SAT or ACT Scores Stuck? What to Do When Practice Scores Stop Improving

As of October 2026: When scores stall, we start with the homework, because nine times out of 10 the work at home is not being done the full way through. Our published results show ACT students starting at 27 or below gain 5.7 points on average, and SAT students starting at 1350 or below gain 190 points, with 63% gaining 200 points or more -- and those numbers count only students who finished 12 or more instruction hours with at least 80 percent homework completion.

A flat practice score is feedback, not a verdict. Here is how we read it, what we change first, and what the research says about why.

By Nick Standlea, Test Prep Gurus. Figures may be quoted with attribution to Test Prep Gurus.

What to Do If Your Scores Arent Improving (And How to Fix It) -- Nick Standlea, Test Prep Gurus (YouTube).

Nine times out of 10, the stall is homework, not ability Nick says it plainly in his video: when a student's scores are not going up, it almost always comes back to the amount of effort put into the homework at home. That is the first place we look, before we change books, change teachers, or change tests. Our published results count only students who completed the program, which the results page defines as 12 or more instruction hours and at least 80 percent homework completion. The gains in that table are what the work looks like when it is finished, not started.

Our completed-program gains are 5.7 ACT points and 190 SAT points Across 491 ACT students from 2017-2023, those starting at 27 or below gained an average of 5.7 points. Across 291 SAT students in 2024, those starting at 1350 or below gained an average of 190 points, and 63% of them gained 200 points or more. That is 782 students in the published results. One flat practice score early in a program does not predict the end of that curve.

Taking tests beats rereading, and the research is clear about it Dunlosky and colleagues (2013) rated practice testing and distributed practice as high utility, while rereading and highlighting got low utility ratings. Roediger and Karpicke (2006) found that repeated study looked better after 5 minutes, but prior testing produced greater retention at 2 days and 1 week. So the study method that feels most productive in the moment is often the one that fades. If your student is rereading explanations, that is a fixable cause of a flat score.

What are Test Prep Gurus' published results?

These are the numbers from our results page. They cover students who completed the program, which the page defines as 12 or more instruction hours and at least 80 percent homework completion. Read them as averages, not promises.

Measure	Value
ACT students starting at 27 or below: average gain	5.7 points
SAT students starting at 1350 or below: average gain	190 points
SAT students starting at 1350 or below who gain 200 points or more	63%
ACT cohort	491 students, 2017-2023
SAT cohort	291 students, 2024
Total students in the published results	782
Founded	2009

Source: prepgurus.com/results-at-test-prep-gurus. Quote these exactly as printed; they are averages for students who completed the program.

What does the research say about how to study?

Three studies sit behind the advice on this page. Each row names the study, the finding we use, and the figure the study reports. We do not print figures the studies do not state.

Study	Finding we use	Reported value
Dunlosky and colleagues (2013)	High-utility learning techniques	Practice testing; distributed practice
Dunlosky and colleagues (2013)	Moderate-utility techniques	Elaborative interrogation; self-explanation; interleaved practice
Dunlosky and colleagues (2013)	Low-utility techniques	Summarization; highlighting/underlining; keyword mnemonic; imagery use for text learning; rereading
Dunlosky and colleagues (2013)	Why highlighting rated low	Little consistent benefit; possible harm to inference
Roediger and Karpicke (2006)	Design of the study	Two experiments; free recall without feedback
Roediger and Karpicke (2006)	Final retention test intervals	5 minutes, 2 days, or 1 week
Roediger and Karpicke (2006)	Short delay result	Repeated study beat repeated testing at 5 minutes
Roediger and Karpicke (2006)	Long delay result	Repeated testing beat repeated study at 2 days and 1 week
Ericsson, Krampe, and Tesch-Romer (1993)	What deliberate practice is	Effortful activities designed to optimize improvement
Ericsson, Krampe, and Tesch-Romer (1993)	Practice-alone hours by age 20, best violinists	About 10,000 hours
Ericsson, Krampe, and Tesch-Romer (1993)	Practice-alone hours by age 20, good violinists	About 8,000 hours
Ericsson, Krampe, and Tesch-Romer (1993)	Practice-alone hours by age 20, teacher group	About 4,000 hours
Macnamara, Hambrick, and Oswald (2014)	Variance in performance explained by deliberate practice, overall	12% unadjusted; 19% corrected for measurement error
Macnamara, Hambrick, and Oswald (2014)	Variance explained in education	4%

Citations: Psychological Review, 100(3), 363-406 (1993); Psychological Science, 17(3), 249-255 (2006); Psychological Science in the Public Interest, 14(1), 4-58 (2013); Psychological Science, 25(8), 1608-1618 (2014).

