Your analysis pretends that we're talking about some dry arithmetic problem here. In fact, we're talking about politics, and it's not at all linear. For example, I can imagine how progressives could block the next Alito by winning only ONE election: If Ned Lamont beats Joe Lieberman in the primary. I noticed you completely ignored primaries - for example, you didn't mention Chuck Pennacchio in Pennsylvania. Granted, he's a long shot, but primaries have low turnout and they're the sort of elections where committed activists can have a greater impact. Toomey was a long shot against Specter, too, but look how close he came.

In 2004, the Constitutional Convention of the Massachusetts legislature voted 105 to 92 in favor of an amednment that would ban gay marriage. Constitutional amendments in MA need to be approved by two consecutive sessions of the legislature, so the same amendment came up again in 2005. This time, the vote was 39 to 157, and the amendment went down in flames. What happened? Did we defeat 66 anti-gay legislators in one year and replace them with 66 new pro-gay legislators? No. 17 of the 18 new legislators elected in the intervening year, voted against the amendment... and 55 who were there for both votes, switched their vote from yes to no.

There were only a tiny handful of elections that turned the tide. Most significant, by far, was the election of Carl Sciortino, a young first-time candidate who challenged a 16 year incumbent conservative Democrat in the primary, and won by 94 votes. Nobody expected him to even come close, and his win shocked the establishment. In combination with two other less shocking elections that also reinforced the message, legislators literally began thinking, "they're going to come after us". When, several months later, three conservative Democrats who had supported the amendment resigned and the special elections to fill the open seats elected three pro-gay anti-amendment Democrats, the message became even clearer, and a special election for state senate in the summer sealed it. Voting against gay marriage had been seen as a "safe" vote the first time around. By the 2005 Constitutional Convention, it was seen as a serious political risk. Behavior modification followed.

Ned Lamont is much less of a long shot than Toomey was, and the energy behind his campaign may be immense. If he wins, behavior modification will certainly hit several other Senate Democrats. A filibuster of the next Alito may get 40 votes - especially if Democrats pick up 4 or 5 more seats overall.

You have a great point, and thanks for writing (drawing:) about it!

However, machines that "produce a paper" are still a problem, compared with the real gold standard: hand marked paper ballots. Some of the main problems with "paper trail" touchscreen voting:

1. The paper printout is usually cheap, light, and hard to read; surveys show most voters don't review it, and even if they do, they're likely to miss some errors.

2. They're also hard to count, if need be later. Paper ballots are designed to be counted, but "paper trails" aren't. You'd usually end up with the equivalent of a very long roll of cash register receipt and have to have people go through it by hand.

3. Machines that can record votes incorrectly, due to bugs or fraud, can just as easily print incorrectly - the fact that your votes are right on the paper mean nothing if the machine has problems. So paper trails only really matter if they are the BALLOT OF RECORD - the official votes. Unfortunately, because they're just an "audit trail", legally, in many places, the machine's electronic count is the official count. Even if the paper has a different result.

Touchscreen voting machines are often touted as being great for disabled voters, but that makes the problem even worse: they're less likely to be able to "verify" the paper trail, so to catch problems with the machines you really need to have all of the abled voters using them as well.

The right solution is to provide a Vote-PAD option for disabled voters ( http://vote-pad.us/ ) and have everyone hand-mark their paper ballots. Those ballots are inherently "verified", legally the ballots of record, and designed to be counted.

It's very misleading to talk about a "popular vote" for Super Tuesday, because nobody knows what it as. As your article mentioned, most caucus states don't release (or have) vote counts, they give precinct delegate counts. The total number of precinct delegates was fixed in advance, regardless of actual voter turnout, and the higher the turnout, the higher the ratio of voters to delegates. Adding up the vote counts from all the primary states to the precinct delegate counts from the caucus states leaves you with sums that mean nothing.

Since Obama won every single caucus state, while Hillary won more primary stats than Obama did, that fantasy-sum you call "popular vote" obviously severely undercounts Obama's vote. So with the totals as close as they are, it's very likely Obama won the popular vote too. But we don't know and we probably won't know.