Advertisement IC Markets

Broker switch post-mortem: what improved and what did not

Compare ECN/Raw spread brokers, analyze execution speeds, report slippage, and evaluate commission structures for high-frequency traders.
Post Reply
PropScalpDesk
Posts: 521
Joined: Sat Sep 19, 2026 7:50 pm

Broker switch post-mortem: what improved and what did not

Post by PropScalpDesk »

I switched raw accounts once last year after a month of rising reject rates on stops during London open.

What improved: fewer rejects, slightly tighter median spread on EURUSD in the first hour, and a deals tab I trusted more for forensics.

What did not: my expectancy. Same playbook, same mistakes, same revenge click after CPI. The broker was not the edge. Execution pain had been a convenient story.

Post-mortem rule I keep: after any switch, run thirty live sessions before declaring victory, and keep the old account's cost log for comparison. If behaviour metrics are unchanged, the switch was plumbing, not progress - which can still be worth it, just name it honestly.

I also kept size unchanged for those thirty sessions. If you size up to celebrate a switch, you will not know whether execution or behaviour moved the needle. If rejects return after the switch, I reopen the post-mortem instead of assuming I imagined the old problem. The honest label for most of my switches is "plumbing upgrade". That label keeps me from rewriting the playbook every time the login changes.
Recommended broker for automated trading & scalping IC Markets
Shadow Trader
Posts: 5844
Joined: Wed Sep 09, 2026 5:19 pm

Re: Broker switch post-mortem: what improved and what did not

Post by Shadow Trader »

Naming the switch as plumbing rather than progress is the honest label most people skip. I switched raw accounts twice chasing reject rates, and expectancy barely moved until I fixed the revenge click after data.

The thirty live sessions before declaring victory is a good bar. I use twenty, but I keep the old account's cost log open beside the new one for the whole stretch. Median spread in the first London hour and reject count on stops are the only two numbers I let decide whether the plumbing improved.

Keeping size unchanged for those sessions matters. I sized up once after a week of cleaner fills and then could not tell whether the better week was execution or the extra risk. That mistake cost me a clean comparison.

If rejects return after the switch I reopen the post-mortem instead of blaming myself first. Once it was my VPS route, not the new broker. The deals tab you trust more is worth the switch on its own when forensics used to be guesswork.

What reject rate made you switch in the first place, as a percent of stop modifies during London open?
Shadow Trader
Posts: 5844
Joined: Wed Sep 09, 2026 5:19 pm

Re: Broker switch post-mortem: what improved and what did not

Post by Shadow Trader »

PropScalpDesk, one line my first post-mortem missed was weekend gap and rollover cost. The new broker looked cleaner on London rejects, then quietly cost more on triple-swap Wednesdays and on Sunday open slips.

I only caught it because the old cost log stayed open for a full month, not just the twenty sessions around the switch. The comparison sheet now has four lines: London reject rate, median EURUSD spread in the first hour, round-trip commission, and Sunday open slip.

The attached chart is an illustration of reject rate on stop modifies during London open. It compares the old account versus the new one across the first twenty sessions after the switch.
img_13215_r2.png
img_13215_r2.png (95.98 KiB) Viewed 213 times
I also refuse to change playbook rules in the same week as a broker switch. Too many variables. Plumbing gets a verdict first, playbook tests wait.

My tighter median spread drifted about 0.1 pip within six weeks once I stopped watching the deals tab every morning. Curious whether yours held after the first month or did the same.
Amanhasnoname
Posts: 2295
Joined: Fri Oct 30, 2026 9:45 am

Re: Broker switch post-mortem: what improved and what did not

Post by Amanhasnoname »

Calling a broker switch plumbing rather than progress is the right frame. Fewer rejects and a cleaner deals tab remove friction, but they only matter if the old problems were actually costing trades. Keeping size and rules frozen while the new account settles is what makes the comparison honest, otherwise any change in results could come from anything.

One thing worth adding to the cost log is the timing of fills, not just whether they were rejected. Note the gap between the click or trigger and the confirmed fill during the London open on both accounts. A broker can accept every stop but fill it a little later and worse, and that only shows up if you record it alongside spread and swap.

On the review point, it helps to set the reopen triggers in advance as numbers, such as a reject rate or median open spread above a set level for a week. Then you are checking against a threshold instead of a feeling, and you will notice slow drift in spreads before it quietly eats into expectancy.
Post Reply