Sorry, not sure how to phrase this question more succinctly, otherwise the title of the post would be more targeted.
See the attached screenshot. You can ignore the bars; basically, I'm using a measure which tracks actual performance against a 0% target (see horizontal reference). I am using the same measure as the "color" mark, see legend at bottom. Week 25 (first week) has a value of ~ -40%, or almost completely red, next week / week 26 has a value of ~ +200%, or completely green. Based on the legend, I'd expect all values below the 0% reference line to be red, and all above to be green; instead, it creates a full balanced gradient shift between the two points, meaning the "halfway point" for those two weeks is ~ +75% (-50 + 200 / 2 = 75), which is where the gradient visually diverges for these points.
The issue is potentially even more pronounced on the last two data points - week 28 is +400% , week 29 is ~ -10%, so all values from ~195% and below are reddish.
I guess another way of stating my problem / question - I am expecting ALL points of this line, independent of whether they fall on a specific X axis tick or not, to line up with the Y values listed in palette legend.
Please let me know if I can elaborate further; also, I'm sorry to not include a .twbx initially, but thought it'd be better to post and perhaps i'm missing something obvious. If the specific data would be helpful and anyone is willing to dig in, let me know and I'll anonymize the data I'm working with and attach. Also, I'm decent in illustrator / photoshop, so I could mock up the output pretty easily if this would help someone understand what I am trying to work towards.
Thanks very much!
(EDIT: inserted same downloadable image into post as attachment)
Please, don't use lines for this,
use (thinner) bars or dots instead.