Knowing what each bias is doesn't stop it from happening. That's the painful lesson every trader learns the hard way. You read about overconfidence, you nod, you agree — and a week later you're scaling into a position because "this time the setup is obvious." The knowing and the doing are different muscles.
So this chapter is the doing muscle. Four checklists — pre-trade, post-trade, weekly, and "when the red light starts flashing" — each just a few questions. Read slow, answer honestly, move on. If you can run through these every time, you'll be in the top 5% of retail traders by decision discipline alone. Not because the questions are hard, but because almost nobody actually asks them.
Checklist #1 — Before Every Trade (60 seconds, every time)
Run these six before you click BUY or SELL. If you can't answer yes to all six, the trade isn't ready.
Pre-trade: Before committing capital, verify you haven't fallen into at least one avoidable trap. Six questions. One minute. Every time.
-
Is this trade because a rule fired — or because it "feels right"? → If it's "feels right," you are the hero of a story you're about to lose.
-
Would I still take this trade if my bot said the opposite? → If no, you're overriding a system you built to protect yourself from you.
-
What specific evidence would make me close this trade at a loss? → If you can't name it in advance, you will rationalize holding forever.
-
Am I sizing this based on my strategy's max drawdown — or on my excitement? → If the position would hurt more than 2x my strategy's tested max DD, I'm overconfident.
-
Have I checked: is today an FOMO day? → Signs: BTC just ran +10% in a day, Twitter is euphoric, a friend called you. Historical base rate of FOMO-day entries: usually the local top.
-
Is this the same setup I've taken the last 5 times? What happened those 5 times? → If you don't know, check the log. If you know and the answer is "lost 4 out of 5" — you're in a confirmation-bias loop.
Checklist #2 — After Every Closed Trade (2 minutes)
These five questions force you to attribute outcome to cause honestly, so next time's decision improves.
Remember: Outcome ≠ decision quality. A winning trade from a bad decision trains you to make more bad decisions. A losing trade from a good decision is the price of an edge. Separate them explicitly.
-
Did I follow my system, or did I override it?
-
What was the outcome? (Win / loss / breakeven)
-
Was the outcome consistent with the decision quality, or was I lucky/unlucky? → Four cells: Good decision + Win. Good decision + Loss. Bad decision + Win. Bad decision + Loss. Only the top-left is to be replicated.
-
What would I do differently if this exact setup appeared tomorrow?
-
Did any bias show up? Which one? (FOMO / Loss Aversion / Confirmation / Anchoring / Overconfidence / Sunk-Cost / Broker Signal / Other) → Tally these weekly. Your worst bias is the one that shows up three times in ten trades.
Checklist #3 — Weekly Review (Sunday evening, 10 minutes)
Zoom out. Patterns invisible on a single trade become loud over a week.
Bottom line: The trader who reviews weekly outperforms the trader with the "better" system who never reviews. Pattern recognition on your own behaviour is the single highest-ROI activity in trading.
-
What percentage of my trades followed my written rules this week? → Target: 100%. Reality for most: 60-80%. The gap is where the money is made or lost.
-
Which bias showed up most often? (Count from Checklist #2 tallies) → Then: write one sentence about how you'll catch it earlier next week.
-
Did any of my "gut calls" outperform my system? → If yes: was it skill or luck? Apply the Checklist-#2 four-cells test over 10+ trades before updating the system.
-
Is my system still being tested live — or has it accumulated 30+ trades? → Under 30 trades = still evidence-gathering. Don't scale yet. Over 30 trades with worse-than-backtest performance = the backtest was optimistic, investigate.
-
Did I add or remove any rule based on this week's drama alone? → If yes: revert. Recency bias is the #1 system-destroyer. Wait 30 trades before any rule change.
-
Am I paying attention to broker signals (push notifications / in-app suggestions)? → Those signals go out to millions simultaneously. By definition, you cannot be early. Mute them.
Checklist #4 — When the Red Light Flashes
The four conditions below mean your judgement is compromised right now. When any of them are true: no trades. Step away from the screen.
The setup: Every major retail loss I've seen happened in one of these four mental states. If you recognise any, the correct action is the same: close the app, do something physical, come back in 6 hours.
-
You're in a drawdown streak and thinking "I need to make it back." → Revenge-trading is how 40% drawdowns become 70% drawdowns. Step away.
-
You're on a 3+ winning streak and thinking "I've got this figured out." → Overconfidence peak. Historical base rate: the 4th trade after 3 wins is the worst-performing trade in the series.
-
You haven't slept well, you had a fight, you drank, or it's past 11pm your time. → Impulsivity is up, prefrontal cortex is down. Your rules still apply — but your adherence won't. Don't trade.
-
You're refreshing the chart or TradingView more than once every 30 minutes. → You're looking for an excuse to trade. There isn't one. Close the app.
Checklist #5 — Before You Ship a New Backtest Result
If you're building a strategy, not just following one, these six questions save you from the 10 backtest traps covered in Part I of this guide. If you can't answer yes to all six, your backtest is not ready for live capital.
Verdict: 195 of every 200 strategies that look profitable in backtest fail live. The questions below filter out roughly 180 of those 195 before you touch real money.
-
Did I check robustness in a neighbourhood of parameter values, or only the single "best" one? → See Parameter Robustness. Plateau = keep. Spike = reject.
-
Does the strategy still beat HODL in all three walk-forward sub-periods? → If it only beats in 1 of 3, it's regime-dependent. See Benchmark Tunnel Vision.
-
Would the strategy still work if I removed my 5 worst-looking trades? → If no: a handful of lucky trades are carrying it. That luck won't repeat live.
-
Did I use data the strategy wouldn't have seen in real-time? (look-ahead bias) → Every close, every rolling indicator, every "adjusted" price should be checked. See Look-Ahead Bias.
-
How many strategy variants did I test before picking this one? → If >10, your "winner" might just be p-hacked. See P-Hacking.
-
Did I survivorship-filter my data? (only using coins/stocks that still exist today) → Dead coins have useful lessons. Missing data is one-sided lying. See Survivorship Bias.
Remember: The goal of this checklist isn't paranoia — it's humility. Five of the smartest traders I know ran these checks before deploying real capital. The other five didn't. The first group compounds. The second group posts on /r/wallstreetbets.
How to actually use this
Print this page. Put it next to your keyboard. Or save the PDF on your phone and open it before every trade.
If that's too much friction: just commit to one checklist per week. Start with #1 (pre-trade). Get it automatic in 30 days. Then add #2. Then #3. After 90 days you'll have a decision discipline most professional traders envy — and most never built.
Biases don't disappear when you learn about them. They disappear when you build the habit of checking for them, one question at a time.
That's the whole guide. 17 biases. 4 checklists. 24 questions.
If this guide saves you from even ONE bad decision, the hours I spent writing it were worth it. If it saves you from a full drawdown, buy me a coffee and write back — your story is the best possible proof that this stuff works.
— Dominic
