I am not sure I understand what you are describing.
Can you send us your Skyline document and maybe a screenshot of what you are seeing?
In Skyline you can use the menu item:
File > Share
to create a .zip file containing your Skyline document and supporting files including extracted chromatograms and iRT database.

Files which are less than 50MB can be attached to these support request. You can always upload larger files here:
https://skyline.ms/files.url

When you are looking at the "Score to Run" regression plot, by default, it uses the retention time calculator that produces the best fit. That is, the thing chosen on the "Calculator" right-click menu is "Auto".
This would be helpful if you had a lot of candidate retention time predictor and wanted to choose the one that most closely matches your data. However, it's not really the way that most people want to use this graph, so you would want to choose something specific from the "Calculator" right-click sub-menu. It's possible that this is the symptom that you are describing.

-- Nick

I can verify this bug reported by Whitney. To reproduce:

Open a Skyline document which contains molecules imported using an iRT library.
Using the right-click menu in the chromatogram view, select "Show X Score" (where X is the name of your iRT library)
Using ctrl+Down arrow, pan through the chromatograms. Some will display as index (RT) score along the X axis, and some will switch to display as RT randomly.

Seems to be worse for files containing large numbers of replicates.