A clock is required for IFR flights in the US but any clock doesn't do; the clock should be the one mounted on the instrument panel. What's the idea behind this requirement? Is it because random clocks the pilots might carry with them for their flights cannot be trusted?
The clock needs to be positioned such that you can see it easily as part of your instrument scan, and the sweep second hand needs to be there to make it easy to time intervals that are not whole minutes.
The regulation has been there for a long time, so the requirement should be considered in the light of how things used to be as well as what they are now.
A very practical example from 1969 follows. The regulation predates that by many years, I am sure, but the 1969 environment shows the need.
You're in a light single engine airplane (think Cessna 150 and later a Cessna 172). You're working on your instrument rating in actual conditions in the Pacific Northwest in a non-radar environment. You're in moderate turbulence and having to keep on eye on the icing. And, though it seems ridiculous now, we didn't use headsets but instead picked up a mic for each transmission, of which there were many because in the non-radar environment there were lots of reports.
You're inbound to the VOR on the airport, after which you'll proceed outbound to the NDB that is the final approach fix. Once at the NDB, you will fly 30 seconds farther outbound on the Rwy 16 ILS, which by the way since you have no HSI you're having to navigate by reverse sensing on the old VOR/ILS CDI with its swinging left/right needle. At the 30 second point, you're to turn right to the procedure turn outbound course and fly for 45 seconds (but maybe if that wind is from the east you should go for 60 seconds - better watch that CDI needle to see how fast you get full deflection).
Oh, damn, the instructor just covered up the directional gyro to simulate its failure. Now your only heading reference is the wet compass, which is bouncing around in the turbulence, and, besides, you can't remember whether it's going to precede or lag (and by how much) in the turn you have to make to reverse course for the inbound leg of the procedure turn. Ah, but wait, all you have have to do is fly a standard-rate turn for 60 seconds and you'll be in good shape.
So, thus far, we've had three separate needs to time things to seconds in a situation where we have our hands full - left hand on the yoke, right hand for throttle, carb heat, microphone.
Given all that, do you really want to have to keep pulling back the sleeve on your left arm to monitor a relatively small clock face? Much better to have a larger clock face right in front of you on the instrument panel.
Do we really need such a clock now? Perhaps not, but it doesn't hurt, and regulations are notoriously resistant to change.
And does anybody actually fly procedure turns anymore?
We actually had a guy get busted for this. No approved clock installed on the aircraft.
After that happened, we took out all the broken mechanical clocks on our aircraft, and installed digital clocks, with batteries and a 12V feed, and a 12V panel lighting wire.
Legally, you can have all the stopwatches and wristwatches you want, but for instrument flight, there must be a clock installed. I have been told by an FAA Avionics Inspector that a GPS or other piece of avionics with a clock satisfies the requirement. Back when we had LORAN-C, one of the units had a timer in it, and we did not have a clock on that aircraft. I cannot remember an aircraft GPS which does not have a clock feature.
Among all clocks and watches it should be many models that are just unsuitable (cheap/unreliable, strange design, etc). There is a certain risk that clearly unsuitable model would formally match the requirements, or may appear it is doing. For instance, "digital presentation" - binary ok?
If the watch is a part of the plane it probably has passed the certification so should be bearable.