Order routing transparency: what retail traders can and can't know

[FXS London Scalper]

"Is my broker A-book or B-book?" is one of the most common questions on any trading forum. The honest answer is that a retail trader usually can't know for certain, and many brokers run a mix.

What you can find out:
- The broker's order execution policy, which regulated brokers must publish. It describes how orders are handled and what factors are considered.
- Whether the broker acts as principal (counterparty to your trades) or agent. This is often in the client agreement.
- Regulatory disclosures, where some regulators require brokers to publish data on execution quality.
- Your own execution statistics: slippage symmetry, rejects, fills during news.

What you generally can't see:
- Which liquidity provider filled a specific order.
- Whether your account is classified differently from others.
- Internal risk management decisions.

A more useful question than "A or B book?" is: are my fills fair and consistent? Symmetric slippage, reasonable rejects and no unusual treatment after profitable periods are good signs regardless of the model.

What's your approach? Do you care about the execution model, or only the statistics?

[Shadow Trader]

Some brokers publish the percentage of their clients who are long or short each pair. The data looks like a window into crowd behaviour, and a common reading is contrarian: if most retail traders are long, price is more likely to fall.

There's some truth to the idea, since retail traders as a group tend to lose, and they tend to be wrong-footed at turning points. But the indicator has several weaknesses.

It reflects only one broker's clients. A broker with a particular type of customer, such as swing traders, scalpers or beginners, will have a different mix from another.

It updates slowly relative to a scalper's horizon. A reading that refreshes every few minutes says little about the next thirty seconds.

It mixes positions of all ages. A trader who entered a long yesterday and a trader who just entered one count the same.

It says nothing about size. A thousand micro lots on one side can be outweighed by a single large institutional order that doesn't appear in the data.

As a background indicator for the day's bias, it may add a small amount of information. As a timing tool for scalping, it's close to useless.

A more honest use is as a check on personal bias. If the crowd and the trader are on the same side, it's worth asking whether the trade is original or just popular.

Has retail positioning data ever added anything to a short-term decision?