Get more FiveThirtyEight
- Terms of Use
- Privacy Policy
- Your California Privacy Rights
- Children's Online Privacy Policy
- Interest-Based Ads
© 2018 ABC News Internet Ventures. All rights reserved.
Chance the Democrat wins (99.8%)
Chance the Republican wins (0.2%)
The Classic version of our model projects a race’s outcome by taking a weighted average of polls of the race (if available), polls of similar races (CANTOR) and non-polling factors (fundamentals). It is then reverted toward a mean based on long-term trends in midterms and presidential approval ratings.
D+23.2
Adjusted polls
D+29.0
CANTOR
D+35.5
Fundamentals
D+27.1
Experts
D+23.2
Lite
D+26.1
Classic
D+26.4
Deluxe
<0.1
<0.1
Historical adjustment
Key
More weight
Less
We've collected 10 polls for the Massachusetts Senate election. We’re adjusting poll results in three ways: Polls of registered voters or all adults are adjusted to a likely-voter basis; older polls are adjusted based on shifts in the generic congressional ballot since the poll was conducted; and polls are adjusted for house effects (the tendency for a firm’s polls to lean toward Democrats or Republicans). Polls with larger sample sizes and those conducted by higher-quality polling agencies are given more weight, as are more recent polls.
Adjustments | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
dates | pollster | sample | weight | Warren | Diehl | margin | likely voter | Time-line | House effects | Adjusted margin | |
Oct 25-28 10/25-28 | MassINC Polling Group | 502 | LV | 1.54 | 54% | 32% | D+22.0 | — | <0.1 | 1.4 | D+20.6 |
Oct 24-28 10/24-28 | Suffolk University | 500 | LV | 1.35 | 56% | 34% | D+21.2 | — | <0.1 | 0.3 | D+20.9 |
Oct 10-27 10/10-27 | Western New England University | 402 | LV | 0.82 | 57% | 27% | D+30.0 | — | <0.1 | 1.0 | D+29.0 |
Oct 1-7 10/1-7 | University of Massachusetts Lowell | 485 | LV | 0.42 | 56% | 31% | D+25.0 | — | <0.1 | 0.8 | D+25.8 |
Sep 17-21 9/17-21 | MassINC Polling Group | 506 | LV | 0.04 | 56% | 30% | D+26.0 | — | 0.1 | 1.4 | D+24.5 |
Sep 13-17 9/13-17 | Suffolk University | 500 | LV | 0.03 | 54% | 24% | D+29.4 | — | 0.1 | 0.3 | D+28.9 |
Jun 8-12 6/8-12 | Suffolk University | 500 | LV | 0.00 | 55% | 33% | D+22.4 | — | 1.0 | 0.3 | D+23.1 |
May 22-26 5/22-26 | MassINC Polling Group | 501 | RV | 0.00 | 54% | 19% | D+35.0 | 0.2 | 1.1 | 1.4 | D+34.6 |
Mar 16-18 3/16-18 | MassINC Polling Group | 504 | RV | 0.00 | 55% | 20% | D+35.0 | 0.2 | 0.9 | 1.4 | D+34.3 |
Nov 9-12 11/9-12 | MassINC Polling Group | 503 | RV | 0.00 | 58% | 32% | D+26.0 | 0.2 | 0.5 | 1.4 | D+24.0 |
Our state similarity scores are based on demographic, geographic and political characteristics; if two states have a score of 100, it means they are perfectly identical. These scores inform a system we use — CANTOR, or Congressional Algorithm using Neighboring Typologies to Optimize Regression — to infer what polling would say in unpolled or lightly polled states, given what it says in similar states.
Sim. score | Polling avg. | ||
---|---|---|---|
Rhode Island | 74 | D+22.2 | |
Connecticut | 70 | D+18.8 | |
New Jersey | 63 | D+10.1 | |
New York | 62 | D+24.7 | |
Maryland | 50 | D+33.6 | |
Delaware | 38 | D+20.8 | |
Illinois | 38 | — | |
New Hampshire | 30 | — | |
Vermont | 27 | D+39.7 | |
Pennsylvania | 26 | D+11.7 |
The Deluxe version of our model calculates an implied margin for each race based on expert race ratings from The Cook Political Report, Inside Elections and Sabato's Crystal Ball; it then adjusts that margin toward its estimate of the national political environment.
Equivalent Margin | |||
---|---|---|---|
Expert | rating Rating | Raw | Adjusted |
Cook Political Report | D+26.7 | D+27.3 | |
Inside Elections | D+26.7 | D+27.6 | |
Sabato's Crystal Ball | D+26.7 | D+26.3 | |
Average | D+26.7 | D+27.1 |
Nate Silver explains the methodology behind our 2018 midterm forecasts. Read more …
Comments