Replies

Using the expert ratings above, the results of 32,000 simulated elections are listed in the table below. The definitions of the columns are:

"GOP House Seats P10” – the value where the probability of the actual number being lower is 10%.
”GOP House Seats EV” – the expected value (mean) or probability-weighted average of the simulation.
”GOP House Seats P90” – the value where the probability of the actual number being lower is 90%. Or, it is the value where the probability of the actual number being higher is 10%.
”Probability of 218” – The probability associated with value 218 (the House majority goal).
”GOP Gain” – The difference between the GOP EV and the current number of seats (179), rounded down.

Week	GOP House Seats P10	GOP House Seats EV	GOP House Seats P90	Probability of 218	GOP Gain
09-Apr-10	199	204.77	210	0.18%	26
16-Apr-10	199	205.09	211	0.22%	26
22-Apr-10	200	205.28	211	0.37%	26
01-May-10	201	206.22	212	0.73%	27
08-May-10	201	206.33	212	0.66%	27
19-Jun-10	203	208.44	214	2.29%	29
10-Jul-10	203	208.49	214	2.43%	29
17-Jul-10	203	208.49	214	2.34%	29
24-Jul-10	203	209.25	215	3.37%	30
31-Jul-10	203	209.06	215	2.96%	30
07-Aug-10	203	209.15	215	3.11%	30
14-Aug-10	204	209.33	215	3.33%	30
21-Aug-10	204	209.86	216	4.47%	31
28-Aug-10	205	210.91	217	7.03%	32
04-Sep-10	206	211.88	218	10.45%	33
11-Sep-10	208	213.67	220	20.31%	35
18-Sep-10	208	214.1	220	23.14%	35
25-Sep-10	209	214.72	221	27.54%	36
02-Oct-10	210	216.44	222	41.34%	37
09-Oct-10	212	218.48	225	58.13%	39
16-Oct-10	215	220.78	227	75.42%	42
23-Oct-10	217	223.28	229	88.59%	44
30-Oct-10	220	226.71	233	97.00%	48

Alternatively, below is a model that uses the most recent poll instead of the expert assessment, when there is one. I "expire" the poll if it is greater than 60 days old and revert back to the expert assessment until a new poll is taken. If the trend is moving downward, I suspect it is because a poll expired and the expert asssessment is lagging public opinion.

This week, the polls cover 79% of the 110 races being tracked. 87 polls are being used out of 99 that were captured, for a poll utilization of 88%.

Using the expert ratings above and recent poll data, the results of 32,000 simulated elections are listed in the table below. The definitions of the columns are:

"GOP House Seats P10” – the value where the probability of the actual number being lower is 10%.
”GOP House Seats EV” – the expected value (mean) or probability-weighted average of the simulation.
”GOP House Seats P90” – the value where the probability of the actual number being lower is 90%. Or, it is the value where the probability of the actual number being higher is 10%.
”Probability of 218” – The probability associated with value 218 (the House majority goal).
”GOP Gain” – The difference between the GOP EV and the current number of seats (179), rounded down.

Week	GOP House Seats P10	GOP House Seats EV	GOP House Seats P90	Probability of 218	GOP Gain
31-Jul-10	207	211.93	217	8.43%	33
07-Aug-10	205	210.66	216	4.51%	32
14-Aug-10	205	210.56	216	4.44%	32
21-Aug-10	208	212.88	218	11.83%	34
28-Aug-10	211	216.78	220	42.58%	38
04-Sep-10	213	217.53	222	50.40%	39
11-Sep-10	214	219.05	224	66.39%	40
18-Sep-10	214	218.54	223	62.30%	40
25-Sep-10	214	218.46	223	61.17%	39
02-Oct-10	217	220.83	225	83.86%	42
09-Oct-10	219	223.48	228	96.15%	44
16-Oct-10	222	225.93	230	99.33%	47
23-Oct-10	224	228.38	233	99.94%	49
30-Oct-10	227	231.29	236	100.00%	52

And in the Senate...

This final week has brought little change from recent weeks.

