Broker threads can be the most useful part of a trading forum or the worst. I'd like to suggest a few standards so ours stay on the useful side.
A good broker thread usually has:
The entity and account type, not just the brand. "Broker X, EU entity, raw account" is much more useful than "Broker X".
Dates. Conditions change. A 2022 complaint may not apply today.
Numbers where possible: spreads you logged, slippage in points, withdrawal time in days.
How you got the data: own trades, statements, support emails.
Things that make a thread worse:
Referral links without disclosure.
"Scam!!!" with no details.
Screenshots cropped so the time and symbol aren't visible.
Comparisons based on the broker's own marketing page.
And one more: please separate execution problems from strategy problems. A losing month is not automatically the broker's fault, and a winning month doesn't make a broker good.
If people agree, maybe moderators could pin a short version of this. Anything you'd add or remove?
A broker that takes the other side of client trades ends up with inventory. If clients are heavily long EURUSD, the broker is short. To manage that, it may skew the quotes: lower the bid and offer slightly to discourage more buying, or to encourage selling.
The skew is small, often a fraction of a pip, and changes as the net position changes. The spread may stay normal while the whole quote shifts.
For a scalper, the effect is subtle. Entries on the crowded side get slightly worse prices. Exits on that side may be slightly better. A strategy that follows the crowd pays a small, systematic tax.
The giveaway is a quote that sits off the consensus price of other brokers by a consistent small amount, in a direction that depends on the time of day or news flow.
Comparing the midpoint of a broker's quote against a reference feed over a few weeks can reveal such behavior. A drifting midpoint that correlates with client positioning suggests skew.
Whether that's good risk management or an unfair practice is a debate, but it's real and measurable.
Any case where a broker's midpoint drifted away from a reference feed in a way that looked like skew?
