TradingView's Paper Trading account is the most-used simulator in retail trading, and it is good at the thing simulators are usually bad at: it fills your orders at the real bid and ask, from the same live feed as the chart, with the same order ticket you would use on a connected broker. What it cannot do is make you treat the result as if it mattered — and that, not the fill logic, is why most paper-trading records say nothing about how the same person will trade real money.
The fix is not a better simulator. It is a review discipline that removes the three ways paper results flatter you: the account that resets, the commissions that are off by default, and the trades nobody writes down. This guide sets TradingView's simulator up so it behaves like a funded account, puts numbers on the friction it leaves out, and shows how to journal each paper trade so that after fifty of them you know something — including whether to go live at all.
What the simulator modelsHonest about the spread, silent about the rest
TradingView's own support pages describe how Paper Trading fills: a market buy executes at or near the current ask, a market sell at or near the bid, and a limit or stop is triggered by the bid or ask rather than by the mid-price line you see on the chart. That is more honest than many demo accounts, which fill at the last trade. The rest of the list is what you have to add yourself.
What TradingView simulates
Real-time bid/ask fills from the live feed; market, limit, stop and stop-limit orders; take-profit and stop-loss brackets on a position; leverage and margin per instrument; a running P&L, order history and account history; stocks, forex, crypto and futures. Commissions can be switched on and set per your broker — percentage, fixed per order or per contract.
What it leaves out
Commissions and exchange fees are OFF until you enable them. There is no slippage beyond the quoted spread — a stop in a fast market fills at the stop price, not three ticks through it. No partial fills, no queue position, no rejected orders. On a free plan some exchanges' data is delayed, so the 'live' fill is at a price fifteen minutes old. And the balance resets to $100,000 whenever you like.
The last item is the one that matters. A real account that loses 30% is a real account that is 30% smaller, and every trade after that is sized from a smaller base under a worse mood. A paper account that loses 30% is a menu option away from being whole again — and the trader who uses it has just deleted the only information the simulator produced.
Set it up like a funded accountSix settings, once
Set the balance to the account you would actually fund
Not the default $100,000. If the real account would be $5,000, reset the paper account to $5,000 before the first trade. Position sizes, drawdowns and the temptation to size up all scale with the balance; practising with twenty times the money teaches the wrong sizes and the wrong feelings.
Switch commissions on and enter your broker's real rates
In the Paper Trading settings, tick 'Include commissions in paper trading' and choose the type: a percentage of value for stocks and crypto, a fixed amount per order, or an amount per contract for futures. Look the rate up on the broker's fee page you would really use, and include exchange and regulatory fees where the broker lists them separately.
Decide the risk per trade in dollars and never change it
One number — say 1% of the starting balance — written down before the first trade. The stop distance then sets the size, not the other way around; the Risk/Reward Calculator does that arithmetic. A paper record where the size drifted with confidence is a record of confidence, not of the strategy.
Never reset the account
Not after a bad week, not after a rule change, not to 'start clean'. If the account halves, keep trading the halved account at the same percentage risk. The drawdown is the most valuable thing the simulator will ever show you, and the reset button exists to throw it away.
Trade the instruments and hours you would trade live
One or two symbols, one session. A paper account is the only place a trader can hold twelve positions across four asset classes at 3 a.m. without consequences, and the record that produces is worthless for the two symbols you will actually fund.
Check the data is real-time for what you trade
If the exchange you practise on is delayed on your plan, every fill is at a stale price and the record is fiction. Either pay for that exchange's real-time data or practise on one that is live for you.
The arithmetic the simulator skipsFriction is a fraction of R
With commissions off and no slippage, the paper account reports a strategy's gross edge. The live account pays a toll on every round trip, and because the toll is charged in dollars while the edge is earned in R, the honest way to see it is as a fraction of what you risk per trade.
