The present study was an attempt to explore the relationship between the readability index of an essay written by an L2 learner and the assessment of that. To do so, samples of participants' writing were scored by two raters using TOEFL writing scoring rubric. The readability index of each text was then calculated using eight readability formulas and graphs, i.e., Flesch-Kincaid index, Reading Ease index, the FOG index, Coleman's readability index, the SMOG formula, Fry's graph, Spache formula, and Dale-Chall readability index. Pearson product-moment correlation coefficient formula was used to check the extent to which the two sets of scores matched. The correlation coefficients obtained ranged from .02 to .15, none of which significant. This indicates that readability index of a text and the writing assessment procedure through holistic rubrics are dealing with two different constructs and have very little in common. This also calls into question the reliability and validity of some computerized assessment programs such as PEG, LSA, or E-rater, which take into account factors very similar to those examined by readability formulae.