SIP Trunk Failure Rate vs Performance Dashboard Discrepancy

Quick question about SIP trunk failure metrics in our EU-West BYOC environment. The Performance Dashboard shows a 2% failure rate for a specific queue, yet the underlying Architect flow logs indicate a 15% rejection rate due to SIP 488 errors. SIP 488 Not Acceptable Here

Is the dashboard aggregating these errors differently, or is there a known latency in metric calculation for complex routing flows? Need to align these figures for the weekly operations review.