When using/testing mdc-web on the demo site, there seems to be an issue with the js button.
0.11.1 via this page:
https://material-components-web.appspot.com/
Microsoft Edge 41.16193.1001.0
Windows 10 Build 16193
Open Edge to https://material-components-web.appspot.com/button.html and click on a button that is not CSS only.
It behaves normally or degrades gracefully to a CSS only button (which appears to be what happens on IE 11)

Definitely a bug. Thanks for filing.
A similar issue happens with radio buttons and checkboxes when using Edge (Microsoft Edge 40.15063.0.0).
https://material-components-web.appspot.com/radio.html or
https://material-components-web.appspot.com/checkbox.html
Only on the Javascript components
Thanks
The same happens with tabs too:

But at the tabs the bounding box for clicks also appears to be too large
Tested on:
Microsoft Edge 40.15063.0.0
Microsoft EdgeHTML 15.15063
Windows Build 15063.296
Apologies for being non specific. Edge doesn't seem to like material-components-web much at all (yet) - just browsing through the demos at https://material-components-web.appspot.com/ with Edge reveals quite a few issues. I have no doubt that it'll be fixed in time though. I'm really impressed with this library in general and have to say thanks for the work done so far.
Thanks for all the contributions and patience, we have an PR out for this issue #1008.
The bug is caused by Edge's partial support to CSS Variables: there are known issues when apply CSS Variables to pseudo-elements and animations, which in this case it breaks rendering of all ripples that have enabled CSS Variables feature.
PR is about to feature detect Edge and fallback to use a CSS only ripple effect under that circumstance and we will closely monitor when a future version of edge fixes this bug.
After some discussion within #1008 , we decided go with a scoped FD for Edge but rather than banning Edge as whole. The new FD will monitor the actual behavior of how CSS Variable is applied to a pseudo element and if it behave in the above describe way, then we will stop upgrade ripple.
The benefit we get from this method is if users on a future version of Edge that fix the bug, all of our components will start working with ripple while the lower version Edge will still have it fallback to CSS solution.
Most helpful comment
After some discussion within #1008 , we decided go with a scoped FD for Edge but rather than banning Edge as whole. The new FD will monitor the actual behavior of how CSS Variable is applied to a pseudo element and if it behave in the above describe way, then we will stop upgrade ripple.
The benefit we get from this method is if users on a future version of Edge that fix the bug, all of our components will start working with ripple while the lower version Edge will still have it fallback to CSS solution.