Temperature vs Power has strong positive correlation

So, I just bought super high RPM fan for my laptop, it's worked fine, it reduced temperature. I noticed multivariate time series real time graph between temperature and power that they have similarity drawn (of course not totally similar). But as you see when I turned on the fan, both temperature and power goes down. While temperature is designed to know is the hardware overheating, how about power, why I need to know the state of power if I can just see temperature monitor?

Question by Muhammad Ikhwan Perwira on Super User, licensed CC BY-SA. We show a short excerpt and link to the original thread as credit.