Protection: The Liquidity You Lean Oneric

In the previous post I argued that profitable intraday traders make decisions from market depth rather than charts. This post is about the specific depth idea that came up more than any other across the journals I read. Different traders used different words for it. The most common one was protection.

What protection means

Protection is a large resting order sitting behind your entry, on the side that would hurt you.

• If you buy, protection is a large bid just below your entry price.
• If you short, protection is a large offer just above it.

The logic is simple. If the protective order is genuine, price has to chew through it before your trade goes badly wrong. And if it starts getting eaten or disappears, you have a very clear signal that you were wrong, before the loss gets big. Your stop isn't an arbitrary number. It's somebody else's order.

Protection has a twin that matters just as much: no resistance ahead. A long with a giant bid behind it is still a poor trade if there's an equally giant offer one tick above. What these traders wanted was a lopsided book: heavy behind, light in front.

Protection is relative, not absolute

This is the part beginners get wrong, and the part the more experienced journals were most explicit about. A large-looking order is only protection if it's large relative to how much the stock actually trades.

Consider two synthetic cases, both with a 250,000 share bid just under the entry.

In Stock A, it would take more than twelve minutes of normal trading to chew through that bid. In Stock B, it could be gone in under a minute, and probably before you can react. Same order, completely different meaning.

A simple way to think about it:

Copy protection ratio = protective depth within N ticks / volume expected to trade through that zone

What counts as "expected volume" depends on the moment. At the open it's the auction volume and the first minute or two. Mid-session it's the recent per-minute volume. Into the close it's the closing auction. There's no magic threshold, but a ratio well above 1 is a different animal from a ratio below 1.

Sizing from protection

Protection also tells you how big you can be. Several of the journals sized positions by the liquidity they could exit into, rather than by a fixed dollar amount.

The reasoning: if the protective bid is your exit, you must be able to sell into it without being the one who eats it. If you're long 400,000 shares and the protective bid is 250,000, you are the reason it disappears when you try to leave. A rough rule I saw expressed several ways was to keep your size comfortably smaller than the liquidity you're leaning on, so that on a bad day you can get out in one go at roughly the price you planned.

This flips the usual sizing question. Instead of "how much do I want to make?", it's "how much can this book absorb if I'm wrong?"

Where exactly to enter

If the protective bid is at 0.485, where do you buy? The journals described two approaches.

Just in front of it. Join the bid at 0.490, one tick ahead of the big order. You get filled if sellers come down, and you're first in line before the wall. Your risk is a tick or two. The danger is that the wall is fake and gets pulled the moment you're filled.

On the offer, with the wall well behind. Pay the offer at 0.500 when the book is clearly lopsided and you expect an immediate move. You pay more but you're in when the move starts. This suits fast situations like the open.

Either way, the risk is defined by the distance to the protection, not by a percentage you picked in advance.

When protection fails

The exit rule followed directly from the entry logic. Across the journals, the signals to leave were:

What you do not do is wait for price to go through the protection and then hope it comes back. That turns a one or two tick loss into the kind of loss these traders spent years learning to avoid.

The traps

Spoofed walls. Some large orders exist to influence other traders and are cancelled before they can be hit. Watch whether a wall holds when price actually reaches it. We'll cover this properly in the post on spoofs and icebergs.

Crowded protection. If everybody can see the same wall and everybody is leaning on it, the queue in front of the wall fills up. When it breaks, everyone exits at once into the same thin air.

Protection without a reason. A big bid in a stock with no catalyst and no volume is just a resting order in a quiet stock. Protection works best combined with a reason for the stock to move: a stock in play, a mispriced open, a fresh piece of news.

Turning it into a checklist

Before a scalp entry, the depth-reading version of a pre-trade checklist looks something like this:

If you can't answer the last question in one sentence before you click, you don't have protection. You have hope.

Why this matters for systematic traders

Protection is one of the most codifiable ideas in discretionary scalping. Depth within N ticks, divided by recent volume, is a number. Whether it's consumed or cancelled is observable. Whether price moves within T seconds after entry is measurable. In the final post of this track we'll sketch how to compute these features from order book snapshots.