What should we do first when SAT or ACT practice scores stop improving?

Take a long, hard look at how hard your student is working on the homework. That is Nick's first move in his video, and he is blunt about why: nine times out of 10, a flat score comes back to the amount of effort put into the work at home.

This is not a scolding. It is a sorting step. If the homework is being done the full way through and the score still sits flat, we have a teaching problem to solve, and we can solve it. If the homework is being half-done, no new book and no new tutor will fix that. We would just be changing the wrapper on the same habit.

So the first conversation is not about strategy. It is about what actually happens at the kitchen table on a Tuesday night. How many problems got reattempted? How many explanations got read before the second try instead of after it? Those answers tell us more than the score report does.

Our published results give the context. ACT students starting at 27 or below gained an average of 5.7 points, and SAT students starting at 1350 or below gained an average of 190 points. Those are students who completed the program, which the results page defines as 12 or more instruction hours and at least 80 percent homework completion. The gains belong to the finished work.

Why do we call a flat practice score feedback instead of a bad result?

We see the test as feedback, and we see the practice tests inside the program the same way. That is Nick's framing in his video, and it changes what a flat score means in your house.

A score report is a measurement of what is happening right now. It is not a prediction and it is not a grade on your child as a person. If what we are doing is not working, the practice test told us that early, while there is still time to change it. That is useful information, not bad news.

Parents, students, and teachers are all on the same side here. Nick says it that way on purpose. Everyone in the room wants the score to go up. When the number sits still, the right response is a shared question -- what should we adjust? -- not a search for who to blame.

This matters for the mood at home. A student who hears "your score is feedback" will usually tell you the truth about the homework. A student who hears "your score is a failure" will usually hide it. Nick is direct about that risk, and we cover it in the section on faking the work.

What is the three-step homework process our students are supposed to follow?

It is three passes through the same problems, and most stalled students are stopping after the first one. Nick lays out all three in his video, and he repeats them because this is what he comes back to nine times out of 10.

Step one: attempt the problems and check whether they are right or wrong. That feels like finishing. It is not. In Nick's words, you are only halfway done at that point.

Step two: reattempt every problem you answered incorrectly, plus the ones you circled along the way because you were not really sure. Being unsure and being right is still a gap. That circle is a flag for future you.

Step three: go back a third time to the ones still left over -- the ones you could not figure out, or the ones you do not feel you have mastery over. Only after that third honest try do you check the explanations. Nick is firm here: do not look at them while you are trying to solve it the third time.

Whatever is left after all that, a small handful of problems, is what you bring back to your teacher. That is the right use of class time. A student who brings back forty unattempted problems is asking for a lecture. A student who brings back four is asking for the exact help that moves a score.

Why does checking the explanations too early keep a score flat?

Reading an explanation before you have tried hard yourself turns practice into reading, and Dunlosky and colleagues (2013) rated rereading as a low-utility technique. In the same review, summarization, highlighting and underlining, the keyword mnemonic, and imagery use for text learning also received low utility ratings.

The two techniques that received a high utility rating were practice testing and distributed practice. That is the whole list at the top. Our three-step homework process is built on the first one, because each pass asks the student to pull the answer out of their own head.

Roediger and Karpicke (2006) show why the early-explanation habit fools people. In their two experiments, students studied prose passages and took free-recall tests without feedback. When the final test came after 5 minutes, repeated study beat repeated testing. But at 2 days and 1 week, prior testing produced greater retention.

Read that gap carefully, because it explains a lot of flat scores. Rereading an explanation feels effective right away. The feeling is real; it just does not last to test day. Pulling the answer out of your own head feels harder and worse, and it is the one that holds.

So when your student says the homework "made sense," that is not the same as knowing it. Making sense is the 5-minute result. The official exam is the 1-week result.

Our student says the homework is done -- how do we know if it really is?

Nick says it out loud in his video: there are ways to fool your instructors that you were really doing the work. He says it because pretending costs the student more than it saves.

His argument is simple. If you convince a teacher to mark 100% instead of 50%, what does it achieve? You still have to take an official exam at the end. You still find out where you really are. The marked sheet does not sit for the test.

So we tell students to prioritize learning over not getting in trouble. Nick promises no judgment from teachers if a student was too busy to do the homework. It is far better to say "I did not do it" or "I did not do it thoroughly" than to pretend.

