Kamran or anyone, how often should one expect a cash_return value to differ in the 252 trading days per year?

If this is correct, in the past year, cash_return value changed nearly every day for 5 companies in QTradableStocksUS, while that fundamental changed only 1 time for 520 companies. All of the others ranging between those. While somewhat confusing, the comments at the top of this are the output for 252 trading days (1 year), however ...
I also ran this for a 1 year rolling window (shown here) and you can see in the custom chart a dramatic change from June 2014, with the rolling window effect reflecting that change in the linear ramp upward for a one year duration . My highest ideal wishes would be for these results to be faulty as it is unfortunately an example of the reason I was forced to abandon all trust in fundamentals some time ago.

How sweet I imagine real numbers must be.

Since Enterprises Value (EV) is in the denominator, and MarketCap is part of EV (plus debt minus cash), I would have expected cash_return to update daily along with ‘price’...

Yes, that is correct. The value of free cash flows is probably updated quarterly but since this is a ratio using the Enterprise value, it should change daily. BTW the value of Fundamentals.enterprise_value seems to be stuck too although should be unrelated as these are calculated separately by Morningstar.

We need some clarity on why this is the case and if there are more of these.

To compare morningstar and quandl, what is the route to import a Zacks collection containing ep_val?

https://www.quandl.com/databases/ZFA  
  Zacks Fundamentals Collection A  
    Zacks Market Value Supplement

      ep_val

        Enterprise Value (Market Cap + Longterm Debt (excluding Mortgages & Conv Debt) + Current Portion of LT Debt + Preferred Stock (Cash + Marketable Securities)  

Some import examples out there:
from quantopian.pipeline.data.zacks import broker_ratings
from quantopian.pipeline.data.zacks import EarningsSurprises

I was looking at this post

Upcoming changes to morningstar fundamental data

I am wondering if this issue is related to the old Morningstar changes. Reading the post these particular fields should remain unaffected. So I am still very confused.

Perhaps @Josh at Q can respond too. Thanks