Poll Margin of Error Calculator

Poll margin of error from sample size, the margin of error on a candidate’s lead, and whether the lead is real or a statistical tie — with the chance the leader is truly ahead.

%
%
Weighted polls lose precision; 1.3–2 is common. Leave 1 if unknown.

Results

Reported margin of error
± 3.1 points
Lead
A +3 points
Margin of error on the lead
± 5.97 points
Is the lead real?
Statistically tied — the lead is inside its margin of error
Chance A is really ahead
≈ 83.7% (if the poll is unbiased)
Range for candidate A
44.9% to 51.1%
Sample sizeMargin of error (95%)
100± 9.8 pts
400± 4.9 pts
600± 4 pts
800± 3.46 pts
1,000± 3.1 pts
1,500± 2.53 pts
2,000± 2.19 pts
5,000± 1.39 pts
💡 The margin of error on a lead is roughly double the reported margin — so a 3-point lead in a ±3 poll is not a clear lead.
💡 Margin of error covers only random sampling error. Who answers polls, how “likely voters” are chosen, and question wording can add more error than this.
💡 Quadrupling the sample only halves the margin of error — that’s why most polls stop around 1,000 people.

Formula

MOE = z × √(p(1 − p) ÷ n) (reported at p = 50%); MOE of the lead = z × √((pA + pB − (pA − pB)²) ÷ n); × √(design effect) for weighted polls

More Math calculators