Post Mortem for the 2016 Presidential, Senate, and House Elections: December 21, 2016 |
I have put below a D or R for the actual outcome for each of the three
elections. I added Wisconsin to
the Presidential election data, which had such a high probability of the
Democrats winning that it was not included in the description.
Remember that all the probability data are from PredictWise and are
betting-market probabilities except for the House elections, where only
25 probabilities are from betting markets---in this case all from Predictit.
For the Presidential election, the Republicans won Wisconsin, and so according to the ranking assumption, they should have won all the states below it. They in fact lost Nevada, Colorado, and New Hampshire. So three errors for the ranking assumption. Ignoring Wisconsin, there was one error---the Democrats should not have won New Hampshire given that they lost Pennsylvania and Michigan. Note, however, that the ranking assumption did better than the betting markets. The Republicans won five elections that they were not favored to win: Wisconsin, Pennslyvania, Michigan, Florida, and North Carolina. For the first three the probabilities were 90 percent or above of the Republicans losing. For the Senate election, the Republicans won Wisconsin, and so according to the ranking assumption, they should have won all the states below it. They in fact lost Nevada and New Hampshire. So two errors for the ranking assumption. Regarding the betting markets, the Republicans won two elections they were not favored to win: Wisconsin and Pennslyvania, which had probabilities of 84 and 79 percent, respectively, of the Democrats winning. For the House election it is probably best just to focus on the 25 elections for which there are betting-market data. The Republicans won 186 Nebraska 2, and so according to the ranking assumption, they should have won all the elections below it. They in fact lost (using only the betting market data) 191 Florida 7, 196 New Jersey 5, and 198 Illinois 10. So three errors for the ranking assumption. Regarding the betting market, the Republicans won six elections that they were not favored to win: 186 Nebraska 2, 188 Texas 23, 192 Minnesota 2, 195 Virginia 10, 197 California 25, and 200 California 49. (The 0.501 probabilities in the table are really 50/50, and so these are not counted here.) |
Ranking Assumption for 2016 Presidential election |
Background
The paper, Interpreting the Predictive Uncertainty of Elections, Journal of Politics, April 2009, provides an interpretation of the uncertainty that exists on election morning as to who will win. The interpretation is based on the theory that there are a number of possible conditions of nature than can exist on election day, of which one is drawn. Political betting markets provide a way of trying to estimate this uncertainty. (Polling standard errors do not provide estimates of this type of uncertainty. They estimate sample-size uncertainty, which can be driven close to zero with a large enough sample.) This paper also introduces a "ranking assumption," which puts restrictions on the possible conditions of nature that can exist on election day. Take as an example the vote in each state for the Democratic candidate for president. Rank the states by the probability on the day of the election that the candidate wins the state. The ranking assumption says that if the candidate wins state i, he or she wins every state ranked above state i. Given some ranking, the ranking assumption can be tested by simply looking to see after the fact if the candidate won a state ranked lower than one he or she lost. In the paper the assumption was tested for the 2004 and 2008 presidential elections using Intrade probabilities at 6am Eastern standard time on the day of the election. The test is thus a test of the joint hypothesis that the Intrade probabilities on November 7, 2016, are right and the ranking assumption is right. Using the Intrade probabilities, the ranking assumption was perfect in 2004 and off by one in 2008. In 2008 Missouri was ranked above Indiana, and Obama won Indiana and lost Missouri. Both of these elections were very close. Obama won Indiana with 50.477 percent of the two-party vote and lost Missouri with 49.937 percent of the two-party vote. Otherwise, 2008 was perfect. Evidence is also presented in the paper, although this is not a test of the ranking assumption, that Intrade traders used the ranking assumption to price various contracts. 2016 Presidential Election I plan to collect market probabilities by state from PredictWise on various days, the last day being the day before the election, November 7, 2016. The November 7 data will be used to test the ranking assumption. The test is a test of the joint hypothesis that the PredictWise market probabilities are right and the ranking assumption is right. The data as they are collected are presented in Table 1. Data collected from PredictWise on November 7, 2016, 9pm These are the final data that will be used to test the ranking assumption. On this date, all but 10 states had market probabilities on the PredictWise website above 96 percent or below 10 percent. These states can be ignored. The 10 states, ranked by market probabilities for the Democratic candidate, are: |
state | prob | votes | actual | sumvotes | |
WI | 96+ | 10 | R | 224 | |
NV | 95 | 6 | D | 230 | |
CO | 94 | 9 | D | 239 | |
PA | 94 | 20 | R | 259 | |
MI | 90 | 16 | R | 275 | pivot |
NH | 89 | 4 | D | ||
FL | 77 | 29 | R | ||
NC | 61 | 15 | R | ||
OH | 32 | 18 | R | ||
AZ | 24 | 11 | R | ||
IA | 15 | 6 | R | ||
89 = PredictWise market-based probability that Clinton
wins the presidential election.
