Calc: position measurements for cell-anchored jpgs

In trying to let some picures appear uniformily positioned each in their cell, I realised that the X and Y measurements for “position” are relative to the page and not as expected relative to the cell borders, even when the graphics are anchored to the cell. Seems strange to me.
Is this as expected?
A better way to do this?

I don’t want to forestall any proactive replies here. I’ve noted this also, and think it’s a design issue (that is, I’ve not found a way to change it). It’s especially annoying if you have a need to change row height for preceding rows in the sheet and your graphic is located, for example at row 200. The position is maintained with reference to the page, and you’re required to change the y-coordinate. This would also apply to the x-coordinate for a change in column width. I will guess implementation of cell-based coordinate references is part of the plan but has not yet implemented. I don’t have programming skills enough to help in that, but will be a part of the squeaky wheel.

Win7x64 - LibreOffice

Maybe I don’t understand well the question, but if I right-click and select anchor to cell, the graphic goes with the cell inserting/deleting rows/columns.

If the graphic is bigger than the cell, then changing the size of any cell under the graphic, change the graphic size. But if the graphic is protected to change position/size (right-click/position and size/protect) then only goes with the cell.

If the graphic is smaller than the cell, then goes with the cell but does not change their size, even changing the cell size.

Yep, VBAS96’s remarks are a consequence of what I was trying to say.
My concern was more aimed at the measurements (i.e. I was expecting to adjust a picture’s anchor point relative to a cell border (low left corner or whatever). I thought, that giving the same X/Y value to all my equally sized cell-anchored pictures, then locking their position, would yield nicely positioned graphics inside their cells, but even though they are cell anchored, they all have y-values that reflect the position on the page.)
So, basically, I think, cell anchoring might not really work as expected, as far as positioning goes, even though MarioSV has outlined some difference in respect to deleting and resizing.

As a result, when some of the preceding other rows on the page change their heights, the graphics are no longer inside their cells (which is what VBAS96 is saying).

Many thanks for your input, to both of you!