Why Tick-by-Tick Data Matters for Backtesting

Every candle on your chart tells you four things: where price opened, how high it went, how low it went, and where it closed. That is it. Everything that happened between those four points is lost. And for backtesting, what happened inside each candle often matters more than the final OHLC values.

What Is Tick Data?

A tick is a single price update. Every time a trade is executed or a price quote changes, that is one tick. On an active instrument like EUR/USD during the London session, there can be hundreds of ticks per minute. On a quiet exotic pair during the Asian session, you might see only a few.

Tick data records every one of these price updates with a timestamp. When you replay tick data, you see exactly how price moved from one moment to the next, in the exact order it happened.

OHLC data, by contrast, summarizes all of that activity into just four numbers per time period. A 5-minute candle might contain 500 ticks, but your standard chart only shows you the open, high, low, and close. The other 496 price points are thrown away.

What OHLC Bars Hide

The problem with OHLC data is not that it is wrong. The four values it shows are accurate. The problem is what it does not show. Here are the specific issues:

The Stop Hunt Problem

Imagine a 15-minute candle with a high of 1.1050 and a low of 1.1020. Your long position had a stop at 1.1025. Looking at the OHLC bar, your stop was clearly hit because the low is below your stop level. But what if price actually went up first to 1.1050, hit your take profit, and then dropped to 1.1020 after you were already out of the trade with a profit?

With only OHLC data, you cannot know. Most backtesting engines assume the worst case or use a simple rule (like "if both stop and target are within the bar, assume the closer one was hit first"). These assumptions can make a winning strategy look like a loser, or vice versa. Tick data removes the guesswork entirely because you can see the exact sequence.

Entry Precision

If you use limit orders, you need to know whether price actually reached your entry level. OHLC data might show that the low of a candle was 1.1030, and your buy limit was at 1.1030. Did you get filled? With OHLC only, you might assume yes. But tick data could show that price briefly touched 1.1031 and bounced, never actually reaching 1.1030. That fill never happened.

Backtesting Inflation

This is the sneaky one. Backtests run on OHLC data tend to produce results that are slightly better than reality. The reason is that OHLC bars force the backtest engine to make assumptions about the order of events, and those assumptions often favor the trader.

For example, many backtest engines process the high of a bar before the low if you are long (assuming your target was hit before your stop). This is not necessarily what happened. Over hundreds of trades, these small inaccuracies add up and paint a rosier picture than what you would actually experience in live trading.

The Pain of Downloading Tick Data

Traditionally, getting tick data for backtesting has been a frustrating process:

• Finding a source. Not many brokers or data providers offer free tick data. Dukascopy is one of the few, but downloading from their servers is slow and the format needs conversion.
• File sizes. Tick data is massive. A single year of EUR/USD tick data can be several gigabytes. Storing and managing these files is a hassle.
• Format conversion. Different platforms expect different data formats. You often need to run conversion scripts to get the data into the right shape for your backtesting tool.
• Data gaps. Downloaded data often has gaps, especially around rollover times or during low-liquidity periods. You need to check for and handle these gaps.
• Keeping it updated. If you want recent data, you need to download new files regularly and append them to your existing datasets.

This is one of the reasons many traders give up on tick-level backtesting and settle for bar data. The effort of maintaining a tick data library is real work.

StrategyTune solves this by streaming tick data directly from multiple broker sources in the browser. There is nothing to download, no files to manage, and no conversion needed. You pick an instrument and the data is there, ready to replay.

Tick Data vs OHLC: Quick Comparison

Data from Multiple Brokers

Another thing worth mentioning is that tick data varies between brokers. Each broker has slightly different spreads, slightly different feeds, and slightly different execution. A strategy that works perfectly on one broker's data might show different results on another's.

StrategyTune provides market data from multiple broker sources for forex, indices, commodities, and crypto. This means you can test your strategy against different data feeds to see if the edge holds up across providers, not just the one you happen to use for live trading.

When OHLC Is Good Enough

Not every strategy needs tick data. If you trade on daily charts with stops that are 100+ pips wide, the events inside a single candle probably do not affect your outcome. For position traders and long-term swing traders, OHLC data is generally fine.

But if you trade intraday, use tight stops, or care about the difference between a 45% win rate and a 50% win rate, tick data is not optional. It is the difference between backtesting results you can rely on and results that are contaminated by guesswork.

For more on the practical differences between tick and bar replay, see chart replay vs bar replay. And if you want to start backtesting with tick data right now, StrategyTune is free and runs in the browser with no setup needed.