In the previous installment of this series, we discussed a curious new feature (or bug, depending on your perspective) of the Enhanced ACT. Its shorter length and fewer sections mean that test results are noisier. This is great news for students willing to sit for the test five times. On average, they’ll net enough lucky sections to push their ACT superscores up by nearly three additional points, in addition to any gains they’ll see from working with a qualified test prep tutor. While this may be beneficial for motivated students, it’s not considered good psychometric practice. An important quality for measurements such as the ACT is reliability, or the degree to which the test consistently captures whatever it’s measuring. By shortening the test, the ACT has undermined reliability. According to the ACT’s own figures, students will now score anywhere from 2.7 points above to 2.7 points below their true ability on any given test, 95% of the time.
In theory, any test used for competitive admissions should meet a higher bar for reliability, especially when stakes are high and decisions hinge on small differences of just one point. A 2 or 3 point swing isn’t a rounding error. It’s the difference between admission to a safety school versus a reach school. Under superscoring, this issue intensifies because some students effectively have the opportunity to take multiple attempts, while others may have only one. Yet colleges continue to treat superscores as if they reflect stable academic potential. They do not. Instead, superscores reflect which students had the time, money, and tolerance to keep testing repeatedly, and even then, success isn’t guaranteed. Some students who sit for multiple sittings will see their superscores jump by 5 points, while others may only gain one or two. Thus, the improvement itself is unpredictable, making the reputation strategy just another gamble—one where there is a house that always wins.
Winners and Losers of the “Enhanced ACT” Update
Winner: ACT Inc.
The ACT earns revenue by administering its test, so it has every incentive to maximize the number of sittings per student. The organization recognizes that its test design, combined with its superscore reporting policy, will incentivize students to test repeatedly. Students speak to one another and share the observation: those who invest in multiple sittings significantly increase their competitiveness. But the ACT isn’t waiting for these conversations to happen organically. It’s incentivizing major test prep providers, such as Revolution Prep, to promote multiple sittings by allowing them to resell multi-test packages. For $299, students are encouraged to prepay for four ACT sittings. Revolution Prep explicitly markets the superscore strategy, openly stating in its promotional materials: “Multiple Sittings is the Secret Behind a Top ACT® Score.” The result is straightforward. The ACT, which is owned by a private equity group, generates greater profit and, crucially, launders the message—take our test as many times as possible to get into the best college—through its test prep partners.
Winner: Motivated Students
Students who are willing to take the ACT multiple times now have a clear statistical advantage. The increased variance caused by the test’s shorter sections creates more opportunities for lucky breaks. Superscoring captures these high points while ignoring lower scores. Even two test sittings yield nearly a full point of average improvement, and five sittings can push gains to more than two points. Although each additional sitting provides diminishing returns, the data clearly indicates that more sittings produce better outcomes, particularly at colleges that accept superscores. Unlike the previous version of the ACT, where repeat testing primarily confirmed a student’s existing skill level, the new test rewards persistence by allowing score variability to work in the student’s favor. For high schoolers who approach the ACT strategically, and who have the resources to repeatedly test, this new system offers more than just a second chance. It provides a scalable method for improving scores.
Loser: Typical Students
For the average student—the one who takes the ACT once or twice—the new format quietly stacks the deck against them. Increased score volatility turns a single test day into a roll of the dice. If they land on an off section, there is no buffer to absorb the damage. Unlike their better resourced peers, they are unlikely to sit for five rounds just to collect a few lucky breaks for a superscore. As a result, their scores reflect more randomness than skill, without the benefit of repetition to convert that randomness into upside. Worse still, colleges that rely on superscores may end up rewarding volume over actual ability. This creates a growing divide between students who can afford to play the long game and those who cannot. In this environment, typical students are structurally disadvantaged.
Loser: Colleges & Universities
In theory, colleges should not be at the mercy of test makers competing for market share. They are under no obligation to accept exams that fail to meet basic psychometric standards, and the ACT arguably no longer does. Yet colleges are caught in their own competitive race and often resist policies that might reduce application volume. A school that distances itself from the ACT risks losing motivated applicants and shrinking its pool to just SAT takers, diminishing selectivity. As a result, institutions remain tethered to a test that no longer meets reliability benchmarks. Worse still, they help sustain a system that students and advisors are increasingly learning to exploit, even if unintentionally. The end result is a landscape in which colleges feel forced to rely on a measure they know is gamable and lacks sufficient reliability.
How Colleges Can Push Back: Ditch the Superscore
The problem with superscoring is the inconsistency. Students who can afford or are willing to test five times will significantly outpace their peers. But even if all students played the superscore game, the outcome would still be inequitable. It’s a game: some testers will win big; others will win small. That asymmetry undermines the fairness colleges claim to support. If institutions were serious about leveling the playing field, the simplest solution would be to stop accepting superscores altogether. That wouldn’t eliminate the problem—students can still report their best single-day result, which is itself subject to cherry-picking the best outcomes from a volatile measure—but it would rein in the excesses of the superscore system.
It’s likely that these conversations are already happening behind closed doors. After the test-optional movement, many colleges were able to conduct a natural experiment by comparing students who submitted scores with those who did not. The results were clear: standardized tests outperformed every other admissions metric in predicting college success. Raw GPA contributed little to no additional predictive value. Given that entrance exams remain the only truly objective data point in most applications, colleges may finally be ready to eliminate a major source of noise and bias by dropping ACT superscores.
Disclaimer: The views expressed here are the author’s personal opinions, based on publicly available information and analysis. This post does not reflect the views of ACT Inc., Revolution Prep, or any college or university.
