Some of you may remember that in the early morning hours of February 6 (the morning after Super Duper Tuesday) the Obama campaign "inadvertently" attached an excel spreadsheet that contained their projections on the pledged delegate count for the previous night's contests and all remaining contests.
``This is only one of an infinite number of scenarios,'' Obama campaign spokesman Bill Burton said. He added that the information was released unintentionally.
Still I thought it might be fun to revisit those projections three weeks later after eleven more contests...
UPDATE (1:20 PM PST): Thanks to Searching for Pericles I updated the Tennessee numbers giving one more delegate to Clinton (and one less to Obama).
First let's compare how Obama's campaign thought he had done on February 5th with how it turns out they did (positive margins favor Obama, negative margins favor Clinton)...
State | Pledged Total | Actual Obama | Actual Clinton | Actual Margin | Predicted Obama | Predicted Clinton | Predicted Margin | Error |
Alabama | 52 | 27 | 25 | 2 | 29 | 23 | 6 | -4 |
Alaska | 13 | 9 | 4 | 5 | 9 | 4 | 5 | 0 |
Am. Samoa | 3 | 1 | 2 | -1 | 1 | 2 | -1 | 0 |
Arizona | 56 | 25 | 31 | -6 | 25 | 31 | -6 | 0 |
Arkansas | 35 | 8 | 27 | -19 | 8 | 27 | -19 | 0 |
California | 370 | 167 | 203 | -36 | 161 | 209 | -48 | 12 |
Colorado | 55 | 37 | 18 | 19 | 37 | 18 | 19 | 0 |
Connecticut | 48 | 26 | 22 | 4 | 26 | 22 | 4 | 0 |
Delaware | 15 | 9 | 6 | 3 | 9 | 6 | 3 | 0 |
Georgia | 87 | 60 | 27 | 33 | 60 | 27 | 33 | 0 |
Idaho | 18 | 15 | 3 | 12 | 15 | 3 | 12 | 0 |
Illinois | 153 | 104 | 49 | 55 | 101 | 52 | 49 | 6 |
Kansas | 32 | 23 | 9 | 14 | 23 | 9 | 14 | 0 |
Massachusetts | 93 | 38 | 55 | -17 | 37 | 56 | -19 | 2 |
Minnesota | 72 | 48 | 24 | 24 | 48 | 24 | 24 | 0 |
Missouri | 72 | 36 | 36 | 0 | 36 | 36 | 0 | 0 |
New Jersey | 107 | 48 | 59 | -11 | 48 | 59 | -11 | 0 |
New Mexico | 26 | 12 | 14 | -2 | 13 | 13 | 0 | -2 |
New York | 232 | 93 | 139 | -46 | 95 | 137 | -42 | -4 |
North Dakota | 13 | 8 | 5 | 3 | 8 | 5 | 3 | 0 |
Oklahoma | 38 | 14 | 24 | -10 | 14 | 24 | -10 | 0 |
Tennessee | 68 | 28 | 40 | -12 | 29 | 39 | -10 | -2 |
Utah | 23 | 14 | 9 | 5 | 13 | 10 | 3 | 2 |
Total | 1681 | 850 | 831 | 19 | 845 | 836 | 9 | 10 |
Those actual numbers are based on the best information I can find. When possible I've been using official sources for the results. Overall Obama did better than his campaign had initially thought as more votes were counted. In particular he picked up 6 more delegates from Clinton in California (a 12 delegate swing) and 3 more in Illinois (a 6 delegate swing). In New York he lost two delegates (a 4 delegate swing) and in every other state his count appears accurate up to +/- a delegate.
Now let's look at the remaining February contests we've had since then...
State | Pledged Total | Actual Obama | Actual Clinton | Actual Margin | Predicted Obama | Predicted Clinton | Predicted Margin | Error |
Louisiana | 56 | 34 | 22 | 12 | 31 | 25 | 6 | 6 |
Nebraska | 24 | 16 | 8 | 8 | 15 | 9 | 6 | 2 |
Virgin Islands | 3 | 3 | 0 | 3 | 2 | 1 | 1 | 2 |
Washington | 78 | 53 | 25 | 28 | 49 | 29 | 20 | 8 |
Maine | 24 | 15 | 9 | 6 | 10 | 14 | -4 | 10 |
Democrats Abroad | 7 | 4.5 | 2.5 | 2 | 5 | 2 | 3 | -1 |
District of Columbia | 15 | 12 | 3 | 9 | 9 | 6 | 3 | 6 |
Maryland | 70 | 42 | 28 | 14 | 37 | 33 | 4 | 10 |
Virginia | 83 | 54 | 29 | 25 | 43 | 40 | 3 | 22 |
Hawaii | 20 | 14 | 6 | 8 | 11 | 9 | 2 | 6 |
Wisconsin | 74 | 42 | 32 | 10 | 40 | 34 | 6 | 4 |
Total | 454 | 289.5 | 164.5 | 125 | 252 | 202 | 50 | 75 |
As you can see his campaign knew he would do well in February. They thought they would win every contest except Maine. Their projections were modest, though. With the exception of Democrats Abroad (where it looks like they'll take a 4.5-2.5 victory instead of a 5-2 victory) they did better in every single contest. In Maine, Maryland, and Virginia they did much better than they had projected. In total they got a swing of 75 more delegates than they had thought. That's huge. So before Tuesday, let's look at what the campaign was projecting for the remaining contests...
State | Pledged Total | Predicted Obama | Predicted Clinton | Predicted Margin |
Ohio | 141 | 68 | 73 | -5 |
Rhode Island | 21 | 8 | 13 | -5 |
Texas | 193 | 92 | 101 | -9 |
Vermont | 15 | 9 | 6 | 3 |
Wyoming | 12 | 7 | 5 | 2 |
Mississippi | 33 | 20 | 13 | 7 |
Pennsylvania | 158 | 75 | 83 | -8 |
Guam | 4 | 2 | 2 | 0 |
Indiana | 72 | 39 | 33 | 6 |
North Carolina | 115 | 61 | 54 | 7 |
West Virginia | 28 | 13 | 15 | -2 |
Kentucky | 51 | 23 | 28 | -5 |
Oregon | 52 | 28 | 24 | 4 |
Montana | 16 | 9 | 7 | 2 |
South Dakota | 15 | 8 | 7 | 1 |
Puerto Rico | 55 | 25 | 30 | -5 |
Total | 981 | 487 | 494 | -7 |
So back in early February Obama's campaign was projecting--yes, just one scenario--that they would not do quite as well as Clinton in the remaining contests, but that they would keep it close. For Tuesday's states they thought they would only win Vermont, but stay within 10 delegates in each of the other states. Interestingly it looks like they thought they would do better in Ohio than Texas. After that they thought they would win Wyoming, Mississippi, Indiana, North Carolina, Oregon, Montana, and South Dakota. They thought they would split Guam and lose Pennsylvania, Kentucky, West Virginia, and Puerto Rico. Seems reasonable to me except that (1) I'm not sure what they believed would give them such an edge in Indiana and (2) I think their margins of victories in those other states they predict they'll win will be once again significantly larger. Well, we'll see.