In page 126, "it is common to say that algorithm A is better than algorithm B if the upper bound of the 95 percent confidence interval for the error of algorithm A is less than the lower bound of the 95 percent confidence interval for the error of algorithm B .

My understanding toward confidence interval is like above figure. So how does it make sense to compare higher bound of one algorithm to the lower bound of another? Thanks in advance of any insight.

ysdfysdys