Lito told me SPK_ADD_KS should show if it's adding or removing spark as a negative or positive number, but mine only went to .250 twice and 0.000 on one whole datalog. Again I haven't a clear idea on what the numbers actually mean. There are way to many KS PIDs, I'm going to see if I can get some tuners in here!
Lito is right. That value will either go - or + based on how much the knock sensor is adding or subtracting at that point in time.
What you would do (as a tuner) is look at the number you're after in your timing tables as well as any adders or subtractors for temperatures, etc. and find out what the timing is supposed to be at in a given range. Your spark_advance datalog should match. If not, where is the timing going? If your SPK_ADD_KS is showing -1, and you're 1 degree lower in your datalog than you had commanded, the knock sensor is taking it away.
It doesn't sound like you have knock, though.
Another tool is spark_source. Spark_source is a numeric value that can help you determine where the commanded timing numbers are coming from, since there are multiple tables, and also help you find out if timing is being pulled for some reason.
On our tunes, if you go WOT, spark source should be a flat "2" across the entire pull, and then change when you let off. This means that the timing is coming from the correct table. I'm pretty certain you'll find this to be true. That, plus seeing one of your datalogs, I could determine on our end that you are/aren't getting the correct timing based off of what we have inside the table. And if your SPK_ADD_KS is showing 0, then we can say everything looks good.
You will need to datalog AIR_CHARGE_TEMP, COOLANT_TEMP, SPARK_SOURCE, SPARK_ADVANCE, RPM, LOAD, MAF_AD_COUNTS, AIR_FLOW_#/MIN, SPK_ADD_KS, stft1, stf2, ltft1, ltft2, and that should be good for verifying that your spark is perfect. You can email it to me at chris[dot]rose[at]americanmuscle[dot]com or just attach it here, if it allows you to.