ARIZONA POLLSTERS: Kelly +4.3 - Kelly 48.8, McSally 44.5 REF'S AVG: Kelly +2.5 - Kelly 48.2, McSally 45.7 RETURN TO SENATE RACES 2016 Exit Poll - 28D/40I/32R without leaners (PDF) 2018 Exit Poll - 41D/12I/47R with leaners (PDF) Pollsters' Averages Chart Political Ref's Averages Chart Poll Date Sample + Party Distribution Result Pollster Avg Ref's Avg Data Orbital 10/28-10/30 550LV Kelly 48, McSally 46 x   Political Ref's Take on the above poll No adjustment necessary Included   x Emerson 10/29-10/31 732LV Kelly 50, McSally 48 x   Political Ref's Take on the above poll No adjustment necessary Included   x Rasmussen Reports 10/27-10/29 800LV, Kelly 48, McSally 43 x   Political Ref's Take on the above poll No adjustment necessary Included   x Suffolk Univ 9/26-9/30 500LV, 33D/34I/33R Kelly 49, McSally 40 x   Political Ref's Take on the above poll The even turnout undersamples Republicans. Adjusting to R+4 28D/40I/32R. According to the poll data, Kelly won D's 99-0, I's 39-36 and McSally won R's 86-9. Applying the voter ratios to the more likely turnout, Kelly receives 46.2 to McSally 45.5. Kelly 46.2, McSally 45.5   x Susquehanna Polling & Research Inc. 9/25-9/28 500LV, 35D/25I/40R Kelly 48, McSally 44 x   Political Ref's Take on the above poll Party distribution is reasonable. No adjustment necessary. Included   x ABC/Washington post 9/15-9/20 579LV, 25D/38I/31R Kelly 49, McSally 48 x   Political Ref's Take on the above poll Party distribution is reasonable. No adjustment necessary. Included   x Morning Consult 9/11-9/20 907LV, Party distribution not provided Kelly 49, McSally 40 x   Political Ref's Take on the above poll Neither party distribution nor voter ratios of those parties is provided. Excluded     Fabrizio, Lee & Assoc (GOP) 9/14-9/16 800LV, 34D/24I/39R Kelly 48, McSally 46 x   Political Ref's Take on the above poll The party distribution oversamples Dems and Republicans with too few Independents. Will adjust to the more likely 28D/40I/32R. Voter ratios are not provided but can be inferred. McSally has a 5-point partisan advantage, but trails by 2, suggesting she did 7 points (21%) worse with Democrats and Independents than Kelly did with Republicans and Independents. Will assume Kelly won D's 93-2, I's 50-35 and McSally won R's 90-5. Applying these voter ratios to the more likely turnout, Kelly receives 47.6 to McSally 43.4. Kelly 47.6, McSally 43.4   x New York Times/Sienna 9/10-9/15 653LV, 29D/36I/33R Kelly 50, McSally 42 x   Political Ref's Take on the above poll This poll fails to release the number of registered voters polled, an essential number for sufficient transparency. Excluded     Monmouth Univ 9/11-9/15 420LV, 32D/33I/35R Kelly 49, McSally 48 x   Political Ref's Take on the above poll Party distribution is reasonable. No adjustment necessary. Included   x Kaiser Family Foundation 8/29-9/13 1298RV, 28D/28I/31R Kelly 44, McSally 36 1month   Political Ref's Take on the above poll The sample is reasonable. This is a registered voter poll so a registered voter screen will be applied. Kelly adjusts to 27 and McSally to 23.3. Apply the 27/23.3 ratio to the 80% of respondents choosing one of the candidates, Kelly receives 42.9 to McSally 37.1. Kelly 42.9, McSally 37.1   1month YouGov 9/9-9/11 1102LV, 36D/30I/33R Kelly 49, McSally 42 1month   Political Ref's Take on the above poll The sample oversamples D's and undersamples I's. Will adjust to 28D/40I/32R. According to the poll data, Kelly won D's 92-3, I's 45-41 and McSally won R's 85-7. Applying the voter ratios to the more likely voter turnout, Kelly receives 46.0 to McSally 44.4. Kelly 46.0, McSally 44.4   1month Gravis Marketing 9/10-9/11 684LV, 32D/34I/34R