How to Use the Exit Poll Probability Tables
Probability Table I
Assume a polling Margin of Error (MOE):
MOE = 3%
Assume a deviation between the exit poll and vote tally:
Dev = 4%
From the table:
The probability that this could occur due to chance: 0.45%
or
1 out of 223
PROBABILITY TABLE I
Vote Tally Deviation from Exit Poll
1% 2% 3% 4% 5%
MOE PROBABILITY OF OCCURRENCE
0.50% 0.00% 0.00% 0.00% 0.00% 0.00%
1.00% 2.50% 0.00% 0.00% 0.00% 0.00%
1.50% 9.57% 0.45% 0.00% 0.00% 0.00%
2.00% 16.35% 2.50% 0.16% 0.00% 0.00%
2.50% 21.65% 5.84% 0.93% 0.09% 0.00%
3.00% 25.68% 9.57% 2.50% 0.45% 0.05%
3.50% 28.77% 13.14% 4.65% 1.25% 0.26%
4.00% 31.21% 16.35% 7.08% 2.50% 0.71%
4.50% 33.16% 19.18% 9.57% 4.07% 1.47%
5.00% 34.75% 21.65% 11.98% 5.84% 2.50%
.........................................................
Probability Table II
Assume the number of states (N) for which the Polling Margin
of Error is exceeded:
Say N = 16 (this actually occurred)
Assume a probability that the Polling Margin of Error is
exceeded:
Prob = 2.5% (the case with a 3% MOE)
Note: there is a 95% confidence level (19 times out of 20)
that the deviation will fall WITHIN the MOE. There is a 5.0%
probabilty (or 1 out of 20) that the deviation would fall
OUTSIDE the MOE. So the actual probability is one-half of 5.0%
(2.5%, or 1 out of 40) that it would fall outside the MOE for
one of the candiates.
The probability that this could occur due to chance:
0.0000000000000738 (7.38E-14 in scientific notation)
or
1 out of 13.5 trillion
PROBABILITY TABLE II
FOR NUMBER OF STATES EXCEEDING THE MOE
States With Vote Tallies Exceeding MOE
Prob. 5 8 10 13 16
> MOE PROBABILITY OF OCCURRENCE ODDS FOR 16
5.00% 1.E-01 4.E-03 2.E-04 1.E-06 2.E-09 1 in 491 mil
4.00% 5.E-02 9.E-04 3.E-05 8.E-08 8.E-11 1 in 12.4 bil
3.00% 2.E-02 1.E-04 2.E-06 3.E-09 1.E-12 1 in 879 bil
2.50% 9.E-03 4.E-05 5.E-07 3.E-10 7.E-14 1 in 13.5 tril
2.00% 4.E-03 8.E-06 6.E-08 2.E-11 3.E-15 1 in 333 tril
1.50% 1.E-03 9.E-07 4.E-09 5.E-13 ZERO Impossible
1.00% 2.E-04 4.E-08 9.E-11 4.E-15 ZERO Impossible