B-book execution lag vs simulated slippage on news events
I have been running a fairly standard mean-reversion model across a few of the larger multi-asset evaluation firms, specifically focusing on the $EURUSD and $GBPUSD cross-sections during pre-NY session overlap. Most of these platforms claim 'raw spreads' with a commission pass-through, but the actual execution delay under simulated conditions has been highly variable.
I am noticing a pattern where simulated fills during moderate volatility (think secondary tier data releases, not FOMC) are getting pushed by 1.2 to 1.8 pips consistently. This isn't just normal market depth thinning out; it feels like artificial latency built into their MT5 gateway to mimic worst-case B-book internalization. When you are targeting 6-8 pips per trade, that slippage completely guts the edge.
Has anyone done a structured latency comparison between the providers using proprietary tech stacks versus those still white-labeling standard gray-label setups? I am trying to figure out if it is worth migrating to a firm that routes through a cleaner liquidity bridge, or if this level of 'synthetic friction' is just the cost of doing business in the evaluation space now. I would love to hear from anyone who has run parallel accounts with identical API orders to compare raw fill times.