Nomination Math 101: How Obama Is Ahead by over 1 Million in the Popular Vote (Even Counting FL & MI)
The rules of the nomination process in the Democratic Party make it very clear that the only measure which counts toward the nomination is the delegate count. Nevertheless, the idea that the so-called "popular vote" still has relevance continues to have traction, with the Clinton campaign and its backers easily, and understandably, the loudest proponents.
At least on this point, however, the Clinton campaign should perhaps be forgiven; afterall, most election sites list these numbers alongside the delegate totals as if they truly represented what they claim to, and pundits never seem to tire of bringing the popular vote up as an important measure of each nominee's prospects. Real Clear Politics even lists popular vote totals including Florida and Michigan which show Clinton ahead, so no wonder so many keep pushing this notion.
The fact is, however, that the "popular vote" in the Democratic nomination is a chimera. It not only is meaningless because of its lack of legitimization in the official selection process, but also because the numbers truly do not represent anything meaningful. Wikipedia explains quite well the intrinsic impossibility of attaining a true popular vote count, but it boils mostly down to the idiosyncracies of the caucus states, since either:
1) official vote totals have not been released (IA, NV, WA, ME) or
2) the time-consuming nature of caucusing lowers relative turnout dramatically compared to standard primaries.
What this means is that, even if all the individual caucusing numbers were there, they still wouldn't in themselves give us a representative popular vote because all of the caucus states would be severely underrepresented in relation to their total voter participation in the general election. Pledged delegates are apportioned relative to population and voting history for this very reason -- to compensate for the resulting disparities in voter turnout due to different selection methods at the state level -- and this is exactly why only the delegates matter.
But would it perhaps still be possible to compensate for a caucus state's lower turnout to get a rough approximation of what a "real" popular vote might be?
Yes!
I decided a good approach would be to compare the ratios of voters to pledged delegates between primary states and caucus states and to use the result to increase the caucus numbers by that factor. Once I've transposed the caucus numbers upwards for each state's adjusted popular vote, I could add all of these to the popular vote from the primary states for the national total.
So essentially it's a three-step process:
I. Calculate the average number of voters per delegate in primary states. (I chose five states from different parts of the country to get this number.)
II. For each caucus state, calculate the same ratio (whether with votes, county delegates, etc.) and modify the caucus numbers by the ratio I/II to get the adjusted popular vote by state.
III. Using the popular vote total without caucus states as a base, add the adjusted state numbers for the adjusted, national popular vote.
And here's what it looks like (scroll to the bottom for the good stuff)...
I. Calculation of average number of voters per delegate in primary states (5 states as representative)
Voters per delegate IN
1274993/72 = 17708.24
Voters per delegate PA
2307759/158 = 14606.07
Voters per delegate AZ
455635/56 = 8136.34
Voters per delegate MS
434152/33 = 13156.12
Voters per delegate MD
878141/70 = 12544.87
Average voters per delegate = 13230.33
II. Calculation of voter/delegate ratios in caucus states and the resulting adjusted popular vote (by interpolation)
Precinct delegates per delegate IA
2501/45 = 55.58
Transposed popular vote IA
13230.33/55.58 x 940 = 223759 (O)
13230.33/55.58 x 737 = 175436 (C)
County delegates per delegate NV
10742/25 = 429.68
Transposed popular vote NV
13230.33/429.68 x 4844 = 149152 (O)
13230.33/429.68 x 5459 = 168088 (C)
Votes per delegate AL
8880/13 = 683.08
Transposed popular vote AL
13230.33/683.08 x 6674 = 129266 (O)
13230.33/683.08 x 2194 = 42495 (C)
Votes per delegate Am. Samoa
285/3 = 95.00
Transposed popular vote Am. Samoa
13230.33/95.00 x 121 = 16851 (O)
13230.33/95.00 x 163 = 22700 (C)
Precinct delegates per delegate CO
120971/55 = 2199.47
Transposed popular vote CO
13230.33/2199.47 x 80113 = 481899 (O)
13230.33/2199.47 x 38839 = 233626 (C)
Votes per delegate ID
