I was excited to read in the latest ArcUser Magazine that the Python SciPy stack will be integrated from version 10.3 of ArcGIS Pro and version 10.3.1 of ArcGIS Desktop. I think this is great news for making the power of Python for scientific computing more accessible to users of ArcGIS.
I initially started Python programming with the arcgisscripting module of version 9.2 of ArcGIS, and initially did research using only this approach (Etherington 2011). However, from those early beginnings I quickly learnt that Python has massive potential for scientific analysis and visualisation, and I've been making use of the NumPy, SciPy, and Matplotlib packages for many years. As I discovered more and more about what these packages could do I even discovered that I could take a purely Python approach for some of my research (Etherington et al. 2015).
So I would urge users of ArcGIS who haven’t programmed with these newly available Python packages to seriously consider spending some time learning how to make use of them.
Note to self - must make time to learn how to make better use of Pandas...
Etherington TR, Holland EP, O'Sullivan D (2015) NLMpy: a Python software package for the creation of neutral landscape models within a general numerical framework. Methods in Ecology and Evolution 6(2): 164-168.