I’m testing brokers for a first live account and care more about platform stability than a tiny spread difference.
For MT4/MT5 during fast moves, what specific checks do you run before going live? I’m thinking: demo and then micro live during major releases, track disconnects in the journal, ping to server, compare spread spikes with what you see in normal times, and see if rebates change the effective cost at all.
What stability thresholds do you accept, and what would be a deal-breaker before funding?
Run news time demo with ping monitor.
Check platform logs for disconnects during spikes.
Use a VPS near the broker server. In MT4 or MT5 enable journal logging and export it. During CPI or NFP place micro orders with market and limit types. Record time to fill and slippage. Run a script that timestamps tick gaps longer than one second. Track spread peaks and how fast they normalize. Repeat on two brokers for three events. Rebates do not fix a platform that drops. Choose the one with fewer drops and steadier fills, then factor rebates into cost.
Do not trust demo alone. Live data feeds behave differently under stress. Test with 0.01 lots and hard stops. Note re quote rate, order reject codes, and freeze times. If you see more than one disconnect per event or slippage bigger than your edge, pass. Rebates are a tie breaker only after stability and fills pass your checks. Keep screenshots and the log files so you can compare fairly.
Run three news tests on micro size, then a normal London session.
If the chart keeps ticking and orders fill within your plan, you are fine. Rebates are just a bonus after that.
MT4 froze once during CPI for me. Moved to a cheap VPS close to the server and it stopped. Rebates did not change that choice.
I tracked three events with MT5 on a small live account. Average slippage was 0.6 pip on entries and 0.8 on exits, with spreads spiking from 0.2 to 2.4 pips for under ten seconds.
That broker stayed. Rebates trimmed cost a little, but stability was the real filter.
Set your own pass fail rules. For me it is zero platform disconnects during a release and fills inside one second on micros. If a broker misses that twice, I do not fund.