TradingView Strategy Tester Explained: Read Every Metric
The Strategy Tester is the panel TradingView opens under the chart whenever a strategy script is loaded. It replays the script's rules over the bars on the chart and reports the result as numbers, a trade list and an equity curve. This guide explains what each number means, which settings quietly inflate results, and how to compare two strategies without fooling yourself.
The Overview tab: six numbers and what they hide
The Overview tab shows the equity curve, the drawdown chart and a row of headline metrics, none of which is enough alone.
| Metric | What it measures | What it doesn't tell you |
|---|---|---|
| Net profit | Closed-trade profit minus losses, after the commission you set | Whether it came from one lucky trade or many |
| Total closed trades | Completed round trips | Whether that is enough to trust the rest |
| Percent profitable | Share of closed trades that ended positive | How large winners and losers were |
| Profit factor | Gross profit divided by gross loss | How many trades the ratio rests on |
| Max drawdown | Largest fall from an equity peak to a later low | Whether a worse one is coming |
| Avg trade | Net profit divided by number of trades | The spread between best and worst trade |
Net profit is also shown as a percentage of the initial capital set in Properties, so changing initial capital changes the percentage without changing the strategy. Max drawdown is measured on equity including open positions, so it can exceed any single closed loss; what it means for sizing is covered in max drawdown explained.
Performance summary and the list of trades
The Performance summary splits almost every metric into All, Long and Short columns. A strategy that looks balanced overall may earn everything on the long side and bleed on the short side. It also lists gross profit and loss, the largest winner and loser, average bars per trade and commission paid.
The List of trades is the raw material behind every summary: entry and exit names, time, price, size, profit, cumulative profit, and the run-up and drawdown inside each trade. Three habits pay off:
- Sort by profit. If removing the top two or three trades flattens the result, the strategy depends on outliers.
- Click a row to jump to that trade on the chart and check the fill was realistic.
- Scan the dates. Long gaps or clusters in one month say more about the period than any ratio.
Why "percent profitable" alone misleads
A hypothetical strategy that wins 90% of trades, earning 1 unit each time and losing 15 units on the rest, loses money: ten trades produce +9 and -15. Another that wins only 35% of the time, earning 3 per winner and losing 1 per loser, makes +21 and -13 over twenty trades. The lower win rate is the profitable one.
Profit factor fixes part of this: gross profit divided by gross loss, so above 1 means winners outweighed losers in total, and 1.5 means every unit lost was matched by 1.5 won. But it says nothing about sample size; 3 over twelve trades deserves less trust than 1.4 over four hundred. Avg trade, the expectancy per trade, combines win rate and win size and is the better first glance, provided you compare it against the cost of one trade.
Commission and slippage: why the default of zero inflates everything
Unless the script author sets them in the code, commission and slippage in Properties both default to zero. Every trade is then filled at the exact signal price with no fee, which never happens in a live market.
- Commission can be a percentage, a fixed amount per order, or an amount per contract. Use the taker fee of the market you trade; it is applied on entry and exit.
- Slippage is entered in ticks and added against you on every market-type fill. Crypto perpetuals on short timeframes deserve a non-zero value; thin markets more.
Costs hit frequent, small-move strategies hardest, because they repeat on every round trip while the edge does not. The quickest honesty check: enter realistic costs and watch whether the equity curve keeps its shape. If it flattens or turns down, the strategy was trading the absence of fees, not an edge. The same check belongs in how to read a backtest.
Repainting, lookahead and fill assumptions
Some results are inflated not by settings but by how the script computes.
- Lookahead. A script that requests higher-timeframe data can, depending on how it is written, read a higher-timeframe close before that bar has finished. The backtest "knew" the future; a live chart cannot.
- Calculation on every tick. Such a strategy can act inside a bar in real time, but the historical test only sees open, high, low and close, so live behaviour and backtest become two different systems.
- Fill assumptions. On historical bars the tester has to guess whether the high or the low came first. A stop and a target inside the same bar can both look reachable, and the default guess may favour the strategy.
The test is the same for all three: watch the strategy live for a few days and compare its new trades with what the tester shows afterwards. If trades change after the fact, the backtest is not describing what you would have traded.
Bar magnifier and deep backtesting
Both are available on higher-tier plans; check TradingView's plan page for the current tier. Bar magnifier fills orders using lower-timeframe data inside each bar instead of the four OHLC points; results usually get slightly worse and more realistic. Deep backtesting runs the strategy on more history than the chart has loaded, which matters on short timeframes; some features behave differently in that mode, so label its results separately.
How to compare two strategies honestly
A comparison is only fair when everything except the rules is identical:
- Same symbol, timeframe and date range.
- Same initial capital and order-size method. Prefer fixed quantity or fixed cash; percent of equity compounds and favours whichever strategy won early.
- Same commission and slippage.
- Same fill settings: both with or without bar magnifier, both in or out of deep backtesting.
Then compare more than net profit: trade count, max drawdown relative to capital, profit factor, avg trade against cost per trade, and how each behaved in the rising, falling and sideways stretches described in bull, sideways and bear markets. The smaller drawdown is often the one you could actually have held.
Pullwave publishes its tester settings, including fees, slippage and funding for futures, along with every trade on its performance page, so these checks can be run rather than taken on trust.
FAQ
Q. What is a good profit factor?
Above 1 means the strategy made more than it lost in gross terms. Whether a value is "good" depends on the trade count and the costs included: 1.3 after realistic costs across several hundred trades says more than 4 over fifteen trades with costs at zero.
Q. Why do live trades differ from the Strategy Tester?
Usually zero commission and slippage, recalculation on every tick, higher-timeframe lookahead, or fill assumptions inside bars. Set realistic costs and compare a few live days against the tester before trusting it.
Q. Does the Strategy Tester place orders on my exchange?
No. It simulates trades on historical bars inside TradingView, has no connection to any exchange account, and any real order is placed by you.
Note: This article is general education, not investment advice. Backtest results describe the past only and do not guarantee future results. Futures can be liquidated and you can lose your capital. Start small.