Monitor, Analyse, Iterate
    Deploy Paper Bot + Monitor + Iterate
    56 of 57

    Module 18 · Lesson 2 of 3

    Monitor, Analyse, Iterate

    16 min readUpdated

    Execute the first week of paper bot monitoring with rigorous data collection and zero manual overrides.

    Goal

    Run the paper bot for one full week with strict no-override discipline, collecting clean, unbiased performance data.

    Mistake This Prevents

    Overriding bot signals in Week 1 because "conditions don't feel right", and contaminating the data with personal judgement before you have evidence that your judgement improves results.

    Week 1: Zero-Override ProtocolDaySignalOverrideLoggedMondayBUY signal, 10:05NONE✓ YesTuesdayNo signalNONE✓ YesWednesdaySELL signal, 14:22NONE✓ YesThursdayBUY signal, 09:48NONE✓ YesFridayNo signalNONE✓ YesSaturdaySELL signal, 16:10NONE✓ YesSundayBUY signal, 11:30NONE✓ Yes0OverridesClean data= valid resultsWeek 1 goal
    Week 1 has one goal: zero overrides. Every signal gets logged exactly as the bot generates it. Concerns can be added as notes, but no overrides. Clean, unbiased data is the foundation for all future analysis.

    Week 1 Monitoring Protocol

    Daily Check-In (5 minutes, after daily candle close, approximately 5:30 AM IST)

    1. Open your paper trade log
    2. Did the bot signal fire today?
      • Yes: Verify conditions, log the trade
      • No: Note "No signal. Market condition: [condition]"
    3. Are there any open positions?
      • Yes: Check stop and TP levels. Note current price vs. both levels.
      • No: Nothing to do.
    4. Update the running P&L (paper) column

    The No-Override Rule This week: EVERY signal the bot generates is logged as a paper trade. No exceptions for:

    • "The market feels uncertain today"
    • "Funding rate is high" (this is a valid concern, but note it in the log, don't override)
    • "A better setup might come later"

    Note: You CAN log concerns alongside the trade ("logged but note: funding +0.04%, higher risk"). These notes become valuable data.

    End-of-Week Data Capture By Sunday, record:

    • Total signals generated: ___
    • Total paper trades logged: ___
    • Signals overridden (Week 1 should be 0): ___
    • Open positions at week end: ___
    • Week P&L (paper, in R multiples): ___
    • Market condition this week: [Trending / Ranging / other]

    What You Are Learning in Week 1

    • Signal frequency: How often does PB1 fire in current conditions?
    • Signal quality: Do the conditions all hold at close or do alerts sometimes fire incorrectly?
    • Execution mechanics: Is the log easy to update? Any process friction to fix?
    Log Every Day, Including No-Signal DaysMonSIGNAL: BUYPrice: $42,100Size: 0.5 BTCTime: 10:05✓ LoggedTueNO SIGNALTODAYConditions not met✓ LoggedWed?Not logged✗ GAPThuSIGNAL: SELLPrice: $43,800Size: 0.5 BTCTime: 14:22✓ LoggedFriNO SIGNALTODAYConditions not met✓ LoggedWednesday gap = cannot determine if strategy had a signal or notGaps in data make pattern analysis impossible: log every day, signal or no signal
    Logging only trade days creates gaps that make pattern analysis impossible. A missing day could hide a false signal, a missed entry, or a quiet period. Log every day, even when the bot generates no signal.

    Checklist

    • Daily 5-minute check-in completed every trading day this week
    • Zero overrides (all signals logged as paper trades)
    • End-of-week data table completed
    • At least one "no signal" day documented

    Common Mistakes

    • Checking alerts only when price moves significantly (you miss signal-trigger moments)
    • Logging trades as "would have taken" rather than "alert fired at [time]"
    • Not logging "no signal" days (creates gaps in the data record)

    Fedha Tool Tie-in

    FedhaAcademy's Monitoring Dashboard shows the daily signal check status, open positions, and running P&L for your paper bot. A daily 5-minute prompt is sent at 5:30 AM IST (after candle close) with a checklist to complete.

    Key takeaways

    What to remember

    1. Week 1 requires zero overrides, because clean, unbiased data is the goal.
    2. Log signals whether you agree with them or not, including your concerns as notes alongside the trade.
    3. Log 'no signal' days explicitly, because gaps in the record make pattern analysis impossible.
    4. Signal frequency reveals whether the strategy trades too often or too rarely for current conditions.
    Lesson quiz

    Quiz

    Question 1 of 9

    Pass: 7 of 9 correct

    Question 1 of 9

    During Week 1 monitoring, what should you do when the bot signals a trade but you disagree with it?

    Keep your place

    Save this lesson and continue tomorrow.

    Create an account to mark lessons complete, build a daily practice streak, and use paper trading with $10,000 in virtual capital.

    Create free account I already have an account
    Free forever No card required Continue from any device

    Up Next

    Scaling to Live and Keeping It Honest

    Continue