0% (paper, commissions off)
- Expectancy
- 0.125R
- Year, 250 trades
- $31,250
- Edge lost
- —
2% of R
- Expectancy
- 0.105R
- Year, 250 trades
- $26,250
- Edge lost
- 16%
5% of R
- Expectancy
- 0.075R
- Year, 250 trades
- $18,750
- Edge lost
- 40%
10% of R
- Expectancy
- 0.025R
- Year, 250 trades
- $6,250
- Edge lost
- 80%
12.5% of R
- Expectancy
- 0.000R
- Year, 250 trades
- $0
- Edge lost
- 100%
20% of R
- Expectancy
- −0.075R
- Year, 250 trades
- −$18,750
- Edge lost
- 160%
| Friction per trade | Expectancy | Year, 250 trades | Edge lost |
|---|---|---|---|
| 0% (paper, commissions off) | 0.125R | $31,250 | — |
| 2% of R | 0.105R | $26,250 | 16% |
| 5% of R | 0.075R | $18,750 | 40% |
| 10% of R | 0.025R | $6,250 | 80% |
| 12.5% of R | 0.000R | $0 | 100% |
| 20% of R | −0.075R | −$18,750 | 160% |
Source: CalculatorAI · calculatorai.app · drafts/tradingview-paper-trading-numbers.mjs
That strategy — the same modestly profitable trader our prop-firm challenge simulation uses — has an expectancy of 0.125R, which is a decent edge. It breaks even when friction reaches 12.5% of the risk per trade. How close a real trader comes to that depends almost entirely on how tight the stop is:
The same commission is 10% of one trade and 20% of another
Assume about $2.50 in commission and fees per round trip per contract and one tick ($1.25) of slippage on entry and one on the stop — retail order-of-magnitude figures; your broker's fee page has the real ones. The commission does not change; the stop does.
The paper account cannot show you this unless commissions are on, and it will never show the slippage. So the rule for reading a paper record is: compute the expectancy in R from the journal, subtract your friction as a fraction of R, and judge the strategy on what is left. Our expectancy guide shows the calculation; the friction number comes from your broker's fee page and a realistic tick or two of slippage on the stop, not from the simulator.
Log every trade like real moneyThe record is the product
The simulator keeps an order history; it does not keep a journal. The difference is everything that made the trade a decision: the setup, the plan, the reason, what you did when it went against you. Our chart-to-journal workflow covers the gate between alert and order; this is the record on the other side of it.
Open a separate 'Paper' account in the journal
In the Trading Journal, create an account named Paper with the same starting balance as the simulator. Every paper trade goes there and nowhere else, so its statistics never mix with live trades — and later you can put the two accounts side by side, which is the whole point.
Write the plan before the order, not after the fill
Symbol, direction, setup tag, entry, stop, target, size and the dollar risk — entered in the journal while the order is still a plan. A trade logged after it closed is a story about a trade. Attach the TradingView screenshot with the levels drawn.
Record the result exactly as the panel shows it
Fill prices from the order history, not from the chart; the commission column if you enabled commissions; the exit reason — stop, target, manual, time. If you moved the stop, log the original and the moved one. The panel's Download-data button exports the order history as CSV, and the journal's importer maps its columns; a screenshot of the History tab imports too.
Add one line the simulator cannot
Did you follow the plan? If not, what did you do instead, and why? This single field is what separates a paper record from a paper trade list, and after fifty trades it is the field the review lives in.
Review weekly, in R, by setup
Win rate, average winner and loser in R, expectancy, longest losing streak, and rule-adherence rate — per setup tag. A setup that is profitable only when you break its rules is not a setup.
What twenty trades can tell youLess than it feels like
The most common paper-trading mistake after resetting is stopping too soon — in either direction. Twenty winning trades feel like proof; twenty losing ones feel like a verdict. Neither is.
10
- Observed win rate, 95% range
- 20% – 80%
20
- Observed win rate, 95% range
- 25% – 65%
50
- Observed win rate, 95% range
- 32% – 58%
100
- Observed win rate, 95% range
- 35% – 55%
200
- Observed win rate, 95% range
- 38% – 52%
| Trades | Observed win rate, 95% range |
|---|---|
| 10 | 20% – 80% |
| 20 | 25% – 65% |
| 50 | 32% – 58% |
| 100 | 35% – 55% |
| 200 | 38% – 52% |
Source: CalculatorAI · calculatorai.app · drafts/tradingview-paper-trading-numbers.mjs — exact binomial intervals
Over twenty trades, that same strategy's P&L lands anywhere between −7.5R and +12.5R in 95% of runs, around an expected +2.5R. A twenty-trade paper account that is up 10R has proven roughly nothing about the edge; it has shown you the behaviour — whether you took every setup, held to the stop, and sized the same way each time — and behaviour is measurable in twenty trades even when the edge is not. Our drawdown guide covers the same problem from the other side: how long a real edge can look broken.
