Paper trading: the step between a backtest and real money

Paper trading — also called forward testing — is running a strategy live, against the real current market, with simulated money instead of real money. It sits in the gap between two things people over-trust: the backtest behind it and the live account in front of it. Skipping it is one of the most common and most expensive shortcuts in retail trading.

Why a backtest isn't enough

A backtest is a simulation against historical data. However careful it is, it makes assumptions — about fills, about latency, about data quality — and the strategy was, to some degree, shaped to fit that history. Paper trading removes those weaknesses in one move: it runs on a live market the strategy has never seen and cannot have been fitted to. Every bar is genuinely out-of-sample. It is the cleanest test of a strategy you can get without spending money.

What paper trading catches that a backtest misses

• Data and feed problems — gaps, delayed candles, or a symbol that behaves differently live than in your historical file.
• Execution reality — orders that don't fill at the price your backtest assumed, partial fills, and real spread at the moment of entry.
• Latency — the time between your rule firing and the order landing, which a backtest with end-of-bar fills quietly ignores.
• Plumbing bugs — an API misconfiguration, a timezone error, a rule that behaves differently on streaming data than on a static file.
• Overfitting, exposed — if a strategy looked brilliant in backtest and goes flat or negative on paper, you just learned that for free instead of for money.

How long to paper trade

Long enough to be evidence, not anecdote. A week is not enough — you would just be sampling one market mood. The honest rule is the same as for an out-of-sample window: paper trade until the strategy has taken enough trades for the results to mean something, ideally 30 or more, and ideally across more than one type of market condition. For a strategy that trades a few times a week, that is a couple of months. The discomfort of waiting is the price of not funding a fluke.

The psychological traps

Paper trading fails for human reasons more than technical ones. Three traps recur. First, the comparison trap: you watch the paper account and the live market diverge and conclude you are "missing out" — but you are not, you are buying information. Second, the impatience trap: a good first week tempts you to cut the test short and fund it; the first week proves nothing. Third, the discipline gap: paper trades feel consequence-free, so people tweak the strategy mid-test. Every tweak resets the clock — you are now testing a different strategy from bar one.

What 'passing' looks like

A strategy passes paper trading not when it makes money, but when its live behaviour matches its backtest. Similar trade frequency, similar win rate, similar drawdown profile. If the paper results roughly track the backtest, the simulation was honest and you can fund the strategy small. If they diverge sharply, the backtest was lying — and you want to know which assumption broke before, not after, real money is involved.

In Noon Barbari, paper trading runs the same rule set you backtested against live signals from your exchange, with no money at risk, so you can compare live behaviour to the backtest directly. It is the mandatory step between validation and a funded account — cheap insurance against an expensive surprise.