Trend-Following Backtester · Guide · Concepts 한국어

Concepts

What is quantitative investing? Trading by rule instead of by feel

Quant investing is not advanced mathematics. It is deciding when to buy and when to sell ahead of time, in numbers, and then doing that when the day comes regardless of how you feel about it.

The definition: the rule comes first

“Quant” is short for quantitative. A quantitative strategy is one where the decision is made by a formula fixed in advance, not by judgement in the moment.

“Buy when price closes above its 20-day average” is a quant rule. It holds whether the news is good, whether you lost money yesterday, whether you feel confident. “The chart looks strong” is not — there is no guarantee you would read the same chart the same way next month.

There is one test. Hand the rule to someone else as a sentence. If they would trade exactly as you would, it is a quant rule. If it needs “use your judgement,” it is not.

Why convert judgement into rules

Not because human judgement is bad. Because it is inconsistent. The same signal gets a different response when the account is up than when it is down. Losing positions get held longer; winning ones get cut early. Knowing about the bias does not remove it.

What the difference actually looks like

Below is Bitcoin on Upbit (KRW), 2017-09-25 to 2026-07-30 — 3,215 daily bars, computed by this site's backtester.

CAGRMax drawdownSharpe
Buy and hold41.2%-86.8%0.87
Rule-based
Bollinger band breakout
84.6%-36.6%1.74

The column that matters is not CAGR — it is max drawdown. Holding meant living through a 86.8% fall from the peak. Watching $100,000 become $13,187 without selling is something very few people actually do. That is usually where the value of a rule shows up.

These figures come from parameters optimized over the whole history, so they sit above what live trading would have produced. Optimization and overfitting measures that gap directly.

How to start

  1. Pick one rule. Not a blend. Something well-documented like trend following is enough.
  2. Run it on history. Backtest it and check the drawdown you would have had to sit through.
  3. Paper trade it. Live prices, no money. Most illusions die here.
  4. Go small. An amount whose loss changes nothing about your life.
  5. Automate it. Removing yourself from the loop is the last step, not the first. See automated trading.
Test a rule nowStart with backtesting

Three common misconceptions

“You need to be good at maths”

Most strategies here stop at averages and standard deviations. The hard part is not the formula — it is following it.

“Quant always wins”

It does not. On Samsung Electronics, across 26 years, none of the ten strategies beat simply holding.

“What worked before will keep working”

The most expensive misconception. A rule chosen on past data is fitted to that past. Optimization and overfitting puts a number on how much of it survives.

Read next