POLLS: Smith +5.3 - Smith 46.8, Lewis 41.5 REF'S: Smith +4.4 - Smith 46.6, Lewis 42.2 RETURN TO SENATE RACES 2016 Exit Poll - 37D/28I/35R without leaners (PDF) 2018 Exit Poll - 46D/14I/40R with leaners (PDF) Straight Poll Averages Chart Freedom Window Averages Chart Poll Date Sample + Party Distribution Result Pollster Avg Ref's Avg Survey USA 10/23-10/27 649LV, Not provided Smth 45, Lewis 42 x   Political Ref's Take on the above poll No adjustment necessary. Included     YouGov 9/9-9/11 1031LV, 36D/27I/34R Smith 47, Lewis 40 x   Political Ref's Take on the above poll The voter distribution is reasonable. No adjustment necessary. Included   x New York Times/Sienna 9/8-9/10 814LV, 33D/33I/30R Smith 49, Lewis 40 x   Political Ref's Take on the above poll The sample undersamples D's and R's and oversamples I's. Adjusting to 37D/28I/35R. According to the poll data, Smith won D's 94-1, I's 47-35 and Lewis won R's 90-4. Applying the voter ratios to the more likely turnout, Smith receives Smith 49.3, Lewis 41.7   x PPP 9/3-9/4 877RV, 39D/29I/32R Smith 49, Lewis 41 x   Political Ref's Take on the above poll The D+7 sample oversamples Democrats. Adjusting to the more likely D+2 turnout, or 37D/28I/35R. Will also apply a registered voter screen to the I column. Voter ratios are not provided but can be inferred. Smith had a 7-point turnout advantage and had an 8-point lead, which suggests she won Independents by about a point. Will assume Smith won I's 14-13, or 48-45%. Adjusts to 31-32.3 after registered voter screen is applied. Assuming each candidate won his or her party 90-5. After applying these voter ratios to the more likely turnout, Smith receives 43.8 to Lewis 42.4. Smith 43.8, Lewis 42.4   x Harper Polling 8/30-9/1 501LV, Party distribution not provided Smith 43, Lewis 41 x   Political Ref's Take on the above poll Neither party distribution or voter ratios of the parties are provided. Excluded     Emerson College 8/8-8/10 733LV, 38D/27I/36R Smith 48, Lewis 45 x   Political Ref's Take on the above poll The party distribution looks reasonable. No adjustment necessary. Included   x PPP 7/22-7/23 1218RV, 36D/34I/30R Smith 48, Lewis 39 Repeat   Political Ref's Take on the above poll The D+6 sample oversamples Democrats. Will adjust to D+2, or 37D/28I/35R. Will also apply a registered voter screen to the I column. Smith had a +6 partisan advantage and leads by 9, so she must have done about 3 points, or 9%, better with Independents and Republicans than Lewis did with Independents and Democrats. Assuming Smith won D's 92-4, I's 48-44 and Lewis was R's 88-5. After applying the registered voter screen to the column, it changes to 31-31.3. Applying the voter ratios to the more likely voter turnout, Smith receives 43.6 to Lewis 41.0. Smith 43.6, Lewis 41   Repeat