The current hypothesis is that most likely whatever they made is not a pure sample of the material which actually superconducts - this is expected, since when y

The current hypothesis is that most likely whatever they made is not a pure sample of the material which actually superconducts - this is expected, since when you make YBCO superconductors you also tend to get low yields (i.e. ~20%) that actually superconduct. So it could be the whole sample, or it could one micron-sized link of grains of whatever the "real" material is running through the sample.