"sumvotes" is the sum of the electoral votes of all the states ranked above the state plus the state's vote. 270 votes are needed to win. Michigan is the pivot state, with a probablility of 90. All the states above Nevada in the ranking have probabilities greater than 95 (these are not shown). If Clinton takes Michigan and all the states ranked above it, she gets 275 votes. Of the states ranked below Michigan, she could also win by not taking Michigan but taking, say, Florida. This would, of course, violate the ranking assumption. There is evidence that traders are essentially using the ranking assumption. According to the ranking assumption, the probability that Clinton wins overall is the probability that she wins the pivot state, Michigan, which is 90. On the PredictWise website, the market-based probability that Clinton wins the presidential election (in the electoral college) is 89, close to 90. Table 1 gives more details. |
Ranking Assumption for 2016 Senate elections |
Background
The ranking assumption can also be tested using data from the Senate elections. Consier the vote in each state for the Democratic candidate for Senate. Rank the states by the probability that the Democratic candidate wins the state. The ranking assumption says that if the Democrats win state i, they win every state ranked above state i. The ranking assumption can thus be tested by simply looking to see after the fact if the Democrats won a state ranked lower than one they lost (contrary to the ranking assumption). I plan to collect market probabilities from PredictWise on various days, the last day being the day before the election, November 7, 2016. The November 7 data will be used to test the ranking assumption. Again, the test is a test of the joint hypothesis that the PredictWise market probabilities are right and the ranking assumption is right. Data collected from PredictWise on November 7, 2016, 9pm
These are the final data that will be used to test the ranking assumption.
The current make up of the Senate is 54 Republicans and 46 Democrats, counting
the 2 independents as Democrats.
Only 10 states are in play in having market probabilities on the PredictWise
website less than 97 percent
and greater than 10 percent. The 10 states ranked by the market probabilities
from PredictWise for the Democratic candidate are:
|
state | prob | now | actual | |
IL | 96 | R | D | |
WI | 84 | R | R | |
PA | 79 | R | R | |
NV | 77 | D | D | |
NH | 57 | R | D | |
MO | 36 | R | R | pivot for 51 Democratic seats |
NC | 34 | R | R | |
IN | 29 | R | R | |
FL | 11 | R | R | |
LA | 13 | R | R | |
67 = PredictWise market-based probability of the Democrats controlling the
Senate.
If the Democrats win Missouri and all the states above it, they will have 51 seats, counting the 2 independents as Democrats. If they lose Missouri but win New Hampshire and all the states above it, they will have 50 seats. In this case they will still control the Senate if they win the White House. According to the ranking assumption, the PredictWise market probablility that they control the Senate is thus 57 percent (the probablity they win New Hampshire) if they win the White House and 36 percent (the probablility they win Missouri) if they don't. If we use the PredictWise market-based probabilty of 89 percent that Clinton wins the presidential electon, then the ranking assumption probability that the Democrats control the Senate is 0.89x57 + 0.11x36, which is 55.7 percent. This is considerably lower than the PredictWise market-based probability of the Democrats controlling the Senate of 67 percent. |
