We know that the Mining Efficiency +5% ore extracted talent isn't working properly. Other threads have reported this, and NQ have said via twitter that they changed the talent. However they didn't say what they changed the talent to, even after persistent asking. Could it be +1% or +0.5% per talent level? It still says +5% in game, which is quite unfair to players who choose to train this, and who are unaware that it doesn't work.
In another thread in these forums, there were many wild ideas about what the talent did. So after rolling my eyes at this, @Billy_Boolaand I decided to do the obvious:
Scan a hex and record the ore value. Call this S (in kilo-litres).
Mine a known amount of ore. Call this m.
Re-scan the hex and see how much ore remains. Call this s.
Repeat until bored.
The data we recorded is here: https://docs.google.com/spreadsheets/d/1kIquhrbSAllViRFlCqtxtfS0SJCq76FWpgiyt1u6jr8/edit#gid=0
Yes we got bored fairly quickly. (Mining boring?!)
The talent as written says that
(S-s)(1+xT) = m
where T is the talent level (0-5), and x is the (for the purposes of this experiment) unknown +x% level of ore added per talent level. 0.01=1%.
If the above is correct then we can estimate x via:
x=(m/(S-s)-1)/T for any talent level >0.
The data in the above spreadsheet does this for n=30 datapoints. The results are:
Sample mean = -0.0011 (4 d.p.)
Sample standard deviation = 0.0060
95% confidence interval for the true value of x = (-0.0033, 0.0012)
Conclusion: The Mining Efficiency talent most likely adds ZERO ore per talent level.
Notes:
There is some noise (rounding) in the data as scans report ore in KL to 2 d.p. and only more when the ore amount is low. I'm too lazy to attempt to account for this.
We have no idea if the territory scan amount accurately represents the amount in the hex
I didn't attempt to see if there were any variations via planet, ore tier, player(!) or a different +x% amount per tier level.