Compare your baby's estimated weight to published growth curves. ACOG classification with ±15-20% ultrasound accuracy context.
According to ACOG, fetal weight estimates from ultrasound have a ±15-20% margin of error. Below 10th percentile = SGA (small for gestational age), 10th-90th = AGA (appropriate), above 90th = LGA (large). The growth trend across multiple scans matters more than any single estimate.
Written by Ash K · Last updated: June 2026 · Sources cited below
Your baby's estimated fetal weight (EFW) is measured by ultrasound and compared against growth charts to determine a percentile. A percentile tells you how your baby's size compares to other babies at the same gestational age — the 50th percentile means average, not "ideal."
This calculator takes your baby's estimated weight and gestational week and returns the percentile using WHO or Hadlock growth curves.
Key Takeaway: Fetal weight percentile compares your baby to other babies at the same gestational age. Normal range is 10th to 90th percentile — this is intentionally wide because healthy babies come in many sizes. Below 10th percentile = small for gestational age (SGA), which warrants monitoring but is not automatically a problem. Above 90th = large for gestational age (LGA).
| Percentile | Classification | What it means |
|---|---|---|
| Below 3rd | Severely SGA | Close monitoring required |
| 3rd–10th | Small for gestational age (SGA) | Additional monitoring recommended |
| 10th–90th | Appropriate for gestational age (AGA) | Normal range |
| 90th–97th | Large for gestational age (LGA) | Monitor for macrosomia |
| Above 97th | Severely LGA | Delivery planning may be affected |
Note: A baby at the 20th percentile is not "behind" — they're within the normal range. Similarly, the 80th percentile is not "ahead." The percentile is a size comparison, not a performance score. Genetics (parental height, ethnicity), maternal factors, and fetal sex all influence where a baby naturally falls on the curve.
| Week | 10th percentile (g) | 50th percentile (g) | 90th percentile (g) |
|---|---|---|---|
| 20 | 249 | 331 | 414 |
| 24 | 498 | 665 | 832 |
| 28 | 858 | 1,133 | 1,408 |
| 32 | 1,337 | 1,784 | 2,231 |
| 36 | 2,082 | 2,748 | 3,414 |
| 40 | 2,773 | 3,546 | 4,319 |
Source: Hadlock FP, et al., Radiology, 1991; WHO Fetal Growth Charts, 2017.
Falling percentile is more concerning than a low single reading. A baby measuring at the 30th percentile consistently is healthy. A baby dropping from the 60th to the 15th over two scans may indicate growth restriction — reduced nutrient or oxygen delivery through the placenta.
Warning: Ultrasound weight estimates have a margin of error of ±10–15%. A baby estimated at 2,500g may actually weigh 2,125–2,875g. Your provider interprets percentile alongside amniotic fluid levels, umbilical artery Doppler, and other markers — not weight alone.
Bottom Line: Normal fetal weight percentile ranges from 10th to 90th — a wide, intentionally inclusive band. Your baby's trend over multiple scans matters more than any single measurement. Percentile is a size comparison, not a health score. Your OB interprets these numbers in full clinical context.
What is a normal fetal weight percentile?
10th to 90th percentile is considered normal (AGA). This range is intentionally wide to reflect natural variation in fetal size.
Should I worry about a low percentile?
A single low reading may reflect measurement error (±10–15%), genetic factors, or normal variation. A declining trend across multiple scans is what warrants concern. Discuss with your OB.
How accurate are ultrasound weight estimates?
Ultrasound EFW has a margin of error of ±10–15% in the third trimester. Earlier gestational ages tend to be more accurate. The estimates are useful for trend monitoring but not precise enough for definitive sizing.
Fetal Weight Categories — ACOG Classification
This tool is for informational and educational purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment. Always consult your healthcare provider with questions about your health.