# Introduction

Every day the partisan composition of the 118th Senate is predicted from the state polls. Here are the results. The top graph includes all states. The second graph includes only those states where one candidate has a lead of 5% or more (roughly the margin of error for state polls). Blue is for Democrats, red is for Republicans. January and February reflect the results of the 2016 election. As head-to-head polls of the ultimate candidates came in (which we only knew with hindsight), they replaced the 2016 election results. This graph counts only the senators; the role of the vice president is not considered. Also, the software happens to draw the red line first and then the blue line, so if the values are the same, the blue line covers the red line. This is just an artifact, but conveniently, with a 50-50 Senate the Democrats do have control due to the vice president.

## Senate Including States Where the Candidates are Statistically Tied in 2022

The graph above shows the Senate score for every day, counting, say, 46% to 45% as a win for the candidate with 46%, even though this is really a statistical tie. The gray vertical lines show the boundaries between the months.

## Senate Excluding States Where the Candidates are Statistically Tied in 2022

The graph above shows the 2022 Senate races again but omits the "barely" states. The senate races only count in this graph if the candidate is ahead by 5% or more (roughly the margin of error). In other words, the states with white centers on the map, which are statistical ties, are omitted in the above graph.

For comparison purposes, here are the same graphs for 2016, the last time the Class 3 senators were up for reelection.