# Welcome to Election Analytics

Election Analytics is a web tool that tracks and analyzes polling data to forecast who will win the presidency and which party will secure control of the United States Senate. The analytics used are based on Bayesian statistics and operations research methodologies. Forecasts are based on currently available polling data and are updated as new polls become available. Each forecast provides a snapshot of the current state of the election and represents the outcome if the election was held on that day.

We have begun forecasts for the 2016 presidential election using Clinton and Trump as the likely nominees. Forecasts for the Senate will be added when sufficient polling data becomes available.

### Customize your forecast here!

## Current Forecast for the 2016 Presidential Election (Updated Daily)

## How did we do?

Click tabs below to see how we did in previous elections.

Senate 2014

There were 36 Senate races in 2014. Our model correctly predicted 35 of the
36 races, missing only North Carolina. The election-day forecast incorrectly
predicted that the race in Georgia would be decided by a runoff, but it did
predict the overall winner correctly.

Actual Percentage of Votes | Predicted Probability of Winning | ||||
---|---|---|---|---|---|

State | Democrat | Republican | Democrat | Republican | Correct? |

Alaska | 45.3% | 48.9% | 0.162 | 0.868 | Yes |

Arkansas | 39.5% | 56.5% | 0.000 | 1.000 | Yes |

Colorado | 45.2% | 49.3% | 0.009 | 0.991 | Yes |

Georgia ^{1} |
45.1% | 53.0% | 0.016 | 0.984 | Partial |

Iowa | 43.7% | 52.2% | 0.121 | 0.879 | Yes |

Kansas ^{2} |
42.5% | 48.1% | 0.374 | 0.626 | Yes |

Kentucky | 40.7% | 56.2% | 0.000 | 1.000 | Yes |

Louisiana ^{3} |
44.1% | 55.9% | 0.000 | 1.000 | Yes |

New Hampshire | 51.6% | 48.4% | 0.929 | 0.071 | Yes |

North Carolina | 47.3% | 49.0% | 0.959 | 0.041 | No |

Virginia | 49.2% | 48.4% | 1.000 | 0.000 | Yes |

President 2012

Our model correctly predicted 50 of the 51 states (including the District of Columbia). In the Neutral swing scenario, the expected number of Electoral College votes for President Obama was 304.0, while 332 were actually achieved. This discrepancy corresponds to an error of 2.11 standard deviations (based on a standard deviation of 13.30). The Strong Democratic swing scenario was the closest to the actual results, predicting 309.3 Electoral College votes for President Obama.

All 40 of the so-called non-battleground states were predicted correctly. The remaining 11 battleground states are analyzed in the following table (under the neutral swing scenario):

All 40 of the so-called non-battleground states were predicted correctly. The remaining 11 battleground states are analyzed in the following table (under the neutral swing scenario):

Actual Percentage of Votes | Predicted Probability of Winning | ||||
---|---|---|---|---|---|

State | Obama | Romney | Obama | Romney | Correct? |

Colorado | 51.2% | 46.5% | 0.955 | 0.045 | Yes |

Florida | 49.9% | 49.2% | 0.176 | 0.824 | No |

Iowa | 52.1% | 46.5% | 0.987 | 0.013 | Yes |

Michigan | 53.3% | 45.8% | 0.991 | 0.009 | Yes |

Nevada | 52.4% | 45.6% | 0.981 | 0.019 | Yes |

New Hampshire | 52.0% | 46.7% | 0.982 | 0.018 | Yes |

North Carolina | 48.4% | 50.6% | 0.014 | 0.986 | Yes |

Ohio | 50.1% | 48.2% | 0.999 | 0.001 | Yes |

Pennsylvania | 51.9% | 46.8% | 0.988 | 0.012 | Yes |

Virginia | 50.5% | 48.0% | 0.726 | 0.274 | Yes |

Wisconsin | 52.0% | 46.9% | 1.000 | 0.000 | Yes |

Our model correctly predicted 31 of the 33 Senate races. In the Neutral swing scenario, the expected number of Senate seats for the Democrats and Independents was 53.6, while 55 were actually achieved. This discrepancy corresponds to an error of 1.87 standard deviations (based on a standard deviation of 0.75). The Strong Democratic swing scenario was the closest to the actual results, predicting 53.9 Senate seats for the Democrats and Independents.

All 19 of the so-called non-battleground states were predicted correctly. The remaining 14 battleground states are analyzed in the following table (under the neutral swing scenario):

All 19 of the so-called non-battleground states were predicted correctly. The remaining 14 battleground states are analyzed in the following table (under the neutral swing scenario):

Actual Percentage of Votes | Predicted Probability of Winning | ||||
---|---|---|---|---|---|

State | Democrat | Republican | Democrat | Republican | Correct? |

Arizona | 45.3% | 50.4% | 0.057 | 0.943 | Yes |

Connecticut | 55.1% | 43.2% | 1.000 | 0.000 | Yes |

Florida | 55.1% | 42.3% | 1.000 | 0.000 | Yes |

Indiana | 49.9% | 44.4% | 0.931 | 0.069 | Yes |

Massachusetts | 53.7% | 46.3% | 0.990 | 0.010 | Yes |

Missouri | 54.7% | 39.2% | 1.000 | 0.000 | Yes |

Montana | 48.8% | 44.7% | 0.474 | 0.526 | No |

Nebraska | 41.8% | 58.2% | 0.000 | 1.000 | Yes |

Nevada | 44.7% | 45.9% | 0.011 | 0.989 | Yes |

North Dakota | 50.5% | 49.5% | 0.135 | 0.865 | No |

Ohio | 53.3% | 45.1% | 1.000 | 0.000 | Yes |

Pennsylvania | 53.6% | 44.7% | 1.000 | 0.000 | Yes |

Virginia | 52.4% | 47.6% | 0.986 | 0.014 | Yes |

Wisconsin | 51.5% | 45.9% | 0.960 | 0.040 | Yes |

Our model predicted 50 of the 51 states (including the District of Columbia). In the Neutral scenario, the expected number of Electoral College votes for President-Elect Obama was 356.3, while 365 were actually captured. This discrepancy corresponds to an error of 0.77 standard deviations (based on a standard deviation of 11.24). The Strong Democrat Swing scenario was the closest to the actual results, predicting 362 Electoral College votes.

All 40 of the so-called non-battleground states were predicted correctly. The remaining 11 battleground states are analyzed in the following table:

All 40 of the so-called non-battleground states were predicted correctly. The remaining 11 battleground states are analyzed in the following table:

Actual Percentage of Votes | Predicted Probability of Winning | ||||
---|---|---|---|---|---|

State | Obama | McCain | Obama | McCain | Correct? |

Colorado | 53% | 46% | 1.00 | 0.00 | Yes |

Florida | 51% | 49% | 0.97 | 0.03 | Yes |

Indiana | 50% | 49% | 0.27 | 0.73 | No |

Missouri | 49% | 50% | 0.30 | 0.70 | Yes |

Montana | 47% | 50% | 0.26 | 0.74 | Yes |

Nevada | 55% | 43% | 1.00 | 0.00 | Yes |

New Mexico | 57% | 42% | 1.00 | 0.00 | Yes |

North Carolina | 50% | 49% | 0.71 | 0.29 | Yes |

North Dakota | 45% | 53% | 0.25 | 0.75 | Yes |

Ohio | 51% | 47% | 1.00 | 0.00 | Yes |

Virginia | 52% | 47% | 1.00 | 0.00 | Yes |