I opted for the 174 N data because the I have no 415 N data for the thinnest line that give us total friction very close to the pure block friction. I have attached a 415 N plot. As you see there is still little difference between the ball bearing block and the plain bearing blocks.
It's not plotted in log-log space like Figure 4 (not that that's a bad thing), but it looks like the ball bearing curve is shifted down by about 20%. Hard to say for sure with only 3 data points, though.
