How do electoral math differ from popular votes?

After the 2000 election one might think that everyone in the world understands how an American president is elected. For those who don’t remember or don’t understand, electors are allocated to each state based on how many congressmen and senators that state has. For example, South Dakota has one congressman and two senators, and is therefore entitled to 3 electoral votes. Therefore, with the total number of congressmen being 435 and there being 100 senators, along with three electors allocated by Constitutional amendment to Washington D.C., there are total of 538 electors.

So how far off from the popular vote could an electoral vote get? To figure this out, you need to know how many voters there are in each state and the number of electors per state. 270 electors can be gotten by winning evenly 50% of the votes in the 11 largest states in the Union. If the loser were to take 50%-1 vote in those states, plus all the votes in the other states, using 2008 voter information, the winner would need only 43 million votes, while the loser could have over 110 million, or 71% of all the votes. This is how President Bush won the election 2001 but lost the popular vote.

The electoral system almost always produces less dramatic results that do not mirror the popular vote. Let’s look at a few percentages:

>Year	Who	Winner Electoral Votes	Loser Electoral Votes	Percentage Electoral Win/Loss	Winner Popular Votes	Loser Popular Votes	Percentage Popular Win/Loss
1904	Roosevelt v. Parker	336	140	70.6%/39.4	7,630,457	5,083,880	56.4%/37.6%
1908	Taft v. Bryan	321	162	66%/34%	7,678,395	6,408,984	51.6%/43.0%
1912	Wilson v. Roosevelt	435	88	82%/17%	6,296,284	4,122,721	41.8%/27.4%
1916	Wilson v. Hughes	277	254	52%/48%	9,126,868	8,548,728	49.2%/46.1%
1920	Harding v. Cox	404	127	76%/24%	16,144,093	9,139,661	60.3%/34.1%
1924	Collidge v. Davis/Follette	382	149	72%/28%	15,723,789	13,217,948	54%/45.4
1928	Hoover v. Smith	444	87	84%/16%	21,427,123	15,015,464	58.2%/40.8%
1932	Roosevelt v. Hoover	472	59	89%/11%	22,281,277	15,761,254	57.4%/39.7%
1936	Roosevelt v. Landon	523	8	98%/2%	27,752,648	16,681,862	61%/37%
1940	Roosevelt v. Wilkie	449	82	85%/15%	27,313,945	22,347,744	54.7%/44.7%
1944	Roosevelt v. Dewey	432	99	81%/19%	25,612,916	22,017,929	53.4%/45.9%
1948	Truman v. Dewey/Thurmond	303	228	57%/43%	24,179,347	23,167,222	49.6%/47.5%
1952	Eisenhauer v. Stevenson	442	89	83%/17%	34,075,529	27,375,090	55%/44%
1956	Eisenhauer v. Stevenson	457	73	86%/14%	35,579,180	26,028,028	57%/24%
1960	Kennedy v. Nixon	303	219	56.5%/40.9%	34,220,911	34,108,157	49.7%/49.6%
1964	Johnson v. Goldwater	486	52	90%/10%	43,127,041	27,175,754	61%/39%
1968	Nixon v. Humphrey/Wallace	301	191 + 96	56%/36%/18%	31,783,783	41,172,957 (total)	43.4%/56.2%
1972	Nixon v. McGovern	520	17	97%/3%	47,168,710	29,173,222	61%/38%
1976	Carter v. Ford	297	240	55%/45%	40,831,881	39,148,634	50.1%/48.0%
1980	Reagan v. Carter	489	49	91%/9%	43,903,230	35,480,115	50.7%/41.0%
1984	Reagan v. Mondale	525	13	98%/2%	54,455,472	37,577,352	59%/41%
1988	Bush v. Dukakis	426	111	79%/21%	48,886,597	41,809,476	53.4%/45.7%
1992	Clinton v. Bush	370	168	69%/31%	44,909,806	39,104,550	43%/37.5%
1996	Clinton v. Dole	379	159	70%/30%	47,401,185	39,197,469	49.2%/40.7%
2000	Bush v. Gore	271	266	50.4%/49.6%	50,456,002	50,999,897	47.9%/48.4%
2004	Bush v. Kerry	286	251	53%/47%	62,040,610	59,028,444	50.7%/48.3%
2008	Obama v. McCain	365	173	68%/32%	69,456,897	59,934,814[	53%/46%

In this table, the closest the popular and electoral votes come together is in 1916, although Bush v. Kerry comes close.

So what do we learn from all of this? I see two key messages:

The nature of the electoral voting system wildly distorts popular will in favor of each state getting at least some voice. This was, after all, the reason for its design.
National polls are, at best, a finger in the wind, and may be entirely misleading.

Loading ...

Here comes World IPv6 Day!

As you may have read in the press some time ago, the world is running out of IP addresses. Really the world is running out of the current version IP addresses. An IP address is the means by which your computer and my computer can communicate with each other. Addresses are similar to phone numbers in that if we each have a unique number we both can call each other.