Kelly 48, McSally 43 1month   Political Ref's Take on the above poll The party distribution is a little off, but not too much. This is also a likely voter poll. No adjustment necessary. Included   1month Benenson Strategy Group/GS Strategy Group 8/28-9/8 1600LV, Party distribution not provided Kelly 48, McSally 45 1month   Political Ref's Take on the above poll Neither party distribution nor voter ratios of those parties is provided. Excluded     Change Research 9/4-9/6 470LV, Party distribution not provided Kelly 51, McSally 45 >1month   Political Ref's Take on the above poll Neither party distribution nor voter ratios of those parties is provided. Excluded     Fox News 8/29-9/1 772LV, 46D/11I/43R Kelly 56, McSally 39 >1month   Political Ref's Take on the above poll The D+3 sample badly oversamples Democrats. Will adjust to R+4, or 42D/11I/46R. According to the poll data, Kelly won D's 95-3, I's 57-21 and McSally won R's 82-14. Applying these ratios to the much more likely voter turnout, Kelly receives 52.6 to McSally 41.3. Kelly 52.6, McSally 41.3   >1month Emerson College 8/8-8/10 661LV, 32D/33I/35R Kelly 51, McSally 41 >1month   Political Ref's Take on the above poll Party distribution seems reasonable. No adjustment necessary Included   >1month Change Research 8/7-8/9 428LV, D+4 across all battlegrounds, but party distribution for each state is not provided Kelly 49, McSally 43 Repeat   Political Ref's Take on the above poll Neither voter turnout assumptions or voter ratios within the parties is provided. Impossible to verify credibility of the poll. Excluded     OH Predictive Insights 8/3-8/4 603LV, 34D/26I/40R Kelly 48, McSally 43 >1month   Political Ref's Take on the above poll Party distribution seems reasonable. No adjustment necessary Included   >1month OnMessage Inc. 8/2-8/4 400LV, Party distribution not provided Kelly 48, McSally 48 >1month   Political Ref's Take on the above poll Party distributions are not provided but voter ratios are. Kelly won D's 92-5, I's 51-40 and McSally won R's 90-9. Apply these ratios to the most likely turnout of 28D/40I/32R, Kelly 49 to McSally 44.8   >1month Data for Progress 7/24-8/2 1215LV, Party Distribution Not Provided Kelly 47, McSally 38 >1month   Political Ref's Take on the above poll Assuming a turnout of 28D/40I/32R. According to the poll data, Kelly won D's 91-3, I's 41-24 and McSally won R's 83-8. Applying these ratios to the likely voter turnout, Kelly receives 44.4 to McSally 37.0. Kelly 44.4, McSally 37.0   >1month PPP 7/21-7/22 816RV, 33D/32I/35R Kelly 51, McSally 42 >1month   Political Ref's Take on the above poll The R+2 sample oversamples Democrats a bit, but not by enough to make much of a difference. No adjustment necessary. Included   >1month Spry Strategies 7/11-7/16 700LV, 33D/26I/39R Kelly 48.1, McSally 41.4 >1month   Political Ref's Take on the above poll The R+6 turnout seems to be a bit toward a turnout assumption that factors in some leaners because the Independent column is low. This matches 2018 turnout when leaners were included and comes close to the R+4 in 2016. No adjustment necessary. Included   >1month Change Research 7/10-7/12 345LV, Party Distribution Not available Kelly 52, McSally 45 Repeat   Political Ref's Take on the above poll Neither party distribution or voter ratios within those parties are provided. Adjustment not possible. Excluded from our average. Adjustment not possible. Excluded from our average.     