Ranking Assumption for 2016 House elections |
Background
The ranking assumption can also be tested using data from the House elections. Consider the vote in each of the 435 districts for the Democratic candidate for the House. Rank the districts by the probability that the Democratic candidate wins the district. The ranking assumption says that if the Democrats win district i, they win every district ranked above district i. The ranking assumption can thus be tested by simply looking to see after the fact if the Democrats won a district ranked lower than one they lost (contrary to the ranking assumption). The data for the House are not as good as the data for the Senate. From PredictWise one can get a ranking of the districts, but only 25 districts have betting markets. The probabilities for the other districts are just the views of PredictWise. The 25 betting markets are from Predictit. What I have done is to take the 25 probabilites from Predictit and the remaining probabilites from PredictWise. The final data were collected on November 7, 2016, 9pm. They are as follows. When the probability ends in 1, this is a Predictit probability; otherwise it is a PredictWise probability. (The Predictit probabilities are in fact only two digits; the 1 is just for presentation purposes.) Rank District Probability Actual 1 AL - Alabama 7 1.000 2 AZ - Arizona 3 1.000 3 AZ - Arizona 7 1.000 4 CA - California 2 1.000 5 CA - California 5 1.000 6 CA - California 6 1.000 7 CA - California 11 1.000 8 CA - California 12 1.000 9 CA - California 13 1.000 10 WI - Wisconsin 4 1.000 11 WI - Wisconsin 3 1.000 12 WI - Wisconsin 2 1.000 13 CA - California 14 1.000 14 CA - California 15 1.000 15 WA - Washington 9 1.000 16 CA - California 17 1.000 17 WA - Washington 7 1.000 18 WA - Washington 6 1.000 19 CA - California 18 1.000 20 CA - California 19 1.000 21 CA - California 20 1.000 22 WA - Washington 2 1.000 23 CA - California 27 1.000 24 VT - Vermont 1 1.000 25 VA - Virginia 11 1.000 26 CA - California 28 1.000 27 CA - California 29 1.000 28 VA - Virginia 8 1.000 29 CA - California 30 1.000 30 CA - California 32 1.000 31 CA - California 33 1.000 32 CA - California 34 1.000 33 VA - Virginia 3 1.000 34 CA - California 35 1.000 35 CA - California 37 1.000 36 CA - California 38 1.000 37 CA - California 40 1.000 38 CA - California 41 1.000 39 CA - California 43 1.000 40 CA - California 44 1.000 41 TX - Texas 35 1.000 42 TX - Texas 34 1.000 43 TX - Texas 33 1.000 44 CA - California 46 1.000 45 CA - California 47 1.000 46 TX - Texas 30 1.000 47 TX - Texas 29 1.000 48 TX - Texas 28 1.000 49 CA - California 51 1.000 50 CA - California 53 1.000 51 CO - Colorado 1 1.000 52 CO - Colorado 2 1.000 53 CT - Connecticut 1 1.000 54 CT - Connecticut 2 1.000 55 CT - Connecticut 3 1.000 56 TX - Texas 20 1.000 57 DE - Delaware 1 1.000 58 TX - Texas 18 1.000 59 FL - Florida 5 1.000 60 TX - Texas 16 1.000 61 FL - Florida 9 1.000 62 FL - Florida 14 1.000 63 FL - Florida 20 1.000 64 FL - Florida 21 1.000 65 FL - Florida 23 1.000 66 FL - Florida 24 1.000 67 TX - Texas 9 1.000 68 GA - Georgia 2 1.000 69 GA - Georgia 4 1.000 70 GA - Georgia 5 1.000 71 GA - Georgia 13 1.000 72 HI - Hawaii 1 1.000 73 HI - Hawaii 2 1.000 74 IL - Illinois 1 1.000 75 IL - Illinois 2 1.000 76 TN - Tennessee 9 1.000 77 IL - Illinois 3 1.000 78 IL - Illinois 4 1.000 79 IL - Illinois 5 1.000 80 TN - Tennessee 5 1.000 81 IL - Illinois 7 1.000 82 IL - Illinois 9 1.000 83 IL - Illinois 11 1.000 84 IL - Illinois 17 1.000 85 IN - Indiana 1 1.000 86 IN - Indiana 7 1.000 87 SC - South Carolina 6 1.000 88 KY - Kentucky 3 1.000 89 LA - Louisiana 2 1.000 90 MA - Massachusetts 1 1.000 91 MA - Massachusetts 2 1.000 92 MA - Massachusetts 3 1.000 93 RI - Rhode Island 2 1.000 94 RI - Rhode Island 1 1.000 95 MA - Massachusetts 4 1.000 96 MA - Massachusetts 5 1.000 97 MA - Massachusetts 7 1.000 98 MA - Massachusetts 8 1.000 99 PA - Pennsylvania 14 1.000 100 PA - Pennsylvania 13 1.000 101 MD - Maryland 2 1.000 102 MD - Maryland 3 1.000 103 MD - Maryland 4 1.000 104 MD - Maryland 5 1.000 105 MD - Maryland 7 1.000 106 MD - Maryland 8 1.000 107 ME - Maine 1 1.000 108 MI - Michigan 5 1.000 109 MI - Michigan 9 1.000 110 MI - Michigan 12 1.000 111 PA - Pennsylvania 2 1.000 112 PA - Pennsylvania 1 1.000 113 MI - Michigan 13 1.000 114 MI - Michigan 14 1.000 115 OR - Oregon 3 1.000 116 MN - Minnesota 4 1.000 117 OR - Oregon 1 1.000 118 MN - Minnesota 5 1.000 119 MO - Missouri 1 1.000 120 MO - Missouri 5 1.000 121 MS - Mississippi 2 1.000 122 NC - North Carolina 1 1.000 123 NC - North Carolina 4 1.000 124 NC - North Carolina 12 1.000 125 NJ - New Jersey 1 1.000 126 OH - Ohio 13 1.000 127 NJ - New Jersey 6 1.000 128 OH - Ohio 11 1.000 129 NJ - New Jersey 8 1.000 130 OH - Ohio 9 1.000 131 NJ - New Jersey 9 1.000 132 NJ - New Jersey 10 1.000 133 NJ - New Jersey 12 1.000 134 NM - New Mexico 1 1.000 135 NM - New Mexico 3 1.000 136 OH - Ohio 3 1.000 137 NV - Nevada 1 1.000 138 NY - New York 5 1.000 139 NY - New York 6 1.000 140 NY - New York 26 1.000 141 NY - New York 7 1.000 142 NY - New York 8 1.000 143 NY - New York 9 1.000 144 NY - New York 10 1.000 145 NY - New York 12 1.000 146 NY - New York 20 1.000 147 NY - New York 13 1.000 148 NY - New York 14 1.000 149 NY - New York 15 1.000 150 NY - New York 16 1.000 151 CA - California 9 0.990 152 NY - New York 17 0.990 153 WA - Washington 10 0.990 154 CA - California 16 0.990 155 WA - Washington 1 0.990 156 CA - California 31 0.990 157 CO - Colorado 7 0.990 158 CT - Connecticut 4 0.990 159 CT - Connecticut 5 0.990 160 TX - Texas 15 0.990 161 FL - Florida 22 0.990 162 NY - New York 4 0.990 163 IA - Iowa 2 0.990 164 IL - Illinois 8 0.990 165 PA - Pennsylvania 17 0.990 166 MA - Massachusetts 6 0.990 167 MA - Massachusetts 9 0.990 168 OR - Oregon 5 0.990 169 OR - Oregon 4 0.990 170 NH - New Hampshire 2 0.990 171 AZ - Arizona 9 0.980 172 CA - California 3 0.980 173 CA - California 26 0.980 174 CA - California 52 0.980 175 CA - California 36 0.970 176 MN - Minnesota 1 0.970 177 NY - New York 18 0.960 178 NV - Nevada 4 0.940 179 MN - Minnesota 7 0.920 180 NY - New York 25 0.910 181 CA - California 7 0.881 betting 182 VA - Virginia 4 0.880 183 MD - Maryland 6 0.870 184 NH - New Hampshire 1 0.870 185 FL - Florida 10 0.840 D and all above 186 NE - Nebraska 2 0.801 betting R 187 FL - Florida 13 0.80 D 188 TX - Texas 23 0.781 betting R 189 IA - Iowa 1 0.770 R 190 CA - California 24 0.750 D 191 FL - Florida 7 0.711 betting D 192 MN - Minnesota 2 0.691 betting R 193 NY - New York 3 0.680 D 194 NV - Nevada 3 0.590 D 195 VA - Virginia 10 0.571 betting R 196 NJ - New Jersey 5 0.561 betting D 197 CA - California 25 0.551 betting R 198 IL - Illinois 10 0.551 betting D 199 MN - Minnesota 8 0.530 D 200 CA - California 49 0.521 betting R 201 CA - California 10 0.501 betting R 202 CO - Colorado 6 0.501 betting R 203 FL - Florida 26 0.501 betting R 204 AZ - Arizona 1 0.490 D 205 CA - California 21 0.430 R and all below 206 NY - New York 19 0.411 betting 207 FL - Florida 18 0.371 betting 208 PA - Pennsylvania 8 0.371 betting 209 ME - Maine 2 0.341 betting 210 MI - Michigan 8 0.320 211 AZ - Arizona 2 0.300 212 IL - Illinois 12 0.290 213 MI - Michigan 1 0.271 betting 214 PA - Pennsylvania 16 0.261 betting 215 PA - Pennsylvania 6 0.230 216 MN - Minnesota 3 0.220 217 VA - Virginia 5 0.200 218 NY - New York 1 0.191 betting, pivot for Democratic control 219 NY - New York 21 0.190 220 MI - Michigan 7 0.190 221 WI - Wisconsin 8 0.161 betting 222 NY - New York 22 0.161 betting 223 KS - Kansas 3 0.160 224 IN - Indiana 9 0.150 225 UT - Utah 4 0.111 betting 226 IA - Iowa 3 0.111 betting 227 NY - New York 23 0.110 228 VA - Virginia 2 0.100 229 MT - Montana 1 0.100 230 AK - Alaska 1 0.090 231 CO - Colorado 3 0.090 232 IN - Indiana 2 0.080 233 NY - New York 24 0.051 betting 234 NJ - New Jersey 3 0.050 235 NY - New York 11 0.040 236 NC - North Carolina 13 0.040 237 IL - Illinois 13 0.040 238 MI - Michigan 6 0.040 239 NJ - New Jersey 2 0.030 240 FL - Florida 2 0.030 241 WI - Wisconsin 7 0.020 242 PA - Pennsylvania 7 0.020 243 NC - North Carolina 2 0.020 244 WI - Wisconsin 6 0.020 245 WA - Washington 8 0.020 246 MI - Michigan 3 0.020 247 WA - Washington 3 0.020 248 MI - Michigan 11 0.020 249 AR - Arkansas 2 0.010 250 PA - Pennsylvania 12 0.010 251 WV - West Virginia 2 0.010 252 MI - Michigan 4 0.010 253 NJ - New Jersey 7 0.010 254 OH - Ohio 10 0.010 255 WI - Wisconsin 1 0.010 256 WA - Washington 5 0.010 257 NJ - New Jersey 11 0.010 258 MI - Michigan 10 0.010 259 OH - Ohio 6 0.010 260 OH - Ohio 16 0.010 261 OH - Ohio 14 0.010 262 NM - New Mexico 2 0.010 263 FL - Florida 15 0.010 264 FL - Florida 16 0.010 265 GA - Georgia 12 0.010 266 NY - New York 2 0.010 267 OH - Ohio 1 0.010 268 IA - Iowa 4 0.010 269 IL - Illinois 6 0.010 270 SC - South Carolina 7 0.010 271 SC - South Carolina 5 0.010 272 KS - Kansas 2 0.010 273 MO - Missouri 6 0.000 274 OR - Oregon 2 0.000 275 IN - Indiana 8 0.000 276 KS - Kansas 1 0.000 277 IN - Indiana 6 0.000 278 IN - Indiana 5 0.000 279 IN - Indiana 4 0.000 280 IN - Indiana 3 0.000 281 MO - Missouri 7 0.000 282 OH - Ohio 8 0.000 283 IL - Illinois 18 0.000 284 SD - South Dakota 1 0.000 285 IL - Illinois 16 0.000 286 IL - Illinois 15 0.000 287 IL - Illinois 14 0.000 288 MO - Missouri 3 0.000 289 MO - Missouri 4 0.000 290 TN - Tennessee 1 0.000 291 NC - North Carolina 5 0.000 292 TN - Tennessee 2 0.000 293 MN - Minnesota 6 0.000 294 TN - Tennessee 3 0.000 295 KS - Kansas 4 0.000 296 TN - Tennessee 4 0.000 297 TN - Tennessee 6 0.000 298 TN - Tennessee 7 0.000 299 TN - Tennessee 8 0.000 300 TX - Texas 1 0.000 301 ID - Idaho 2 0.000 302 ID - Idaho 1 0.000 303 KY - Kentucky 1 0.000 304 KY - Kentucky 2 0.000 305 SC - South Carolina 4 0.000 306 ND - North Dakota 1 0.000 307 TX - Texas 2 0.000 308 TX - Texas 3 0.000 309 GA - Georgia 14 0.000 310 TX - Texas 4 0.000 311 KY - Kentucky 4 0.000 312 GA - Georgia 11 0.000 313 GA - Georgia 10 0.000 314 GA - Georgia 9 0.000 315 GA - Georgia 8 0.000 316 GA - Georgia 7 0.000 317 GA - Georgia 6 0.000 318 TX - Texas 5 0.000 319 TX - Texas 6 0.000 320 GA - Georgia 3 0.000 321 TX - Texas 7 0.000 322 GA - Georgia 1 0.000 323 FL - Florida 27 0.000 324 PA - Pennsylvania 3 0.000 325 FL - Florida 25 0.000 326 TX - Texas 8 0.000 327 TX - Texas 10 0.000 328 NC - North Carolina 10 0.000 329 TX - Texas 11 0.000 330 TX - Texas 12 0.000 331 FL - Florida 19 0.000 332 MS - Mississippi 1 0.000 333 FL - Florida 17 0.000 334 KY - Kentucky 5 0.000 335 KY - Kentucky 6 0.000 336 TX - Texas 13 0.000 337 NE - Nebraska 3 0.000 338 FL - Florida 12 0.000 339 FL - Florida 11 0.000 340 LA - Louisiana 1 0.000 341 TX - Texas 14 0.000 342 FL - Florida 8 0.000 343 SC - South Carolina 3 0.000 344 FL - Florida 6 0.000 345 NY - New York 27 0.000 346 FL - Florida 4 0.000 347 FL - Florida 3 0.000 348 OK - Oklahoma 3 0.000 349 FL - Florida 1 0.000 350 TX - Texas 17 0.000 351 NJ - New Jersey 4 0.000 352 PA - Pennsylvania 4 0.000 353 TX - Texas 19 0.000 354 TX - Texas 21 0.000 355 TX - Texas 22 0.000 356 NV - Nevada 2 0.000 357 LA - Louisiana 3 0.000 358 CO - Colorado 5 0.000 359 CO - Colorado 4 0.000 360 OK - Oklahoma 5 0.000 361 NE - Nebraska 1 0.000 362 TX - Texas 24 0.000 363 TX - Texas 25 0.000 364 LA - Louisiana 4 0.000 365 TX - Texas 26 0.000 366 CA - California 50 0.000 367 NC - North Carolina 3 0.000 368 CA - California 48 0.000 369 TX - Texas 27 0.000 370 TX - Texas 31 0.000 371 CA - California 45 0.000 372 TX - Texas 32 0.000 373 TX - Texas 36 0.000 374 CA - California 42 0.000 375 UT - Utah 1 0.000 376 UT - Utah 2 0.000 377 CA - California 39 0.000 378 UT - Utah 3 0.000 379 OH - Ohio 2 0.000 380 OH - Ohio 7 0.000 381 VA - Virginia 1 0.000 382 LA - Louisiana 5 0.000 383 PA - Pennsylvania 5 0.000 384 NC - North Carolina 7 0.000 385 NC - North Carolina 11 0.000 386 VA - Virginia 6 0.000 387 VA - Virginia 7 0.000 388 VA - Virginia 9 0.000 389 NC - North Carolina 8 0.000 390 LA - Louisiana 6 0.000 391 NC - North Carolina 6 0.000 392 MI - Michigan 2 0.000 393 CA - California 23 0.000 394 CA - California 22 0.000 395 MS - Mississippi 3 0.000 396 OK - Oklahoma 4 0.000 397 MD - Maryland 1 0.000 398 WA - Washington 4 0.000 399 SC - South Carolina 2 0.000 400 PA - Pennsylvania 11 0.000 401 PA - Pennsylvania 10 0.000 402 OK - Oklahoma 1 0.000 403 SC - South Carolina 1 0.000 404 WI - Wisconsin 5 0.000 405 PA - Pennsylvania 9 0.000 406 MO - Missouri 8 0.000 407 PA - Pennsylvania 18 0.000 408 CA - California 8 0.000 409 MO - Missouri 2 0.000 410 OH - Ohio 12 0.000 411 OH - Ohio 15 0.000 412 CA - California 4 0.000 413 OH - Ohio 5 0.000 414 WV - West Virginia 1 0.000 415 CA - California 1 0.000 416 OH - Ohio 4 0.000 417 AZ - Arizona 8 0.000 418 NC - North Carolina 9 0.000 419 AZ - Arizona 6 0.000 420 AZ - Arizona 5 0.000 421 AZ - Arizona 4 0.000 422 WV - West Virginia 3 0.000 423 OK - Oklahoma 2 0.000 424 MS - Mississippi 4 0.000 425 AR - Arkansas 4 0.000 426 AR - Arkansas 3 0.000 427 PA - Pennsylvania 15 0.000 428 AR - Arkansas 1 0.000 429 WY - Wyoming 1 0.000 430 AL - Alabama 6 0.000 431 AL - Alabama 5 0.000 432 AL - Alabama 4 0.000 433 AL - Alabama 3 0.000 434 AL - Alabama 2 0.000 435 AL - Alabama 1 0.000You can see that the pivot is the 1st district of New York with a probability of 0.19. If the Democrats win this district and all the districts above it, they have control of the House. According to the ranking assumption, the probabiity of this happening is 0.19. On the Predictit website the betting probability that the Democrats get 218 or more seats was 0.065 (at 9pm, November 7, 2016). This compares to 0.19 using the ranking assumption. |
|