The TPUs are doing almost 2x the images for the same cost.

That is not all that close is it?

Yeah, pretty big coincidence. However, this may change with the next TensorFlow versions, which supposedly has further speed improvements for the TPUv2.

Note also, that the ~2% performance difference is only on one model (ResNet-50) and cannot be generalized to all workloads/all of deep learning (at least not without further proof).