Does that deviate from a predictor of a standard Republican voter?
Could also say that, absent a bunch of old white guys, Trump would have lost.
Racism speaks to the motive for their vote. But surely you'd recognize that motive may not be singularity focused on any one issue, for any particular person, towards any particular candidate. Not sure how you think their motives are quantified.
Two things here:
1) you’re right that it’s difficult or maybe impossible to determine if racism makes someone vote for republicans or if being in republican circles just eventually makes you racist. I would argue it doesn’t really matter though because...
2) the analysis showed racism predicted a Trump voter even when controlling for other factors like being an old white guy. ie: between two white guys that make the same money, are the same age, live in the same place, etc, the one that reports greater racism is more likely to vote for Trump.
You could argue border security is racism, but when people fear "others" there is often a real underlying cause, such as economic fears. Trump sold the people on scapegoats to address economic woes, for policies such as trade wars and walls. I doubt you'd accuse them of hating Asians for wanting to attack China on trade.
The study used long established ways of measuring racial resentment that aren’t focused on policy issues. If you go back and read the paper you can see the details.