What are the "latency" numbers you're talking about on the LV loop? Like, what is the actual metric you're measuring for that, and what's the value you've been

What are the "latency" numbers you're talking about on the LV loop? Like, what is the actual metric you're measuring for that, and what's the value you've been quoted for that measure. If you count time waiting in traffic, I bet trains have better latency...