When I started trading, I was fairly disciplined: plan, risk management, consistency. One thing that escaped me for a while was spread as a dynamic variable — one that tends to change exactly when it matters most.

I started on CFDs, where spread can widen significantly. Back then I treated it as just another "cost of doing business." It took time to realize that spread isn't just a number you see in a quiet market — it's a variable that:

I remember a period where I thought I was being rigorous: three times a day, for one week, I manually recorded the spread from my platform. It felt like discipline and data collection.

In reality, it was closer to impressions than data. What was missing:

If you want decisions grounded in how markets actually behave, you need history, not snapshots.

What historical spread data actually lets you do

With historical data, you stop speculating and start measuring. Not "what is the spread right now," but:

1) Gaps: size and frequency

For certain instruments (Hello DAX), it's worth having a clear picture of what's normal and what's an outlier. When you know:

it becomes much easier to decide whether it makes sense to hold through close/open.

2) Sharp moves: what spread does at turning points

The most interesting things often happen fast: a sharp move, aggressive order flow, a volatility regime shift. Looking back, you can examine:

This is the line between "the strategy works on paper" and "the strategy survives real execution." Especially useful when I'm anticipating mean reversion significantly past ATR — if a move that large actually arrives, it's likely driven by thin liquidity, and spread widening comes with it hand in hand.

3) Overnight: how often does the market move while you sleep

Not everyone wants — or can afford — to manage alerts at 3am. Historical data gives you a practical answer:

In other words: not "what could happen," but "what actually happens and how often."

4) "Volatile days" vs. normal days

From a risk perspective, the average day isn't particularly informative. What matters in practice are the days when the market behaves differently: macro surprises, geopolitical events, news-driven volatility, regime shifts.

On those days, the question usually isn't "do I need a better setup?" It's about clearer decision-making:

5) Close/open: what happens to spread when you hold overnight

If you're holding through the close/open transition, the gap in price isn't the only thing to think about. You also want to know:

This is exactly the kind of situation where a "small" change in conditions can have a meaningful impact on both P&L and composure.

6) Holidays and thin liquidity: a natural filter

Holidays and low-liquidity periods are a straightforward filter for me. Wider, less stable spreads are one reason why sometimes the right move is to step back rather than forcing a trade. Protecting capital and focus is a legitimate strategy.

Average spread is often the wrong metric

One of the more common mistakes is looking at "average spread" and drawing conclusions from it.

Averages are sensitive to outliers: a single extended spike can pull the number up significantly, even if spread is normal 95% of the time. For trading decisions, I find it more useful to look at:

"Spread doesn't matter on futures" — does that hold?

You often hear that futures are clean: transparent markets, tighter bid/ask, higher liquidity. In normal conditions, that's largely true.

But trading outcomes aren't decided in normal conditions. They're decided in moments when:

The question isn't whether futures typically have tighter spread. The question is what spread does in the moments when it matters most. And without historical data, you can't answer that reliably.

In the next piece, I'll get more practical: the specific metrics I track (median, percentiles, spike frequency and duration, average gap sizes, spread widening on volatile days, long-term spread cycles), how I set alerts without unnecessary noise, and how to turn this into a simple checklist for when not to trade.

Lukas