You are perfectly fine.
I understand that the higher the normalized attribute value numbers (not the raw), the better. Only when the score is low enough to reach threshold, is there worry.
From the application FAQ's: "The manufacturer defines a threshold number against which the “value” is compared, and if the “value” drops below that threshold, SMART reports that attribute as FAILING_NOW, and sets the overall health as FAILED."
"Raw" is the data attribute the drive itself outputs, when the drive is tested. The value itself may not correspond to an actual "human readable" figure, but one that needs to be "translated" to achieve either a relative, interpretive value, or an actual normalized value.
Most of the normalized values in the SMART test are interpretive values, relative to a possible high good score, say of "99".
An example (not from this app): A device may output a Celcius temperature reading. The program might be set up to display that reading in Fahrenheit. Therefore, the raw might be (0) zero, but the "normalized" figure would be then displayed as "32".
Another example might be that the device outputs a figure of 1,000,000 (one million), but the threshold "bad number" is a raw 10,000,000,000 (ten billion). The normalized figure might be a score of "99" out of a possible "99", and the worst it ever tested was "99" as well.