Skip to content
Technical Analysis

From signals to a system

The six components every complete system needs, and why most people never write theirs down.

Technical AnalysisAdvanced11 min read
Browse Technical Analysis(172)

Knowing what a hammer or an RSI divergence is does not make you a trader, any more than knowing what a scalpel is makes you a surgeon. What converts knowledge into results is a written, repeatable process — and the writing-down part is not optional decoration.

The six components

  1. 1
    Universe — what you will trade

    Define it precisely. "NIFTY 100 stocks with 20-day average turnover above ₹50 crore" is a universe. "Good stocks" is not. Narrowing the universe is the cheapest possible improvement to any system, because it removes the instruments where your edge does not exist.

  2. 2
    Setup — the condition that puts a stock on your list

    The context that must exist before you look for an entry. For example: price above a rising 200-DMA, ADX above 25, price within 5% of a 52-week high.

  3. 3
    Trigger — the specific event that makes you act

    A precise, observable event. "A close above the 20-day high" is a trigger. "It looks like it is breaking out" is not. The test: could someone else read your rule and place the identical order?

  4. 4
    Stop — where the idea is proven wrong

    Defined before entry, in ATR multiples or at a structural level. Non-negotiable. If you would not accept this loss, do not take the trade.

  5. 5
    Exit — how you take profit

    The hardest component and the most neglected. Fixed target, trailing stop, structural exit, or time-based. Most people have a detailed entry plan and no exit plan at all, which is why they hold losers and sell winners.

  6. 6
    Size — how much

    Derived from your fixed risk per trade and your stop distance. Never from conviction. Conviction is a feeling; it has no place in an arithmetic calculation.

What "edge" actually means

Expectancy = (Win rate × Average win) − (Loss rate × Average loss)
Positive expectancy
The system makes money over many trades. This is an edge.
Zero or negative
No edge. Any profits are luck and will be returned.

Example: A system that wins 38% of the time with average wins of 2.4R and average losses of 1R has expectancy of (0.38 × 2.4) − (0.62 × 1) = 0.91 − 0.62 = +0.29R per trade. It loses nearly two trades out of three and is highly profitable.

Loading interactive demo…

Testing honestly

An honest test
  • Rules fixed before looking at results.
  • Costs, slippage and taxes included.
  • Tested on data the rules were not developed on.
  • Includes at least one bear phase — 2008, 2011, 2020.
  • Few parameters, and neighbouring values also work.
Fooling yourself
  • Adjusting rules until the equity curve looks good.
  • Ignoring costs because "they are small".
  • Testing only on the last three bullish years.
  • Eleven parameters, each finely tuned.
  • Excluding the trades that "were not really valid setups".

The maximum drawdown question

Before trading any system, find its worst historical drawdown, then ask yourself honestly whether you could sit through it — because you will have to. A system with a 34% maximum drawdown will, at some point, be down 34%. If that would make you abandon it, you do not have that system. You have a system you will quit at the worst moment, which is meaningfully worse than having none.

Check yourself

Strategy A wins 70% with average win 1R and average loss 1.2R. Strategy B wins 35% with average win 3R and average loss 1R. Which has the better expectancy?

Simple bhasha mein
Maa ki recipe likhi hui

Maa ke haath ka khaana isliye ek jaisa banta hai kyunki unke dimaag mein steps fix hain. Aap "andaza se" banaoge toh har baar swaad alag. Trading system wahi likhi hui recipe hai — kab khareedna, kitna, kab nikalna — teeno pehle se likhe hue. Andaza rozana badalta hai, likha hua nahi.

What to remember
  • A system needs six components: universe, setup, trigger, stop, exit, size.
  • Every rule must be precise enough that a stranger could execute it identically.
  • Edge means positive expectancy over many trades, not a high win rate.
  • Overfitting feels like insight. Few parameters, robust neighbours, out-of-sample data.
  • Know your system’s worst drawdown before you trade it, because you will experience it.
You reached the endMark it done and keep your streak going.
Up nextFour strategy templatesPrevious: Pivot points and CPR: the intraday map
Finished this lesson?

Mark it done to track your progress through the curriculum.

Common questions

Short, direct answers to what people ask about this topic.

expectancy meaning in trading
Expectancy is the average result a system produces per trade: (win rate × average win) − (loss rate × average loss). A positive number means the process makes money across a large sample, and that is what the word edge actually refers to; zero or negative means whatever profit appeared so far was luck. Because it is an average over many trades, it says nothing about the outcome of any single one.
the six components of a complete trading system are
Universe, setup, trigger, stop, exit and size. The universe defines what you will trade, the setup is the context that puts an instrument on the list, the trigger is the precise observable event that makes you act, the stop is where the idea is proven wrong, the exit is how profit is taken, and size follows arithmetically from your fixed risk and the stop distance. Each rule has to be written precisely enough that a stranger reading it would place the identical order.
can a trading system with a 35 percent win rate be profitable
Yes — profitability depends on expectancy rather than on how often you are right. A system winning 35 percent of the time with average wins of 3R against average losses of 1R has expectancy of (0.35 × 3) − (0.65 × 1) = +0.40R per trade. A system that wins 70 percent of the time but averages 1R wins against 1.2R losses comes to +0.34R, so the lower win rate is the better system there. Losing two trades in three feels far worse than it performs, which is precisely why rules get written down in advance.
overfitting meaning in backtesting
Overfitting is tuning a strategy so tightly to one stretch of history that it captures that period’s noise instead of a repeatable pattern, so it looks superb in the test and fails in live trading. With enough parameters any past series can be fitted perfectly while forecasting nothing. The standard defences are few parameters, a check that neighbouring parameter values also work, and data reserved that the rules were never developed on.
what is maximum drawdown and why check it before trading a system
Maximum drawdown is the largest peak-to-trough fall in a system’s equity curve over the test period, and it matters because you will eventually sit through something like it. A strategy showing a 34 percent historical drawdown will at some point be down roughly that much. If that would make you abandon it, you do not really have that system — you have one you will quit at the worst possible moment.