Well it is actually based on maths (in an orthonormal plan X is horizontal and Y vertical).
It is actually how coordinates work in the layout view as well (where the origin point is actually top left and the Y axis actually goes "down").
The C2 data editor is a third-part application, there is no actual editor for C2, but according to the video example in the store, columns are used in a spreadsheet way. I haven't been able to see so far how it actually translates once in JSON imported in Construct. But you would have each X representing an actual column, and then each Y being an "index".
I would argue though that "spreadsheet" way makes more sense for humans to read.
The issue then being, you would have to "twist" the position in your mind/events when you are actually reaching for the coordinates in code.
For example, and still according to the example video, at 2:15 for example as it is being displayed:
X (horizontal)
0 = ID, 1 = Name, 2 = Value, 3 = Modifier
Then the Y (vertical, each rows)
Y 0 = skipped, used to name the columns
Array.At(1,2) = Wand +2 (Name of ID 1)
Array.At(3,5) = 4 (Modifier of ID 4)
In a logical way, you would tend to use X as ID, and for each ID have then 3 Y (Name, Value, Modifier) you can relate to.
X> ---------------- 0 ---------------- 1 ---------------- 2 ---------------- 3 ---------------- 4...
Yv
Name ------ Wand +1 ------ Wand +2 ------ Wand + 3 --- Wand +4 --- Wand + 5 ...
Value ----------- 1 ---------------- 2 ---------------- 3 ---------------- 4 ---------------- 5...
Modifier ------- 1 ---------------- 2 ---------------- 3 ---------------- 4 ---------------- 5...
In this setting :
Array.At(1,0) = Wand + 2 (Name of ID 1)
Array.At(3,2) = 4 (Modifier of ID 3) (which is the fourth entry because arrays are 0 based)
Less "readable", but true to logic.