New Accuracy System
Posted: Tue Apr 25, 2017 7:28 pm
I've seen people seem confused about it on Gosu Chat so I thought it would be worth explaining over here if you don't know how it works or haven't read the comments on this feedback topic (Linsk's 2nd comment shows the equation used).
As Linsk has said the calculations for determining your accuracy are as so: 'Accuracy = (net CPM / (gross CPM + corrections)) * 100'
As an example I will use one of my results from the advanced test.
Perhaps the approach you have taken is to directly divide the Net CPM by the Gross CPM and multiplying the received value by 100. In this case I would have received 89.43% (601/672 * 100 ≈ 89.43), which does not match up with the accuracy value shown in the screenshot.
This is fair as you do not directly receive a value for how many corrections you made when you get the result so you cannot factor it into your accuracy calculations.
Once you do have your value for corrections the accuracy (found on your profile for each test result) would then be calculated like this:
Gross CPM = 672, Net CPM = 601, Corrections = 10
601/(672 + 10) * 100 ≈ 88.12%
And the value is equal to the one shown in the screenshot.
As Linsk has said the calculations for determining your accuracy are as so: 'Accuracy = (net CPM / (gross CPM + corrections)) * 100'
As an example I will use one of my results from the advanced test.
Perhaps the approach you have taken is to directly divide the Net CPM by the Gross CPM and multiplying the received value by 100. In this case I would have received 89.43% (601/672 * 100 ≈ 89.43), which does not match up with the accuracy value shown in the screenshot.
This is fair as you do not directly receive a value for how many corrections you made when you get the result so you cannot factor it into your accuracy calculations.
Once you do have your value for corrections the accuracy (found on your profile for each test result) would then be calculated like this:
Gross CPM = 672, Net CPM = 601, Corrections = 10
601/(672 + 10) * 100 ≈ 88.12%
And the value is equal to the one shown in the screenshot.