The general answer will be to use HH::ancovaplot() directly (and not implicitly through HH::ancova() ) together with the HH::ancova() grid with the convenient layer() function and the overloaded + operator. I will leave it to you to figure out the details necessary to make them look exactly the way you want it.
ancovaplot(Sodium ~ Calories * Type, data=hotdog) + layer(panel.ablineq(lm(y ~ x), rot = TRUE, at = 0.5, pos = 3))

Josh o'brien
source share