21224/18 = 1179.11
Transposed popular vote ID
13230.33/1179.11 x 16880 = 189404 (O)
13230.33/1179.11 x 3655 = 41011 (C)
Votes per delegate KS
36723/32 = 1147.59
Transposed popular vote KS
13230.33/1147.59 x 27172 = 313260 (O)
13230.33/1147.59 x 9462 = 109085 (C)
Votes per delegate MN
214066/72 = 2979.14
Transposed popular vote MN
13230.33/2979.14 x 142109 = 631105 (O)
13230.33/2979.14 x 68994 = 306402 (C)
Precinct delegates per delegate ND
19012/13 = 1462.46
Transposed popular vote ND
13230.33/1462.46 x 11625 = 105167 (O)
13230.33/1462.46 x 6948 = 62856 (C)
Votes per delegate NE
38670/24 = 1611.25
Transposed popular vote NE
13230.33/1611.25 x 26126 = 214526 (O)
13230.33/1611.25 x 12445 = 102189 (C)
Votes per delegate WA
31984/78 = 410.05
Transposed popular vote WA
13230.33/410.05 x 21629 = 697863 (O)
13230.33/410.05 x 9992 = 322394 (C)
State delegates per delegate ME
3498/24 = 145.75
Transposed popular vote ME
13230.33/145.75 x 2079 = 188719 (O)
13230.33/145.75 x 1397 = 126811 (C)
Votes per delegate HI
37426/20 = 1871.30
Transposed popular vote HI
13230.33/1871.30 x 28347 = 200417 (O)
13230.33/1871.30 x 8835 = 62465 (C)
Votes per delegate WY
8753/12 = 729.42
Transposed popular vote WY
13230.33/729.42 x 5378 = 97547 (O)
13230.33/729.42 x 3311 = 60055 (C)
Votes per delegate Guam
4521/4 = 1130.25
Transposed popular vote Guam
13230.33/1130.25 x 2264 = 26502 (O)
13230.33/1130.25 x 2257 = 26420 (C)
III. Calculation of the adjusted popular vote totals using the weighted vote results from the caucus states
Popular vote totals w/o caucuses (from RCP)
Obama = 16,319,472
Clinton = 16,059,692
Total transposed popular vote of caucus states (from II)
Obama = 3,665,437
Clinton = 1,862,033
ADJUSTED "REAL" POPULAR VOTE (with transposed caucus totals)
Obama = 19,984,909
Clinton = 17,921,725
------> Obama +2,000,000
ADJUSTED "REAL" POPULAR VOTE + FL and MI (even assuming Obama gets no votes in MI)
Obama = 20,561,123
Clinton = 19,121,020
------> Obama +1,400,000
So what this shows is obvious: that Obama is clear and away the nominee and that only our poor math comprehension is blinding us from that fact.
Your premise is correct. If you have caucuses then the whole point of popular vote is moot.
I think your projection exercise was perhaps not worth the time.
May 22, 2008 7:17 PM | Reply | Permalink
I considered that somewhere about halfway through when I realized just HOW MANY caucus states there are! (I really kicked myself for not using a spreadsheet, too, but by that point it wouldn't have saved time.)
I'm of two minds about whether this was worth doing. On the one hand, there are many who insist on believing the value of the popular vote count AND who are incapable of following the reasoning and math of the argument (even when explained in detail).
But I still think there are many who are a bit brighter and still stubbornly hold onto their belief about the popular vote, BUT could follow a line of reasoning challenging this belief.
I know, it's probably a stretch...
May 22, 2008 7:42 PM | Reply | Permalink
Data is good though...it helps when reason is hiding under a rock somewhere. Certainly less offensive and divisive than sticking out your tongue saying 'nah-nah, nah-nah, naaah-nah'. Props on your work.
May 22, 2008 8:18 PM | Reply | Permalink
Hillary supporters will now ask you if you are using octal, decimal, or hex arithmetic to come to these conclusions.
May 22, 2008 8:03 PM | Reply | Permalink
Could you explain this please: "a good approach would be to compare the ratios of voters to pledged delegates between primary states and caucus states ". I can't see how this is not a completely arbitrary out-of-the sky metric.
May 22, 2008 8:20 PM | Reply | Permalink
This is exactly why the projection exercise was unwise and unnecessary. "Popular vote" is an equally arbitrary metric in a process that is not based on popular votes. Even the delegate apportionments in a particular states are often based primarily on margins within districts and not overall popular vote in the state.
May 23, 2008 12:31 AM | Reply | Permalink
I actually find this quite interesting. Although it looked quite time-consuming, and in the end the "popular vote" doesn't even matter in electing a nominee, it's still a fascinating post.
You might like this:
http://blog.newsweek.com/blogs/stumper/archive/2008/04/04/the-popular-vote-fallacy.aspx
I've been spreading it around. It takes a couple of the arguments you've talked about and expands upon them.
