The Elections Performance Index (EPI) is the premier assessment tool to evaluate election administration across the nation.

The Elections Performance Index (EPI) compares election administration policy and performance across the states and from one election cycle to the next. The index presented here is based on the 2008, 2010, 2012, 2014, and 2016 elections.

The EPI is intended to help policymakers, election administrators, and other citizens:

  • Evaluate elections based on data, not anecdote.
  • Compare the performance of elections across states and time.
  • Identify potential problem areas that need to be addressed.
  • Measure the impact of changes in policy or practice.
  • Highlight trends that otherwise might not be identified.
  • Use data to demonstrate the need for resources to state and local policymakers.
  • Educate voters about election administration.

About the EPI

The index presented here is based on on the U.S. presidential and mid-term elections ranging from 2008–2016. For more information on the EPI indicators, what they measure, and how they were developed, explore this site using the tabs above or read through the comprehensive methodology below.

The EPI was first launched in 2013 by the Pew Charitable Trusts, following close consultation and development with an advisory group of leading election officials and academic experts. The initial edition of the index summarized data from the 2008 and 2010 elections; a second iteration followed the year after with data from 2012, and a third edition for the 2014 election was added in 2016.

In 2017, management and development of the EPI passed to the MIT Election Data and Science Lab, which is dedicated to the nonpartisan application of scientific principles to election research and administration. The Lab is directed by Charles Stewart III, PhD., the Kenan Sahin distinguished professor of political science at the Massachusetts Institute of Technology and a leading voice in the early formation of the index.

For more information on how the EPI has developed and changed over the years, and for a closer look at how its calculations and rankings are updated, please refer to the index methodology linked below. For more information on the MIT Election Data and Science Lab, please visit our website.

The EPI was made possible through the generosity of the Pew Charitable Trusts, the William and Flora Hewlett Foundation, Democracy Fund, and the provost of MIT.


Download the report detailing the EPI’s methodology and background.

Read Methodology

Download Data

Download the full data and code needed to replicate the EPI.

Download Data

Election Overview

Take a bird’s-eye view to see how states performed in the past five national elections. Use the tabs at the top of the graph to compare states’ performances between presidential or midterm elections and see how they've changed relative to the average. Use all 17 EPI indicators for a comprehensive look, or select one or a few indicators of particular interest. Likewise, keep all 50 states (and DC) selected to compare them all between years, or choose a few for a narrower look at a region or group of interest to you.

In 2016, one interesting development to note was that almost every state’s index score was impacted by an increase in the residual vote rate. This was largely due to abstentions in the presidential race, rather than states’ election administration decisions. For more information, see the indicators page.

Compare Elections

EPI Rank

How did each state perform in the last national election? What about the last five elections? This page ranks all 50 states and DC according to their scores in the 17 objective indicators used by the EPI, which measure how well elections are being administered around the country. Select individual election years at the top of the graph for a closer look at each state's performance that year, or compare across election years to see how states' index scores have shifted over time. Compare states according to their performance on one indicator, a few, or all of them.

Since the EPI was first developed, an increasing number of states have begun to distribute ballots by mail to all registered voters. Those states are currently Colorado, Oregon, and Washington. For a closer look at these states, you may wish to uncheck the "mail ballots unreturned" indicator.

Note also that in the comparison view, some states may drop in rank between years while still improving their actual indicator scores, thus keeping a green "Improved" rating despite the apparent drop. For more information on the indicators and how they're calculated, take a look at the full methodology report.

Election Years

Compare Elections

  • Improved
  • No change
  • Worsened

State Profiles

How did each state perform in the past five national elections? This page allows you to take a deep dive into the rankings and indicators for each state (plus DC), along with some of the general voting conditions in that state.

Election Years

Compare Elections

Select a state to view its EPI performance

Select a state to the left to explore its overall EPI average and its performance on individual indicators for the national elections between 2008–2016. To explore changes over time, compare across elections by type (e.g., presidential or midterm elections) within each state’s page to display the changes for the state’s indicators between years.

  • %

    Overall EPI Average2012 Rank #1

  • % +1%

    Provisional Ballots Rejected

  • % +1%

    Absentee Ballots Unreturned

  • % +1%

    Lookup Tools

Note: This content scrolls horizontally on small screens.


Nationwide average
Worsened No change Improved
Trouble Printing dialog.


See how states performed in the past five national elections: explore each indicator to see how well states performed or where they ran into challenges for each election year. Each indicator is also accompanied by a brief explanation of what it measures, how that impacts a state’s election performance, and any interesting details about the issue it looks at or the calculations used to measure it.

Election Years

Select an indicator to view its profile

Select an indicator on the left to learn more about what it measures and how it was incorporated in the calculations for the Elections Performance Index. To see how performance on each indicator has changed across elections, use the tabs above to explore year by year.

Best Worst

Using data from the U.S. Election Assistance Commission’s Election Administration and Voting Survey, this indicator measures the percentage—out of all ballots cast—of absentee ballots that were rejected.

In 1988, only six states allowed individuals to cast an absentee ballot by mail without an excuse. By the 2008 election, that number had grown to 27.

Absentee balloting allows voters to cast a ballot at their convenience without having to go to the polls on Election Day, while also spreading election officials’ workload over a longer voting period.

Unlike in-person voters, however, who may have an opportunity to correct errors, absentee voters have no recourse to fix a ballot if a mistake is made. If there is an errant mark and a machine cannot read the ballot, there is a greater chance that the ballot will not be counted.

States have adopted absentee balloting in different ways:

  • Limited Absentee States require a registered voter to provide a reason when requesting an absentee ballot (e.g., illness, disability, travel, etc.). Such states typically have lower rates of absentee ballots rejected compared with No-excuse Absentee States.
  • No-excuse Absentee States allow any registered voter to request an absentee ballot without providing a reason. Such states typically have higher rates of absentee ballots rejected than do Limited Absentee States.
  • Permanent Absentee States are a subset of No-excuse Absentee States that also give any registered voter the option to receive automatically mailed absentee ballots for future elections. Such states tend to have higher rejected absentee ballots than do other states.
  • Vote-by-Mail States conduct elections entirely by mail.

State Comparison

See how states performed in the past five national elections. Use the list below to choose any number of states or regions to compare their performance during a specific election year, or across two years. To explore differences between groups of states with specific voting policies, select options from “Online Registration,” “Mail Ballot Policy,” or “Election Day Registration.”

Election Years

Compare Elections

Select individual states or a predefined group to compare

Compare the performance of individual states, regions, and groups of states on each indicator during election years. Look at each election separately, or select the “compare elections” option to see how states’ indicator values changed between the midterm elections (2014 vs. 2010), or between two presidential elections

Note: This content scrolls horizontally on small screens.

Change 20102014

Nationwide average
Best Worst
Worsened No change Improved