
{/* banner prompt: Cinematic realistic photograph, 16:9 wide banner composition: three wooden crates of increasing size lined up on a warehouse floor, lit by a shaft of sunlight from a high window, a measuring tape and a clipboard resting on the smallest crate, dust particles in the light. Muted teal and warm amber colour grade, soft natural light, shallow depth of field, high detail, photorealistic, no readable text, no numbers, no logos, no visible faces. */}

This is part eleven of the [ASX order book series](/blog/2026/10/12-reading-the-asx-order-book-how-intraday-traders-actually-find-edge/). The previous post showed how [commission drag](/blog/2026/10/22-commission-drag-the-silent-partner-in-every-scalp/) punishes marginal trades. This one is about where to put your money instead.

Every trader whose journal I read sized by mood at the start. It's the default human behaviour: a bit bigger after a good run because you feel sharp, a lot bigger after a bad loss because you want it back. Over time, independently, they converged on the same fix: **grade the setup first, then let the grade decide the size.**

<KeyPoints>

- Grade every setup before entry and let the grade set the size.
- Moving size from marginal to excellent setups raises expectancy without raising average risk.
- Step base size up slowly, on discipline metrics, and down fast after a drawdown.

</KeyPoints>

## Why grading adds edge

Not all setups are equal. If you track your trades by quality, you'll almost certainly find that your best setups have much higher expectancy than your average ones, and your marginal setups barely break even.

Here's a synthetic example. Suppose a trader's setups break down like this:

<Table data={
    {
        headers: ['Grade', 'Share of trades', 'Expectancy per 1R risked'],
        rows: [
            ['A+', '20%', '+0.90 R'],
            ['A', '40%', '+0.40 R'],
            ['B', '40%', '+0.05 R'],
        ]
    }
} />

Now compare three sizing policies, all with the same average risk per trade.

<Table data={
    {
        headers: ['Policy', 'Size by grade (A+ / A / B)', 'Average size', 'Expectancy per trade'],
        rows: [
            ['Flat', '1 / 1 / 1', '1.0 R', '+0.36 R'],
            ['Graded', '2 / 1 / 0.5', '1.0 R', '+0.53 R'],
            ['Graded, skip B', '2 / 1 / 0', '0.8 R', '+0.52 R, with 40% fewer trades'],
        ]
    }
} />

The graded policy uses the same average risk and earns about 47% more per trade. Nothing about the setups changed. The money just moved to where the edge is. Skipping B setups entirely earns almost as much with less average risk and 40% fewer trades, and it also removes their commissions, which, as the last post showed, can turn a breakeven trade into a losing one.

## What makes a setup A+

The grading criteria differed between traders, but the ingredients overlapped heavily. Most were things we've covered earlier in the series.

<Table data={
    {
        headers: ['Ingredient', 'A+ looks like', 'B looks like'],
        rows: [
            ['Catalyst', 'Fresh, material, unambiguous news or a clearly mispriced open', 'Old news, vague catalyst, or none'],
            ['Book', 'Strong protection relative to volume, little resistance ahead', 'Thin protection, or size on both sides'],
            ['Context', 'Overnight lead and sector agree with the trade', 'Lead mixed or against'],
            ['Liquidity', 'You can exit your full size easily', 'You would be a big part of the volume'],
            ['Timing', 'Your best window, often the open', 'Midday chop or late in the day'],
            ['History', 'Your stats show this setup works for you', 'You have no data, or the data is weak'],
        ]
    }
} />

A simple scoring approach works well: score each ingredient 0, 1 or 2, add them up, and map the total to a grade. The point isn't precision. It's forcing yourself to evaluate the trade on the same criteria every time, **before** you're in it.

## Sizing from the grade

Once you have a grade, sizing is mechanical:

```text
risk dollars   = base R × grade multiplier
position size  = risk dollars / distance to the invalidation point
position size  = min(position size, liquidity cap)
```

The **liquidity cap** matters. Even an A+ setup can't be bigger than the book can absorb when you need to leave. That's the protection idea from earlier in the series: don't be bigger than the liquidity you're leaning on.

