Looking at the original data on 538's GitHub, it seems the variables in the R package are encoded wrong.
For example in season 13 of the Bachelorette, everyone that was eliminated on night 1 has a 2 next to their name in the elimination_1 column – not clear why it is a 2. Also, I’m not seeing an R1 anywhere for Bryan A who got the first impression rose.
Looking at the original data on 538's GitHub, it seems the variables in the R package are encoded wrong.
For example in season 13 of the Bachelorette, everyone that was eliminated on night 1 has a 2 next to their name in the elimination_1 column – not clear why it is a 2. Also, I’m not seeing an R1 anywhere for Bryan A who got the first impression rose.