What are the odds?

Exact face-landing probabilities for bouncing objects: Edge probability in the coin toss and the three-sided die problem

The paper revisits the classical physics problem of what is the probability a thick coin lands on its side. They study the mechanics of a cylinder of a given thickness and radius, being given an initial random angular velocity and linear velocity. The cylinder is then allowed to bounce inelastically until it comes to rest either on one of the faces or on its edge. They then use the areas of phase space which correspond to each of the resting configurations in order to compute the respective probabilities as a function of the thickness to diameter ratio. They find that for example a £1 coin has a probability of landing on its edge of ~1/1000. Comparing to experimental and simulated data they find decent agreement. Further, they calculate the thickness to diameter ratio which would provide a 1/3 probability of landing on the edge. They calculate this to be ~0.831 which is much closer to experimental and numeric studies than previous theoretical suggestions.

https://journals.aps.org/pre/pdf/10.1103/PhysRevE.105.L022201

Hamiltonian memory: An erasable classical bit

The authors consider a model of an information-carrying system in which the information is carried in the phase of a particle moving around a ring. They show that a (magnetic) Hamiltonian can be used to compress a uniform phase distribution to a highly-peaked one, apparently at the cost of no work input. It is unclear to me why this doesn't violate the second law - is this density in phase angle not exploitable as a non-equilibrium store of work? If not, why not?

https://journals.aps.org/prresearch/abstract/10.1103/PhysRevResearch.3.013232

A coarse-grained biophysical model of sequence evolution and the population size dependence

They present a coarse grained model of sequence evolution to ask questions about the speciation rate and how it differs due to effective population size. They rely on a framework analogous to thermodynamics, where the probability of a phenotype is dependent on a balance between its true fitness and the entropy of the phenotype. Using a DNA-protein binding co-evolving system as a framework, they show that, for smaller populations, the most likely phenotype is closer to inviability than for larger populations due to the greater entropic contribution in the former, and hence speciation is faster for smaller populations. This is consistent with experimental evidence, although theirs was a first attempt to explain this occurrence theoretically.

https://www.sciencedirect.com/science/article/pii/S0022519315002039?via%3Dihub