A practice mock cannot produce a real HSAT score. Here is the estimate we show instead, and every step behind it.

Step 1. Your right count is scaled to ACT’s scored count.

Each PreACT 9 Secure section carries some unscored field-test questions. On ACT’s released practice form the scored counts are English 33, Math 30, Reading 25, and Science 26, while the sections show 44, 34, 33, and 32 questions. Our mock uses the full published shape, so we scale your right count to the scored count and round down. Example: 28 right of 44 on English reads as 21 of 33.

Step 2. That count is read off ACT’s own conversion table, and we keep the low end.

ACT’s practice-test scoring guide gives a scale-score range for each raw score, because its practice form is not a live form. We read the range as printed and show ONE number, the LOW end, and that number is the one that feeds everything else. Where ACT’s printed table dips at one raw score, the low end carries over from the raw score below it, since one more right answer can never mean a lower estimate. Example: 21 of 33 on English reads 14 to 18 in ACT’s table; we show 14.

Source: ACT, 2025-26 PreACT 9 Secure Practice Test Scoring Guide (conversion tables, pages 5 to 8). Transcribed row for row; blank cells are raw scores above that section’s scored count.

Step 3. The composite is English, Math, and Reading, averaged the way ACT averages.

The CPS rubric says the composite “is made up of the English, Reading, and Math sections.” We average those three low-end scale scores and round to the nearest whole number, with .5 rounding up, which is how ACT forms a composite. Science is on the mock because the real test requires it, and it is reported, but it never enters the composite or the points.

Step 4. A national percentile is read first, as a benchmark.

ACT’s technical bulletin reports the mean and standard deviation of scale scores for grade 8 students who took the PreACT 8/9 in 2023-24. We place the low-end scale score on a normal curve with those two numbers. That sample is national and general. Chicago Selective Enrollment applicants score higher as a group, so a national percentile overstates a child’s standing among them. Step 5 corrects for that in a way we can name.

Source: ACT, PreACT 8/9 Technical Bulletin, August 2025 (Table 4.10, grade 8 scale-score mean and SD). ACT’s own percentile tables sit on its Knowledge Hub behind a login; the bulletin’s published mean and SD are the closest open benchmark. The columns above are national percentiles before step 5.

Step 5. The percentile shown is re-read against the applicant pool, and it leans low on purpose.

The one documented feature of the Selective Enrollment applicant pool is its old floor. For years CPS required the 24th national percentile in both reading and math on the MAP before a student could apply (CPS's former Selective Enrollment eligibility rule (24th national percentile in reading and math on the MAP; combined 48 for IEP and EL students)). We read the child against a pool that starts there: the national percentile minus 24, divided by 76, then rounded DOWN to the nearest five. Example: a composite of 20 is about the 75th nationally and reads as about the 70th among applicants; a composite of 26 goes from about the 95th to about the 95th.

Why this helps without overshooting: today anyone may sit the HSAT, and the families who choose to are stronger than a bare floor would make them. So the real pool is at least as strong as the one we assume, and a child’s true standing is lower than the figure shown, never higher. We do not guess how much lower, because no one publishes the shape of the pool above the floor. The figure is one number rather than a range because a range invites reading the top of it; the direction of the remaining error is stated instead.

Step 6. In the war room, the composite becomes points on this year’s 640 scale.

The new rubric multiplies the composite by 10 (320 at most) and the final 7th-grade GPA by 80 (320 at most). Every cutoff CPS has published so far came from the older 900-point rubric, so the war room reads a published cutoff on the 640 scale as the same share of the maximum: a cutoff of 855 of 900 reads as 608 of 640. That assumes the applicant pool spreads out the same way under both rubrics, which nobody has been able to check yet. It is a way to set a target, not a prediction of next spring’s cutoff.