So the graduation test is not a P&L figure. It is a record that is long enough, honest enough, and boring enough.
Fifty trades minimum, in one account, with no reset
Enough to narrow the win-rate range to about ±13 points, and enough to have lived through at least one losing streak of five or six — which a 45% strategy produces routinely.
Commissions on, and expectancy positive after friction
Expectancy in R from the journal, minus commission and a realistic slippage allowance as a share of R. If that number is not clearly positive, tighten nothing — widen the stop or find a cheaper instrument, then start the fifty again.
Rule adherence above 90%
Counted from the 'did you follow the plan' field, not from memory. Below that, the record measures your discretion, and discretion is the thing that changes most when money is real.
The paper drawdown is one you could have funded
If the account fell 25% and you would have stopped trading a real account at 15%, the strategy's risk per trade is too high for you, whatever the paper P&L says. Our position-sizing guide works out what the number should be.
Go live at a quarter of the size
The first fifty real trades belong in a second journal account at a quarter of the paper risk. Put the two accounts side by side after that. The gap between them is the cost of being a person, and it is the most useful number you will get from either account.
Our guide to keeping a trading journal covers the fields and the weekly routine in full; the only thing paper trading changes is that the account column has to say so.
Where these numbers come from
The description of what TradingView Paper Trading does and does not simulate is from TradingView's help-centre articles on paper-trading functionality, market-order execution (buys at or near the ask, sells at or near the bid; triggers by bid or ask rather than the mid-price chart line) and on commissions in paper trading (the 'Include commissions' option with percentage, fixed and per-contract types), plus the Trading Panel's CSV download of order history; product details change and are worth re-checking in the app. The friction table and the MES example are arithmetic in drafts/tradingview-paper-trading-numbers.mjs: expectancy = p × W − (1 − p) × L with p = 45%, W = 1.5R, L = 1R, less a friction term expressed as a share of R; the year column multiplies by 250 trades and $1,000 per R. The MES figures assume $5 a point, about $2.50 commission and fees per round trip per contract and one tick of slippage on entry and on the stop — order-of-magnitude retail assumptions, not any broker's schedule. The win-rate ranges are exact 95% binomial intervals for a true 45% win rate; the twenty-trade P&L range is the exact distribution of k wins × 1.5R − (20 − k) × 1R. The model ignores compounding within the sample and assumes every loss is a full 1R, which understates live slippage on stops and so flatters the live case, not the paper one.
Frequently asked questions
Is TradingView paper trading realistic? The fills are: orders execute at the live bid and ask, not at the last price. What is unrealistic is the default configuration — commissions off, a $100,000 balance, a reset button — and the absence of slippage beyond the spread. Configure the first three and account for the fourth in the review, and the record becomes useful.
Does TradingView paper trading include commissions? Only if you turn them on. In the paper-trading settings, enable 'Include commissions in paper trading' and choose a percentage, a fixed fee per order or a fee per contract, using the rates of the broker you would really use.
How many paper trades before going live? At least fifty in one account with no reset. Fewer than that, the win rate is too noisy to judge — a genuine 45% strategy shows 25%–65% over twenty trades — but rule adherence and the drawdown are readable much sooner, and they are the better test.
Can I import TradingView paper trades into a journal? Yes. The Trading Panel's Download-data button exports the order history as CSV, which the Trading Journal imports with a column mapping; a screenshot of the History tab also imports. Keep paper trades in their own journal account so they never mix with live results.
Why do my live results differ from my paper results? Three reasons, in order: friction the simulator did not charge (commissions, slippage on stops), a sample too small to have shown the strategy's normal losing streaks, and behaviour — the plan is easier to follow when nothing is at stake. Two journal accounts side by side show which of the three it is.
