Paper-based TOEFL scores have been used to determine the level of English proficiency for EFL learners for various purposes. However, in repeat tests some lower scores fluctuate despite no additional classroom learning, thus they cannot be used to judge the English level of those taking the test. There is limited research into the lowest score that does not fluctuate outside the Standard Error of Measurement, which the Educational Testing Service (ETS) set at 13 points. Therefore, this research was aimed at determining the lowest score which can be used for distinguishing the students' learning progress or proficiency. Scores of 1,180 test takers who took paper-based TOEFL a minimum of three times over three days to two weeks were analyzed statistically. The analysis revealed that the scores stopped fluctuating outside the Standard Error of Measurement when test takers reached the score of 417. Therefore, not until a test taker obtains the minimum paper-based TOEFL score of 417 can their English level be determined by the TOEFL score. This research has significant implications for employers, universities and high schools that currently use a TOEFL score lower than 417 as the minimum entrance or graduation requirement.