OH Predictive Insights 6/6-6/7 600LV, 34D/26I/39R Kelly 52, McSally 43 >1month   Political Ref's Take on the above poll The party distribution looks about right based on 2016. No adjustment necessary. Included   >1month YouGov 7/7-7/10 1087RV, 34.1D/32.6I/32.7R Kelly 46, McSally 42 >1month   Political Ref's Take on the above poll The D+1.4 sample oversamples Democrats. Will adjust to the more likely 28D/40I/32R voter turnout model we saw in 2016. According to the poll data, Kelly won D's 87-4, McSally won I's 43-39 and R's 81-10. Extrapolating these voter ratios to the more likely voter turnout model, Kelly receives 43.2 to McSally 44.2. Kelly 43.2, McSally 44.2   >1month Global Strategy Group 6/19-6/24 800LV, Party Distribution Not Provided Kelly 49, McSally 42 >1month   Political Ref's Take on the above poll Neither party distribution or voter ratios within the parties is provided. Adjustment not possible. Excluded     Data Orbital 6/27-6/29 600LV, 32.5D/26I/39.5R Kelly 50, McSally43 >1month   Political Ref's Take on the above poll The R+7 sample oversamples Republicans by a bit. It looks like the poll included some leaners as both registered Republicans and Democrats, so I can't adjust fully to 2016 model but a combination of 2016 and 2018. Will adjust to the more likely R+4, assuming 33D/28I/37R. According to the poll data, Kelly won D's 91-4, I's 53-27 and McSally won R's 81-8. Reallocating those vote ratios to the more likely voter turnout, Kelly receives 47.8 to McSally 38.9. Applying the 47.8/38.9 ratio to the 93% of the poll respondents who chose Kelly/McSally, final adjusted result 51.3 Kelly to McSally 41.7. Kelly 51.3, McSally 41.7   >1month Change Research 6/26-6/28 311LV, The overall sample of all battleground states was 39D/27I/34R. Kelly 53, McSally 44 Repeat   Political Ref's Take on the above poll The +5 D sample in battleground states is an oversample of Democrats. Adjustment is impossible because we don't know the party distribution of each state. Excluded from our averages. Adjustment not possible     Gravis Marketing 6/27 527LV, 30D/36I/34R Kelly 42, McSally 46 >1month   Political Ref's Take on the above poll The party distribution matches 2016 almost exactly. No adjustment necessary No adjustment necessary   >1month NYT/Sienna College 6/8-6/16 650RV, 25D/41I/30R Kelly 47, McSally 38 >1Month   Political Ref's Take on the above poll There is a strong chance that Sienna failed to pick up some of the Trump vote among Independents as noted by other poll analysts. The results among Independents are far out of step with other pollsters like YouGov showing a dead heat in the group. Also, this poll includes many respondents it should not. The NYT/Sienna College poll includes many respondents it should not. It includes 13% of Independents, 3% of Democrats and 2% of Republicans who self-label as somewhat likely or not at all likely to vote in Arizona. This is low hanging fruit for a pollster seeking likely voters (accuracy). Likely voter screens should be implemented for this poll to be reliable. Excluded from this average. Excluded from this average     Civiqs 6/13-6/15 1368RV, 34D/31I/35R Kelly 51, McSally 42 >1Month   Political Ref's Take on the above poll The R+1 is off from both the 2016 and 2018 voter turnout distributions. Will adjust to 28D/40I/32R. According to the poll data, Kelly won D's 95-3, I's 52-36 and McSally won R's 85-8. Applying these Kelly/McSally ratios to the more likely voter turnout of 28D/40I/32R results in Kelly receiving 50.0 and McSally 42.5. Kelly 50.0, McSally 42.5   >1Month Fox News 5/30-6/2 1002RV, 43D/14I/43R Kelly 50, McSally 37 >1Month   Political Ref's Take on the above poll While party distribution is provided, the Kelly/McSally voter ratio within each party is required to do a credible adjustment, and those ratios are not provided. No adjustment possible. No adjustment possible. Result excluded from our average.