turn days by time window, per year

04 Β· Traps

Which breakouts trap you, and when

A failed break closes back inside the level within 30 minutes before moving a meaningful distance. A full trap is a failed break that then breaks the opposite side the same day, catching buyers and then sellers. The opening range is the first 15 minutes; previous-day levels are yesterday's high and low.

Opening-range breaks by the time they happen

When full traps spring

Time the trap completed (price broke the opposite side), opening range and previous-day levels together.

Trap rate year by year, and by day type

05 Β· Your auto-trade signal

The 151 signal over 4Β½ years

Win rate by entry time

The dashed line is the win rate these exits need to break even.

Year by year

By market regime (prior day)

By opening gap, expiry and VIX

Filters tested: does skipping a condition help?

Each rule was chosen on 2022–2024 and then checked on 2025–2026, which it never saw. Average result is net % of premium per trade, after 1% brokerage. A filter is only worth keeping if both columns improve.

Other entry setups on the same data

Same option-price model and costs. "151 exits" = +6% / βˆ’8.6% with the app's trailing rules. "Wide 1:2" = βˆ’20% stop, +40% target, break-even after +20%, max 2 trades a day. The last column is the raw direction hit rate 30 minutes after the signal; 50% is a coin flip.

06 Β· Apply in trade151

How to escape the traps in auto-trade

Trap Guard settings for the auto-trader

These are the rule values backed by the tables above. Copy them into the auto-trade settings, or ask for Trap Guard to be built into the server so it applies to every user.


      
07 Β· Method

How these numbers were made