Allow me to use my own state of MN as an example. Minnesota is a caucus state with results of 66-32% in favor of Obama which yielded an excess of caucus voters of 73,115 for Obama. This figure is included in the media created margin of approximately 700,000 for Obama. But, because turnout in Caucus states is considerably lower than in primary states (Clinton is right about that), the margin of victory is also held down. Neighboring Wisconsin has a similar sized electorate. If the same number of voters turned out for a Mn primary as in Wisconsin and if Obama had the same 66-32% majority, his voter margin would have been over 340,00 rather than merely 73,115. Or, if Obama carried a hypothetical Mn primary by the same pct as neighboring Wisconsin (58-41%), then Obama's popular vote margin would have been 190,000 instead of merely 73,115. Even if Obama's margin of victory would only have been 10%, his popular vote margin would have been 110,000 --- still considerably more than he is credited with when only including caucus margin of victory.
Treating the vote totals in caucus states in the same category as primary vote totals significantly disadvantages the caucus states. Even though Mn's electorate size is comparable to Wisconsin, there were 5 times as many voters in the Wis primary as in the Mn caucus. This makes the Wisconsin primary 5 times more important than Mn even though the electorates are essentially equal. Another reason why it is a myth to suggest there is such a thing as a popular vote total. If the rules are unfair, change them but do so before the primary season not after. For this season, stick with the rules.
May 22, 2008 9:18 PM | Reply | Permalink
That's an excellent summary of why the popular vote is completely erroneous.
I began with the intent to only calculate the adjusted popular vote for one caucus state to have an example, but one thing led to another...
May 23, 2008 2:34 AM | Reply | Permalink
I really like what you did here and it's an argument that I've been throwing around in my mind for a while, but I would like to add one caveat.
Obama's ground organization has been exceptional and I believe drives up his percent of the vote in caucus states. Now, there's nothing wrong with this; it's good strategy on his part and it's worked out well for him. However, I do believe that if every caucus state had a primary his vote percent and overall delegates would be lower.
If I had to venture a guess, I still think he'd gain popular vote. In other words, I suspect that he'd drop from a 70%-30% victory to a 58%-42% victory but with 10 times the turnout. So I suspect that his popular vote total would increase overall but not by as much as you're saying.
But kudos to you on bringing up this point; bigger than the math that you did is the point you made, which is that this "popular vote" thing is complete garbage when you have different states nominating in different ways. We really have no idea what the popular vote is within any reasonable margin.
May 22, 2008 9:38 PM | Reply | Permalink
I agree about the influence of the ground organization. I also suspect, since voting decisions aren't private in caucuses as they are in primaries, that people tend to move towards one candidate or another en masse.
So yes, I think my calculations do exaggerate his lead, but that they still are closer to representing whatever "popular vote" can be salvaged from the process than the baseless figures published everywhere.
May 23, 2008 2:43 AM | Reply | Permalink
Your approach is ridiculous on it's face. You ignored Texas which had a primary and a caucus and you can plainly see how caucuses distort the process. You wasted your time taking a distorted number in trying to create an Obama feel-good number. You might as well just make it up. Oh...wait...you just did.
Caucuses are nothing but vote suppression. Obama has a narrow delegate lead because of bullshit caucus wins and wins in states that no democrat will win in the fall. And with no caucuses in November it doesn't bode well for the dems in spite of having the most favorable political climate in memory. Not to mention, the DNC is still arguing about whether to spit in the faces of FL & MI voters once again.
Unbelievable stupidity.
May 22, 2008 9:42 PM | Reply | Permalink
If you hate the caucus system, fine. But don't bitch about it now that the nomination process is over. The time for complaining is long gone. If you had such a problem with the caucus system before, then why not complain then? Why didn't more people complain then? Because no one cared. Now that one candidate effectively used the system to his advantage to win, there's an issue. And you talk of unbelievable stupidity. Man, oh man.
May 22, 2008 10:14 PM | Reply | Permalink
You're right that I ignored Texas, but only to the extent that I didn't take its caucus vote into account at all and only used its primary result.
Here you're wrong. The "made up" number is the popular vote total used repeatedly by Clinton and the media which completely leaves out four caucus states completely and underrepresents the other eleven.
That's a fair argument, but it should be noted, that 1) the DNC rules allow states great freedom to choose the delegation selection process which they deem works best, and 2) the party members in those states apparently DON'T consider a primary system better or else they would push to switch to it.
Of course, the DNC could mandate all states to use the same primary system, but that would deny every state the freedom to decide what's best for themselves. Kind of undemocratic, don't you think?
May 23, 2008 3:04 AM | Reply | Permalink
I think ChronoSpark's post below explains it better than I could, but in essence it's about multiplying caucus results by a factor that compensates for their lower numbers, a result of the diminished participation in caucuses due to their time-consuming nature.
May 23, 2008 2:27 AM | Reply | Permalink
Um, that was meant for Lalo35adm!
May 23, 2008 2:28 AM | Reply | Permalink