A common set of multipliers was something like 0.5x for B, 1x for A and 2x to 3x for A+. Some traders skipped B setups entirely during periods when they were struggling with discipline.

## Grade inflation

Here's the trap. Grading only works if it's honest, and it's hardest to be honest exactly when it matters most.

Several journals described the same pattern after a strong run: every setup starts looking like an A+. Confidence is high, standards slip, size goes up across the board, and suddenly a mediocre setup is carrying triple size. If you look back after a painful loss and find you'd rated almost everything A+ for a week, that's grade inflation, and it's extremely common.

Defences against grade inflation:

- **Record the grade before entry**, in a journal or a spreadsheet, not afterwards.
- **Review grade accuracy** weekly: do your A+ trades actually outperform your A trades? If not, your grading is broken.
- **Cap the number of A+ trades** you allow yourself per day or per week. Real A+ setups are rare.
- **Downgrade automatically** after a losing day or when you notice emotional trading.

## Mood sizing: the failure modes

<Table data={
    {
        headers: ['Mood', 'What happens to size', 'What usually follows'],
        rows: [
            ['After a big win', 'Creeps up on every trade', 'A normal loss arrives at abnormal size'],
            ['After a loss', 'Jumps up to win it back', 'Revenge trading, often in poor setups'],
            ['After a losing streak', 'Shrinks below the plan', 'The next A+ setup is undersized and the streak never recovers'],
            ['Bored', 'Random size in random trades', 'Commission bleed'],
        ]
    }
} />

Grading doesn't remove emotion. It gives you a rule to check emotion against.

## Earning size step-ups

The other sizing question is when to increase your **base** size. The journals had a very consistent lesson here, learned painfully: every step up in base size triggered a relapse in discipline. Bigger numbers on the screen make losses feel worse, which makes cutting them harder, which makes them bigger.

The approaches that worked:

<Steps>

1. **Step up in small increments**, not all at once. Something like 25% to 50% at a time rather than doubling or more.
2. **Step up on discipline metrics, not on P&L.** A clever approach is to tie size increases to process goals: days within loss limits, checklist adherence, trade count limits. Profit alone can come from luck.
3. **Hold the new size for a while** before stepping up again. Let the new size feel normal.
4. **Step down fast** after a drawdown threshold, a rule breach, or a run of bad days. Halving size after a bad day was common. It's much easier to rebuild confidence at small size.

</Steps>

## A template

<Table data={
    {
        headers: ['Rule', 'Example'],
        rows: [
            ['Base risk', 'R = a small fixed fraction of trading capital'],
            ['Grade multipliers', 'B 0.5x, A 1x, A+ 2x'],
            ['Liquidity cap', 'Position no more than a fraction of the protective depth'],
            ['Daily loss limit', 'Stop trading at minus 3R for the day'],
            ['Step up', 'Increase R by 25% after a month within all loss limits'],
            ['Step down', 'Halve R after a day beyond the loss limit or a 10R drawdown'],
        ]
    }
} />

The numbers are illustrative, not recommendations. The structure is what matters: size is a decision made by rules before the trade, not by feelings during it.

<SeriesNav series="asx-order-book" />

<Disclaimer />


94af7449d15e64b8bd0a3ecfdbf850cb54c7495896e819b4cb9dcb7392834753

How to grade trade setups A+, A and B and size positions accordingly, why graded position sizing improves expectancy, and how to scale up size safely.

Most traders size positions by how they feel: bigger after wins, bigger still after losses. The traders whose journals I read converged on grading every setup before entry and sizing from the grade. How to build an A+/A/B grading system, why it adds edge without adding risk, and how to step size up only when you've earned it.

Tag: Position Sizing

Tag: Position Sizing

Size by Grade, Not by Mood