The difference is not the data — both use live prices. The difference is that a live order has to actually find a counterparty, pass a margin check, and survive your own nerves. Those three things are where paper and live results diverge.
The four gaps
1. Slippage
A simulated fill happens at the quoted price. A real fill happens at whatever the book offers when your order arrives. On liquid at-the-money index options the gap is small. On far strikes, in the first minute of the session, or during a volatility spike, it is not.
This compounds on multi-leg strategies. Four legs means four spreads crossed on entry and four on exit — eight chances for the real price to be worse than the quoted one.
2. Margin
Paper trading never checks whether you can afford the position. Live trading checks every time, and option selling margins move with volatility. A strategy that paper trades perfectly can be rejected live for want of funds on exactly the day it would have worked.
3. Rejections
Simulated orders do not get rejected. Real ones do — for margin, session, IP, freeze quantity and order type. A multi-leg strategy that loses one leg to a rejection is holding a different position than the one you designed.
4. Psychology
This is the one traders discount and then discover. A 15% drawdown on paper is a number. The same drawdown on real capital produces an urge to intervene, and intervening in a systematic strategy is usually how it stops being systematic.
The psychological gap, concretely
It is easy to dismiss this as soft, so here is what it looks like in practice.
On paper, a strategy that is down four days running is a data point. You note it, you check the logs, you carry on, because the number on the screen is notional and your attention is elsewhere.
Live, the same four days produce a specific sequence: you check more often than you need to, you start reading the drawdown as evidence the strategy has stopped working, and on the fourth evening you consider skipping tomorrow. Skipping tomorrow is how a systematic strategy becomes a discretionary one, and the day you skip is disproportionately likely to be the recovery.
What actually helps
The defence is deciding in advance, in writing, what would make you stop — and treating anything short of that as noise you have already agreed to sit through.
A checklist worth using
Switch when you can honestly tick all of these, not when you are bored of paper trading:
- The strategy has run on paper across at least two expiry cycles
- It has seen a trending day, a range-bound day and a gap open
- You have read the logs on a day it chose not to trade, and agreed with the decision
- You know the worst drawdown it produced and would accept that in rupees
- You have the margin for the position at a volatility level higher than today's
- Your broker, dedicated IP and daily login routine are all working
- You have decided in advance what would make you stop it
Signals that you are not ready yet
The checklist above says when to go. These are the signs that you should not, however good the paper results look.
- You have changed the strategy during the paper period. You have tested several things briefly, not one thing properly.
- You cannot describe what it does on a quiet day. You have been reading P&L rather than behaviour.
- You do not know the worst drawdown in rupees. The number you have not calculated is the one that will surprise you.
- You have missed the morning routine more than once. Live, each miss is a real cost.
- You are going live because the paper period felt long enough. Time served is not evidence.
None of these are permanent disqualifications. They are all addressable in another few weeks of paper trading, which is a cheaper fix than discovering them with capital committed.
Go live small
Going live is not a switch from zero risk to full size. The sane path is minimum size first — one lot — for a few weeks, which surfaces the real slippage and the real margin behaviour at a cost you will not mind paying for the information.
Compare that live period against the paper period for the same days. The gap between them is your actual execution cost, and it is a number worth knowing before you scale.
| Phase | Size | What you are learning |
|---|---|---|
| Backtest | None | Whether the idea ever worked |
| Paper | Notional | Whether the logic and parameters are sane |
| Live, minimum | One lot | Your real slippage and operational reliability |
| Live, scaled | Your intended size | Whether the edge survives at size |
Each phase answers a question the previous one could not. Skipping a phase does not save time — it just moves the discovery to a more expensive place.
Scaling up
Once the minimum-size period has given you a real execution cost, scaling is an arithmetic question rather than a confidence question.
Take the drawdown you measured, scale it to your intended size, and check it against the rupee figure you said you would accept. If it breaches that, the strategy is not ready for that size — regardless of how the P&L has looked.
What changes as size grows
Watch slippage as you scale. A strategy that filled cleanly at one lot can fill noticeably worse at ten on the same strike, and that cost comes straight out of the edge.
What stays the same
Both modes run on the same engine and the same rules, so a strategy does not need rebuilding to go live. The deployment mode changes; the strategy does not.
What does change is the setup around it: live trading needs a connected broker, a dedicated IP and a daily broker login. Paper trading needs none of them.
The first live month, specifically
The transition is where most of the surprises live, so it is worth having a plan for the first month rather than improvising it.
1
Week one — one lot, watch everything
Read the logs daily. You are checking that the operational routine holds, not that the strategy makes money.
2
Week two — compare fills
For each entry, compare the price you got against the price at that moment. This is your slippage, measured rather than assumed.
3
Week three — check margin behaviour
Note the highest margin the position required, not the average. That peak is what your account has to carry.
4
Week four — compare against paper
Same strategy, same dates. The difference is your real execution cost.
At the end of that month you have three things you did not have before: a measured slippage figure, a peak margin requirement, and proof that you can run the morning routine consistently. Only then is scaling an arithmetic decision rather than a hopeful one.
Going back to paper
Switching back is a legitimate move, not an admission of failure. If a strategy starts behaving in a way you do not understand, running it on paper alongside is a cheap way to observe it without funding the confusion.
Remember that both deployments count towards your simultaneous-strategy limit, so running a paper copy alongside a live one uses two of your three slots.
The short version
The switch to live is a step down in certainty, not a step up in confidence.
- Slippage, margin, rejections and psychology are what change
- Go live at one lot for a few weeks before scaling
- Compare live against paper for the same days to measure your real execution cost
- Decide your stop condition in rupees while you are calm
- Scaling is arithmetic once you have a measured drawdown, not a confidence decision
Switching back to paper is a legitimate move. If a strategy starts behaving in a way you do not understand, observing it on paper is cheaper than funding the confusion.
Frequently asked questions
You can describe what the strategy does on a quiet day, you know its worst drawdown in rupees, you have run the morning routine reliably, and you have written down what would make you stop.
Only in the sense that you learn nothing new after the strategy has seen the conditions it was built for. Beyond that point, minimum-size live trading teaches you more.
It is a sensible way to form your own view. The published backtest tells you what the analyst measured; a month on paper tells you how it behaves now and whether you could sit through its drawdowns.
No, but it answers a narrower question than people assume. It proves the logic works and the parameters are sane. It does not prove the strategy survives real fills.
It varies with liquidity and size, and no honest figure applies to every strategy. Treat paper results as a best case and find your own gap by going live at minimum size.
No. The same strategy runs in both modes on the same engine. Only the deployment mode changes.
The minimum the strategy allows, usually one lot, for a few weeks. The point of that period is information, not profit.
Yes, though both deployments count towards your simultaneous-strategy limit.
After a minimum-size live period has given you a measured execution cost. Scale the observed drawdown to the new size and check it against the rupee loss you already agreed to accept.
No. If a strategy is behaving in a way you do not understand, observing it on paper is cheaper than funding the confusion.
Start with a free 3-day trial
Build a strategy, backtest it and run it on paper — no broker, no IP and no money needed to try it.
