Hah.
Not only do you have to understand the physics, but you would practically have to have experience in similar experiments to be able to intuit which variables you need to include.
Take a random for-instance. There will be several lengths of wire between the elements of the detector and the analyzing circuitry. I know enough (as do several people on this board), to realize that you need to account for the light-speed delay in those wires. At the precision of the measurement necessary, though, we might need to know the speed of light in the actual conductor medium...it might trip us up if we just put down "copper" and called it good enough.
There are a bewildering number of delay stages in the entire arrangement, from the firing of the beam and potential transients there, to the spread of the detector averaging peak (because we are talking about mere DOZENS of neutrinos at the detector; a small enough number they may not statistically cluster at the center of the beam time).
As a start, you'd have to read the paper in detail. But I suspect very strongly you are going to find there are a number of assumptions and procedures that are so well-known within the field they get glossed over in the paper; unless you are an experimentalist in the field yourself you may not even notice their omission.
That's how experiments look, particularly when you are this close to the edge of the detectable.