Score Improvement Methodology
How Miyagi Labs Measures SAT Score Improvement
Students and parents should be able to see exactly how a performance claim was measured.
This page explains the methodology behind Miyagi Labs student results: how baseline scores are established, how improvement is calculated, and what limitations apply.
Results vary by student. No score increase is guaranteed. The sections below explain who is counted and how. Last reviewed: September 2026.
Why We Publish Our Methodology
Score improvement claims are only useful when the methodology behind them is clear. A number on a marketing page tells you nothing unless you know who it describes.
So this page makes it easy for students, parents, educators and journalists to check the following:
- Who is included in the data
- What counts as a starting score
- What counts as an ending score
- How improvement is calculated
- Which students are excluded
- The study period measured
- Whether scores are official or practice-based
How Score Improvement Is Calculated
Baseline score
- Definition
- The student's starting point, taken from their own reported score before preparing, or from a Miyagi Labs diagnostic or full-length practice test at the start of study.
- Why it matters
- A baseline measured too late understates improvement; one measured casually overstates it.
Follow-up score
- Definition
- The student's later score, from their own reported official result or a later full-length measurement on the platform.
- Why it matters
- Official and practice scores are different measurements and should not be mixed without saying so.
Improvement
- Definition
- The difference between the follow-up score and the baseline score for the same student.
- Why it matters
- Per-student differences are averaged. This is not a comparison of two different groups of students.
Study period
- Definition
- The elapsed time between the baseline and the follow-up measurement.
- Why it matters
- Study periods vary widely. The strongest results came from consistent study across weeks, not from single sessions.
Usage criteria
- Definition
- Students counted in the improvement figure are those who actively used the platform to prepare, rather than everyone who created an account.
- Why it matters
- This is the single biggest driver of a reported average, and the one most often left unstated.
Exclusions
- Definition
- Students without both a baseline and a follow-up score cannot be included, because there is nothing to compare.
- Why it matters
- It creates self-selection, which is a real limitation and is described below rather than hidden.
What We Publish Today
Two figures are currently published: a 240-point average reported SAT improvement, and more than 85,000 students having used Miyagi Labs. Individual case studies on the results page show a named student's own reported before-and-after scores, with the exam labeled so an ACT composite is never mistaken for an SAT total.
Several metrics that would make this page stronger are not published yet, and we would rather say so than estimate them: median improvement, the percentage of students who improve, improvement broken down by starting score, improvement by study duration and weekly practice volume, the exact size of the reporting sample and total questions completed.
Those will be added here as they can be verified. Until then, no figure appears on this page that we cannot stand behind.
What Influences SAT Score Improvement?
Results vary, and they vary for identifiable reasons:
- Starting score
- Target score
- Study consistency
- Time available before test day
- Existing skill gaps
- Number of practice questions completed
- Practice-test frequency
- Quality of mistake review
What the Data Does Not Mean
A trustworthy methodology page has to state its limits. These apply to every figure on this page and on the results page:
- Individual results vary, often considerably.
- Improvement cannot be guaranteed by any preparation platform.
- Practice-score improvement does not always equal official test improvement.
- Students frequently use outside resources, including school courses and tutors, alongside Miyagi Labs.
- Different students study for different lengths of time, so an average blends very different study periods.
- Students who keep practicing are more likely to report a result, which biases any self-reported average upward.
Why state the limitations at all?
Because a claim you can check is worth more than a claim you cannot. Naming the limitations makes the remaining figures easier to trust, easier to cite and easier for a parent to evaluate against the other options in front of them.
How Miyagi Uses Performance Data
Aside from reporting, student performance data does day-to-day work inside the product. It is what makes practice personal rather than generic.
- Practice recommendations, weighted toward the skills costing a student points
- Question difficulty, adjusted as accuracy improves
- Study planning, built around a real starting score and test date
- Weak-skill identification at the domain and skill level
- Progress tracking over time, so improvement is visible rather than assumed
See the privacy policy for how student data is handled.
Frequently asked questions
What is the average SAT score improvement?
Miyagi Labs reports a 240-point average improvement for students who actively used the platform to prepare. It is a historical average across a defined group, not a prediction for any individual student, and individual results vary widely by starting score and study consistency.
How many students are included?
More than 85,000 students have used Miyagi Labs. The reported improvement average covers the subset who actively prepared with the platform and reported both a before and an after score. The exact size of that reporting subset is not currently published, and we do not state a number we cannot verify.
Are these official SAT scores?
Reported improvements draw on students' own score reports and their practice performance. Official College Board scores and Miyagi Labs practice or diagnostic scores are different measurements, and a practice-score gain does not automatically translate into the same gain on an official test.
Does Miyagi Labs guarantee score improvement?
No. No score increase is guaranteed. Improvement depends on how much a student studies, what they study and where they started, and any preparation platform is one input among several.
How often is the data updated?
This page was last reviewed in September 2026. It is reviewed when the reported figures change or when the way improvement is measured changes.
Can journalists or schools cite this page?
Yes. That is what it is for. If you need detail this page does not cover, including the inclusion criteria behind a specific figure, contact us and we will tell you what we can verify.
Transparency Builds Trust
The purpose of this page is not to make stronger claims. It is to make our claims easier to understand, verify and cite.