The implication that proton momentum is just divided between the three (valence) quarks makes things seem better than the real situation, where a large fraction of the proton momentum is typically carried by myriad "sea partons". The fraction of the time that more than 1/3 of the momentum is carried by one quark or gluon is very small, although we increasingly rely on it to push exclusion limits upwards.

It depends on the nature of the timer limitations. If there's the possibility of a constant, unknown offset -- e.g. your reaction time always makes things later, never earlier -- then there's an uncertainty related to the unknown nature of that systematic offset. You could attempt to estimate (i.e. calibrate) or reduce that effect. On the other hand, if the inaccuracy of the timing is random, then it will reduce as $\sigma_\mathrm{timer} / \sqrt{N}$, along with other purely statistical effects.

I agree with @innisfree: there's zero phase-space volume for e+e- -> stable Z. But the fact that it's a resonance (with a width), and that initial-state radiation can "return" off-resonance e+e- beams to the peak means that e.g. LEP beams didn't need to be tuned to the Z mass with perfect accuracy.

Something I'd not thought about before: are 2->1 processes still undefined with ep collisions, where there's only one PDF to integrate over? There's still only a single configuration, i.e. zero phase-space volume, that satisfies on-shell constraints... I guess a delta remains unintegrated?

Sure, I agree: that gives you a continuous distribution to integrate over, and hence get rid of your delta functions. I was just rephrasing your statement about 2->1 not being well-defined without phase-space integrals. PDFs (and the constrained situation of 2 incoming hadrons in a head-on collision) are taking us further away from the reverse of a decay, though.

I think an intuitive version of what @innisfree is saying is that the decay happens in the rest frame of the decaying particle, with the outgoing particles' momenta perfectly balancing and their energy exactly adding up to the mass of the decayer. Finding the same set of particles somewhere with exactly the required momenta for that perfect balance to create a perfectly on-shell single particle is vanishingly unlikely. If you have 2 or more particles in the final state, they can absorb some imperfection into their relative momenta, but that's not the inverse of an unprompted decay.

This is a very strange definition of how high-energy experimenters regard photons. Real photons are hard gamma rays which are observend via electromagnetic showers in calorimeters. They are also produced in scintillation detectors, and via bremstrahlung (cf. track curvature). We also do talk a lot about them as Feynman integral propagators, and are probably a bit more cavalier than we should be about calling them "particles" in that context ;-)

Importance of higher-order corrections does not make a theory non-perturbative, it just makes it slowly converging. QCD is non-pertubative at low scales because its beta function has a negative sign, and hence the coupling diverges there... such that "higher order" has no physical meaning at all. QCD at a relatively high scale like $M_Z \simeq 91 GeV$, where $\alpha_s \simeq 0.118$, is quite successfully perturbative. And fundamentally the question here is not about whether the coupling is small, but whether the theory is non-abelian, i.e. has tree-level self-interactions... cf. EW theory.

@Annera If it helps, I think of an effective theory in this context as defining a new effective field which works in the regime where its constituent fields are strongly correlated to the extent that they appear to have no separate dynamics. This is what we do all the time when we interact with macroscopic objects, when chemists and atomic physicists do things with atoms & molecules, and in nuclear physics & engineering. There are more fundamental dynamics but in situations where there's not enough energy to probe them, you can use the composite system as if it were a field in its own right.

@anna As you just said, ATLAS and CMS jet reco work differently. In ATLAS isolation both reduces electron-faking-jet and jet hadrons as a source of non-prompt leptons.

@Dar Not necessarily: calo jets include electrons until analysis level isolation is performed. "Correctly identified" also depends on what the analysis is trying to do... there is little point in lepton id and isolation in inclusive jet analyses, for example