2012 was a long time ago. Seriously. Would you use un-patched software from 2.5 years ago? I wouldn't. Science can move fast too. Relying on old data is foolish when there's a lot more newer data that can be analyzed.
The newer data shows the same feeble trend as before. A little plus, a little minus. Press releases notwishstanding the signal has been weak and remains weak. Best to avoid confirmation bias here.
