Statistical model to test a list of proportions

I am trying to test the accuracy of two methods in determining the makeup of a standard sample. My standard sample has a makeup of 17 species with this breakdown (5%, 5%, 5%,..., 5%, 10%, 15%)= 100%. The two methods each give me a list of 17 percent values that also equal 100%. I want to test to see which method gives the most accurate results using a statistical model. I am analyzing this in R and have been using a chi-square goodness of fit with a list of percents to test against and also calculating RSS. However, these do not seem to be the best/accurate methods. I have pasted the code that I have been using for the chi-squared GOF test

pvalNA<- c(0.05, 0.05, 0.05, 0.05, 0.05, 0.05, 0.05, 0.05, 0.05, 0.05, 0.05, 0.05, 0.05, 0.05, 0.05, 0.1, 0.15)

chip1NA<- chisq.test(x=propCountNA$pCn1, p=pvalNA, rescale.p = T)

where propCountNA$pCn1 is a vector of counts that is getting tested to see if the proportions are equal to the pvalNA vector. When I had kept propCountNA$pCn1 as a list of proportions rather than counts, I wasn't sure if I was getting the correct answers so I had multiplied the proportions by number of rows in the dataset.