How is it we’ve run out? Quite simply the IP version 4 address size is fixed at 32 bits, which allows for at most a little over 4 billion simultaneous computers to connect. Through the use of some sneaky tricks we are able to connect well more than 4 billion under the assumption that not device needs to be able to communicate with ever other device, but that game is getting a bit overplayed.

And so over fifteen years ago, the Internet Engineering Task Force (IETF) created IPv6, which has enough address space to stick an address on every speck of sand we have in the world. More precisely IPv6 can handle 2¹²⁸ or 340,282,366,920,938,463,463,374,607,431,768,211,456 addresses.

NOW THAT’S A LOT OF PASTA!

Nobody wanted IPv6 way back then when we had plenty of IPv4 address space, but now that we’re out of IPv4 addresses, it’s moving day. That’s because we’ve become mobile, and computers have gotten smaller. Not only can a cell phone access the Internet, but so can your printer, a car, a boat, a camera, your television, washing machine, many game systems, and many other things.

Tomorrow is World IPv6 Day. Many service providers and web sites will be enabling the next generation Internet Protocol tomorrow to see what works and what breaks. Will this inconvenience you even just a little? Probably not. Here’s why: your home gateway almost certainly doesn’t support IPv6, unless you’re a geek like me, in which case IPv6 day might inconvenience me. But I had to go to quite some inconvenience already to get IPv6 into my home, so what’s just a little bit more?

Anyway, it’s all one big test to see how painful moving to IPv6 really is, and to see what breaks and what needs fixing. As service providers and web sites kink out bugs you’ll be hearing more about IPv6. Eventually, much like you did when you moved to high definition television, you’ll probably need a new router. If all goes well, the only difference you’ll notice is that eventually services like Skype and iChat AV will improve.

By the way, this blog is IPv6-enabled!

As if On Queue: Google accounts attacked from China

The BBC reports today how China is rejecting Google’s statement that attacks on its users originated from China. It’s very fair for China to call into question from whence attacks originate. The best Google can really authoritatively say is that they saw attacks coming from a particular set of IP addresses that happen to be registered to a network that resides in a particular location, in this case Jinan.

However, the attacks targeted individuals said to be Chinese dissidents or adversaries. In this case, as the BBC writes, while it is very difficult to state with assurance that the attacks were made by the Chinese government, the technique used, spear phishing, leads one to believe that this attack was in fact paid for, in some way, by a government. Spear phishing involves learning about a particular individual, and then crafting a message that that person would think came from someone they knew, and convincing that person to view an attachment that itself contains a virus. That virus must be relatively unknown, or virus checkers will pick it up. The cost of spear phishing is high, and the monetary pay-off tends to be low. Therefore, it is a good fit for an intelligence organization.

In addition, as I wrote not long ago, Cambridge University investigated a break-in of the Office of His Holiness, The Dalai Lama. Those attacks also seemed to originate from China, they were also targeted against an adversary, and worst of all, China apparently acted upon the information stolen by applying diplomatic pressure against those countries who invited the Dalai Lama.

At the very least, China bears some culpability for allowing the attack. Here we have a government that does not believe in the free flow of information, and so they are known for monitoring everything. How, then, did this attack escape their notice?

Win! Dealer supports customer! This must not be America

It’s not- it’s Switzerland.

So here’s a story for my friend Tom who hates BMW America. I was beginning to feel the same way about our national sales organization here in Switzerland. We had been hearing what to me sounds like valves banging around, practically since the day we got the car, but it was intermittent. Well, 4 years and 22,000 miles later (after the warranty is supposed to have expired), we were able to reproduce it. And guess what? It was valves banging around.

Initially, BMW Switzerland said they’d cover parts and 1/2 labor. But our guy at the local garage argued back at them on our behalf and got them to eat the whole valve job.

Yippee!

Is hacking Skype a human rights violation?

Not twenty four hours ago did I write about how the Pentagon is going to announce how cyber-attacks could be casus belli. Now the Wall Street Journal reports that an Egyptian intelligence agency was monitoring Skype communications of dissidents. Let’s first agree on a truth. No one’s right to privacy is absolute or ever assured. However, plotting the peaceful overthrow of a government (in America we call that an election) should not be subject to snooping. If we can go to war over hacking, should we not then also stand up for people’s human rights to peacefully and privately express their views? Ronald Reagan used to rail on how the Soviet Union wasn’t free. He was right. Now here we are in age of the Internet. What do his words mean in today’s context? The free flow of information is a human right. It’s not absolute if, for instance, you’re talking about robbing a bank.

By the way, the Egyptians did not break Skype’s encryption, but instead seem to have infected the system of the dissidents. That’s something Skype can only partially control- that is, if the infection was spread by Skype’s Instant Messaging capability, then they do bear some responsibility. But if it was spread by other means, then there’s really not much they can do.