I agree with PJ’s confidence in a coach holding a watch and relying on it.
I’ve been hand clocking athletes for over 20 years and have settled on a certain methodology that in my view, works pretty well and is accurate enough to know where my athletes are at. I hand time a lot of electronic races at the meets to get an idea where my times sits compared to electronic timing.
Basically I wait to the full body is past the post before stopping the watch and from there I’m generally between 0.18 and 0.22 faster. eg: an athlete runs a 60m trial in 6.85 at training I know they are probably in low 7.0 shape. I don’t accept my times if I miss by any more than the acceptable standard of 0.24; but that rarely happens.
At the pro running meets on grass I have found the (portable) timing mechanism the organisers use is slightly faster than the fixed and more sophisticated timing mechanism used at ‘amateur’ based meets. So my hand times are around 0.14 to 0.18 faster than the electronic timing. This is vitally important to know when I run trials because to hold a ‘fast watch’ can be very costly in our sport as it gives a false idea of where the athlete is really at in the scheme of things.
I always have the starter use a gun that gives off a decent smoke and I stand adjacent to the finish line. I generally take my eyes off the athletes with 5m to go and watch the line.
I have too much faith in my own ability to hold a watch to worry about spending thousands on a timing system. And historically for my own records, it offers consistency in making comparisions between athletes.