PRACTICAL SKILLS / PRACTICAL GUIDE
Decision Making Under Uncertainty
Good decisions under uncertainty do not require perfect prediction. They define possible outcomes, estimate ranges, limit downside and update when new evidence arrives.
In this guide
- The decision: real money, or another month of practice
- Separate the reversible move from the irreversible one
- Build three cases, not one forecast
- Size the first real-money step around the downside case
- Set update triggers before you start, not after
- What a review actually looks like
- Judge the decision separately from the outcome
- Common mistakes
- What to do next
- Sources and further reading
The decision: real money, or another month of practice
You've paper-traded for six weeks with a written rule set and a decent record. The question in front of you isn't "will this work" — nobody can answer that with certainty — it's narrower and more useful: how much real money should go live now, and what would tell you to pull it back? Framing it that way turns an unanswerable question into a manageable one.
Separate the reversible move from the irreversible one
| Move | Reversible? |
|---|---|
| Trading a small amount of real money while keeping the same rules | Reversible — you can stop, withdraw, and return to paper trading with a small, known loss. |
| Quitting a job to trade full-time on the strength of six weeks of paper results | Effectively irreversible — income, benefits and career momentum are hard to rebuild, and six weeks is a thin sample. |
| Increasing size sharply after one good week | Reversible in theory, but the temptation to keep increasing after each win is what turns a reversible step into a large, avoidable loss. |
Build three cases, not one forecast
- Base case: Real-money results track paper results reasonably closely, with a bit more friction from slippage, fees, and the psychological difference of risking actual money.
- Upside case: Results hold up and even improve slightly, because trading real, if still small, money sharpens focus in a way paper trading can't fully simulate.
- Downside case: Real-money results are noticeably worse than paper — a common pattern, since paper trading can't fully capture hesitation, fear of loss or rushed exits under real stakes.
None of these cases requires knowing the future. They require naming, in advance, what each one would look like — so that when results come in, you're checking them against a stated expectation instead of a feeling.
Size the first real-money step around the downside case
The amount that goes live first should be an amount you'd be comfortable losing entirely if the downside case is the one that shows up — not the amount you'd need if the base case holds. That single rule keeps a probably-fine decision from becoming a definitely-painful one if the six weeks of paper results turn out to have been partly luck.
Set update triggers before you start, not after
- A loss trigger: A specific, written drawdown — say, losing 20% of the real-money stake — that means stop and reassess, decided now, before any emotion is attached to a real number.
- A time trigger: A calendar date, such as four weeks or forty trades, whichever comes first, to compare real results against the base case, regardless of how any single day went.
- A behavior trigger: Catching yourself breaking a written rule twice in a week, which usually means the size is too large for your comfort at that moment, whatever the account balance says.
What a review actually looks like
| Question at the trigger date | What it tells you |
|---|---|
| Did results land closer to the base, upside or downside case? | Whether six weeks of paper trading was a reasonable preview or mostly luck. |
| Were the written rules followed, or improvised under pressure? | Whether a poor result reflects the strategy or the size being too large for comfort. |
| What's the smallest change that addresses the gap? | Usually reducing size or tightening one rule — not abandoning the whole approach after one rough stretch. |
Judge the decision separately from the outcome
If the downside case shows up despite a well-reasoned plan, that isn't proof the plan was wrong — a sound process can still produce a bad month, because chance never fully leaves the picture. The real test is whether the reasoning available at the time — the sizing, the trigger, the written rules — held up, not whether the following month happened to go well. Confusing the two is how a good decision gets abandoned, and a lucky one gets copied.
Common mistakes
- Treating six weeks as certainty: A short track record narrows the range of outcomes; it doesn't close it. Size accordingly, not confidently.
- Skipping the trigger and "seeing how it feels": A trigger decided in the moment is really just a decision made under whatever emotion happens to be present that day.
- Moving straight to the irreversible version: Quitting a job before testing with meaningfully sized real money skips the one step that would have actually told you something.
- Widening the downside case after a loss: Redefining "acceptable" after the fact isn't updating on evidence — it's moving the goalposts to match what already happened.
What to do next
- Write your own base, upside and downside case for one real decision you're actually facing this month, in one sentence each.
- Set a loss trigger, a time trigger and a size for the first reversible step — on paper, before you act, not after.
- Put a date on the calendar to review against the cases you wrote, and judge the process on that date, not just the balance.