For parents, the useful check is not a yes-or-no question. Ask what the third pass looked like this week. Ask which problems got circled. Ask how many were left over for the teacher. Those questions cannot be answered by a student who skipped steps two and three.

And when the score is not going up, Nick asks for honest conversations in every direction -- student and parent, director and parent, teacher and student -- about what is really going on with the homework at home. Honesty is part of the process, because only the student knows whether mastery is real.

How does a long-term goal change what a practice score means?

It makes the work a priority, which is the thing Nick says a student must do for scores to move. He starts the whole conversation with the long-term goal, even for students who have no idea what school they want.

Something is a whole lot better than nothing. Figuring out where you do not want to go to school counts. Deciding that you want options when you apply to several schools counts. The more specific the goal gets, the better, but a rough goal beats no goal.

Nick's suggestion is concrete. Visit some schools. Look at some schools on YouTube. Learn about them and see what gets you excited. Somewhere in there is a goal for the future.

He is also clear that we are not here to judge the goal. He mentions students who want to go to USC because they like watching the football team on TV. That is a real goal several students have named, and it is fine. If it motivates a student and they can be passionate about it, run with it.

The reason is practical, not sentimental. The three-step homework process asks for a third attempt at a hard problem on a school night. No student does that for an abstraction. They do it because the score is the path to something they actually want.

Does practice really guarantee a higher score?

No, and the research is honest about that. Macnamara, Hambrick, and Oswald (2014) found that deliberate practice explained 12% of performance variance overall, unadjusted, with an estimate of 19% after correcting for measurement error, and only 4% in education.

That 4% figure is the one to sit with. It means practice is a real part of learning and not the only part. Sleep, stress, reading habits, math background, and test-day nerves all live in the rest of the picture.

Ericsson, Krampe, and Tesch-Romer (1993) described deliberate practice as effortful activities designed to optimize improvement. In their Berlin violinist study, estimated practice-alone hours by age 20 came to about 10,000 for the best group, about 8,000 for the good group, and about 4,000 for the teacher group. Those are estimates of hours, reported by group.

We hold both ideas at once. Practice is not magic, and skipping it is still the most common reason a score sits still. A student who does the three passes is doing the effortful kind of work those researchers describe. A student who checks answers and stops is doing something else.

For a family, the takeaway is calm. Do not promise your child a number. Promise them that the work changes their odds, and point at our published results as a record of what finished work has looked like.

How long should we wait before changing something in the plan?

We change something as soon as the feedback says the current approach is not working, and that is Nick's rule in the video: if what we are doing isn't working, we need to adjust something and try something different.

That is a short leash on purpose. The practice tests inside a program exist to give us that signal. Waiting three more months for a different number out of the same habits is not patience; it is drift.

What changes first is the homework process, not the material. Nick's order is homework effort, then the goal that makes the effort a priority, then everything else. Most of the time the adjustment is in step two and step three of the homework -- the reattempts that were being skipped.

The second thing we look at is spacing. Dunlosky and colleagues (2013) rated distributed practice as one of only two high-utility techniques, alongside practice testing. A student cramming all the homework the night before class is doing the low-utility version of a high-utility plan.

And remember what the published results count. The ACT cohort is 491 students from 2017-2023 and the SAT cohort is 291 students in 2024, for 782 students total. Those averages come from students who put in 12 or more instruction hours with at least 80 percent homework completion. The time frame is a program, not a weekend.

What do the published gain numbers actually include?

They include 782 students who completed the program, split into an ACT cohort of 491 students from 2017-2023 and an SAT cohort of 291 students in 2024. Our results page defines completing as 12 or more instruction hours and at least 80 percent homework completion.

The gains are reported by starting range. ACT students starting at 27 or below gained an average of 5.7 points. SAT students starting at 1350 or below gained an average of 190 points, and 63% of those students gained 200 points or more.

These are averages, so some students gained more and some gained less. We print them that way because that is what the page says, and we do not reshape them into a promise for any one student. Your child's starting score, timeline, and homework habits all matter.

Notice what the completion rule does. It means the published numbers describe students who did the work, which is exactly the group a stalled student is not yet in. That is the good news hiding in a flat practice score. The main variable is still in your family's hands.

Test Prep Gurus was founded in 2009, so these cohorts reflect a long-running program rather than a single season. When a score sits still early on, we compare it to that record, not to panic.

What should a parent say at home when the score has not moved?