In California, Carly Fiorina continues to trail Barbara Boxer, with Boxer gaining an addition 1% over Fiorina. In Colorado, Ken Buck makes back 2% on his lead over Michael Bennet. In Connecticut, Linda MacMahon falls out of the margin of error again. In Illinois, Mark Kirk sustains his 4% lead over Alexi Giannioulias. In Nevada, Sharron Angle edges ahead of Harry Reid with an additional 1% gain. In New Hampshire, Kelly Ayotte pulls safely beyond the margin of error over Paul Hodes. In Pennsylvania, Pat Toomy sustains his 4% lead over Joe Sestak. In Washington, Dino Rossi returns to a 1% lead over Patty Murray. In West Virginia, John Raese trails Joe Manchin again, giving up a 7% lead. In Wisconsin, Ron Johnson sustains his 7% lead over Russ Feingold.

California 1: -1% (R -32%)
Colorado 2: +2% (R +4%)
Connecticut 1: -8% (R -13%) - drops out of the margin of error
Illinois 2: +0% (R +4%)
Nevada 2: +1% (R +4%)
New Hampshire 1: +8% (R +15%) - drops out of the margin of error
Pennsylvania 2: +0% (R +4%)
Washington 2: +4% (R +1%)
West Virginia 1: -10% (R -3%)
Wisconsin 1: +0% (R +7%)

Using the most recent Rasmussen polls, the results of 32,000 simulated elections are listed in the table below. The definitions of the columns are:

"GOP Senate Seats P10” – the value where the probability of the actual number being lower is 10%.
”GOP Senate Seats EV” – the expected value (mean) or probability-weighted average of the simulation.
”GOP Senate Seats P90” – the value where the probability of the actual number being lower is 90%. Or, it is the value where the probability of the actual number being higher is 10%.
”Probability of 51” – The probability associated with value 51 (the Senate majority goal).
”GOP Gain” – The difference between the GOP EV and the current number of seats (41), rounded down.

Week	GOP Senate Seats P10	GOP Senate Seats EV	GOP Senate Seats P90	Probability of 51	GOP Gain
17-Jul-10	47	48.84	50	6.95%	7
24-Jul-10	47	48.95	50	7.69%	7
31-Jul-10	47	48.18	50	1.29%	7
07-Aug-10	47	48.13	50	1.46%	7
14-Aug-10	47	48.46	50	2.92%	7
21-Aug-10	47	48.69	50	3.05%	7
28-Aug-10	47	48.86	50	5.88%	7
04-Sep-10	48	49.21	51	14.19%	8
11-Sep-10	49	50.08	52	37.67%	9
18-Sep-10	48	48.95	50	3.67%	7
25-Sep-10	47	48.35	49	0.30%	7
02-Oct-10	48	49.7	51	21.47%	8
09-Oct-10	50	50.55	51	58.69%	9
16-Oct-10	49	50.08	51	35.48%	9
23-Oct-10	48	49.7	51	21.70%	8
30-Oct-10	48	49.5	51	14.19%	8

-PJ

Nice trend. Charts and data have been updated on KHR.

Looking at the plot which just keeps going up makes one wonder if the estimate that is projected by the "Experts" plus the polls is still lagging behind the reality on the ground. One would have expected a flattening of the curve as is the case with the Senate estimate.

Are the "Experts" still running well behind the trend?

Another explanation is the possibility that there are just too many Toss-Ups in the mix. A toss-up is simply an "Expert" refusing to go on the record - a cop out so to speak. To their credit both Sabato and Election Projection do not have any un-called races on their lists. In fact Election Projection doesn't even have a Toss-Up category. Sabato eliminated the category in his final update. It will be interesting to see which of them is has the more accurate projection.

Just to be clear, I'm picking on the "Experts" and perhaps the pollsters but not the simulation which I think is excellent. The current projection of 52 seats is amazingly close to NYT/Silver at 53, Rasmussen at 55 and Sabato at 55 and those are well established pundits.

Anyone who wants to pick on this simulation needs to explain how it is so close to the projections by these well known sources.