Say that you are on the same side, because Nick's whole frame is that parents, students, and teachers all want the same outcome. Lead with that before you lead with the number.

Then ask about the process, not the score. Ask which problems got a second attempt. Ask which ones got a third. Ask how many were left over to bring back to the teacher. A student who did the three passes can answer those in a minute.

If the honest answer is that the homework was half-done, treat that as useful news. Nick promises no judgment for a student who was too busy, and he prefers that answer over a pretend one. A clear admission gives everyone something to fix this week.

Avoid turning the practice test into a verdict. Roediger and Karpicke (2006) remind us that a method can look strong at 5 minutes and weak at 2 days and 1 week. A single flat score is one snapshot inside a longer process.

Finally, bring the goal back in. Ask what schools your student is curious about. Nick suggests visiting some, or looking at them on YouTube, until something gets them excited. The homework gets easier to do when it is pointed at something real.

What if the homework is honestly getting done and scores are still flat?

Then we adjust something else, and that is the second half of Nick's rule: if what we are doing isn't working, we try something different. The three-step process is the first suspect, not the only one.

At that point we look at how the practice is spread out, because Dunlosky and colleagues (2013) rated distributed practice as high utility. We look at whether the student is testing themselves or rereading, since rereading and highlighting both received low utility ratings in that review.

We also check the quality of the third pass. A student can complete every problem and still read the explanation at the wrong moment. Nick's instruction is specific: do not look at the explanations while you try to solve it the third time. The order is the whole point.

Honesty is part of the process here too, because only the student knows whether they truly reached mastery on a problem or just recognized the answer. That is why Nick asks for honest conversations between student and parent and between teacher and student.

And we keep the research in view. Macnamara, Hambrick, and Oswald (2014) reported that deliberate practice explained 4% of variance in education, so other factors are always in play. The plan can change. The goal does not have to.

What is the move this month?

Take a free mock SAT or a free mock ACT, and use the result as feedback rather than a verdict. A fresh, full-length baseline tells us whether the stall is in one section or across the test.

Then run one week of homework the full three-pass way. Attempt and check. Reattempt every miss and every circled problem. Try the leftovers a third time before any explanation gets opened. Bring the small handful that remain back to your teacher.

Pair that week with one goal conversation. Look at two or three schools together online and find the one that gets a reaction. Nick's point is that something beats nothing, and a rough goal is enough to make the homework a priority.

If you want help setting the baseline, we offer a free mock SAT and a free mock ACT. Call us at 949.229.5359 or start at https://www.prepgurus.com/take-the-next-step. We have been doing this since 2009, and our published results cover 782 students who finished the work.

Questions parents ask

What should we do first when practice scores stop improving?
Take a long, hard look at how hard your student is working on the homework.

Is a flat practice score a bad sign?
We see the test as feedback, and we see the practice tests inside the program the same way.

What are the three homework steps?
Step one: attempt the problems and check whether they are right or wrong.

Why should students wait to read the explanations?
Reading an explanation before you have tried hard yourself turns practice into reading, and Dunlosky and colleagues (2013) rated rereading as a low-utility technique.

Does practice guarantee a higher score?
No, and the research is honest about that.

What do the published gain numbers include?
They include 782 students who completed the program, split into an ACT cohort of 491 students from 2017-2023 and an SAT cohort of 291 students in 2024.

How long should we wait before changing the plan?
We change something as soon as the feedback says the current approach is not working, and that is Nick's rule in the video: if what we are doing isn't working, we need to adjust something and try something different.

What should a parent say at home when the score has not moved?
Say that you are on the same side, because Nick's whole frame is that parents, students, and teachers all want the same outcome.

Method and sources: The gain figures come from the Test Prep Gurus results page at prepgurus.com/results-at-test-prep-gurus, which reports averages for students who completed the program -- defined there as 12 or more instruction hours and at least 80 percent homework completion. The ACT cohort covers 491 students from 2017 to 2023 and the SAT cohort covers 291 students in 2024, for 782 students total; students who did not complete the program are excluded, so these are not figures for everyone who enrolled. The study findings come from four published articles: Ericsson, Krampe, and Tesch-Romer (1993), Roediger and Karpicke (2006), Dunlosky and colleagues (2013), and Macnamara, Hambrick, and Oswald (2014). One known gap: the exact recall percentages from Roediger and Karpicke's Experiment 1 are not established from the available primary-source text, so we report only the direction of their results at 5 minutes, 2 days, and 1 week. Rows in the research table are grouped by study and ordered to match the sequence used in the prose.