I am certain that the Washington Nationals will do fabulously in this year’s playoffs. How do I know? I moved from the DC area to Florida in January. Curses evaporate when I leave.

There’s precedent for this. After I moved to DC from Boston in 2001, the Red Sox built themselves into curse-destroyers in 2004. I suffered along with all of Boston up until then, and I didn’t even get to be there for the payoff! Now, several years later, Washington has its first playoff victory in baseball for the first time since trilobites swam the Potomac. I pondered this while faithfully keeping  a scorecard for tonight’s Orioles-Yankees game. Latent dislike for the Yankees contributed to me pulling for the Orioles, even though their owner deprived Washington of baseball for so long.

Constructing a box score is one of my less marketable skills. Before politics consumed my life after I moved to DC, I was a suburban child in Boston who was a relentless baseball stats nut. I didn’t know it at the time, but spending my elementary school years poring over ways to compare pitchers to Pedro Martinez in Microsoft Excel seems to have foreshadowed my current job at a polling firm, trading left fielders for legislators.

I learned the arcane runes of the baseball scorecard because we didn’t have cable in our basement apartment in Needham and later townhouse in Norwood, Massachusetts, and even fanatical baseball towns like Boston don’t air every game over the air. I clung to a staticky radio that was almost certainly older than I was, the dial carefully tuned to WEEI 850 AM (where Red Sox fans have surely returned to an endless parade of misery and rending of clothing, now that the Sox have sunk into the cellar). If you could watch the game on television, you never needed to draw up a scorecard: the game was right in front of you, and chyrons did all the work of keeping you appraised of previous at-bats…but listening to even the most skilled radio announcer required a bit of imagination and visualization.

Pedro Martinez in 2000. Mark Wilson/Boston Globe

I remember the 1999 Boston Red Sox vividly. Just like how your first loves are seared forever in your memory whether you like it or not, you remember your formative baseball years when you first started obsessively tracking strikeout to walk ratios. I remember them vividly because I was eight, and had by then become well versed in the Curse of the Bambino, and the mythology of suffering it brought to flinty New Englanders. I followed them all season with my radio as if I was a child of 1949 and not 1999, and right as they came back from a 2-0 deficit to beat the Cleveland Indians in the ALDS, including a record 23-7 romp so ridiculous they started running out of numbers for the Fenway Park scoreboard, they lost the pennant to the hated Yankees.

This was one of the first tastes I had of the vengeful wrath of destiny.

2000 came along, and another peculiar habit of flinty New Englanders aside from self-flagellating about the Red Sox captured my attention: the New Hampshire Primary (which triggers more use of “flinty” than any other event). I watched in awe as the primaries gave way to recounts Bush v. Gore and chaos, and I swore that every election going forward would be just this damn exciting. I took some colored pencils to an Electoral College map for the first time. Then in 2001, I moved to Washington, a city that didn’t have baseball at the time, and no cable package to keep up with the Red Sox, and even worse I was way too far to pick up even the faintest hint of WEEI. The internet simply could not do, and I devoted my obsessive energies to elections up and down the ticket and lost track of America’s pastime.

Even when the Nationals came to town I didn’t become obsessive again. I had known them, after all, as the hapless Montreal Expos. Eventually I realized after a rootless childhood that Washington, DC would be my hometown, but by then my interest in baseball had become part of my past. Times had changed, and I had new hobbies to eat up my time, but now that the Nationals bandwagon has picked up even the most jaded Washingtonians, I’ve started watching again. (Sure, heap scorn on me and my bandwagon clinginess, I won’t lie!)

The Boston Red Sox won the World Series in 2004. Now that I’ve moved from DC, it must be the Nationals’ turn. And who knows: if I managed to turn one of my childhood pastimes of coloring in primitive election maps into part of the paying job I hold today, maybe I can figure out a similar arrangement for baseball.

In 2004 and 2008, search volume from June to September of the election remained about even between the two major party candidates, but this year is different.

The chart above illustrates comparative search volumes for the names of the two major party candidates (no quotes) from the 21st to 6th week before the election. The conventions are noted.

Over the last several weeks, search volume for Mitt Romney on Google has far outpaced search volume for Barack Obama. While some of this gap can be explained by Barack Obama’s incumbency, the gap between John Kerry and George Bush in 2004 was much smaller. In all elections from 2004 onward, conventions produced a bounce in search volume for the hosting party, but this year search volume between the two candidates has failed to revert to a rough equilibrium.

Thanks to the vast wealth of data available at Google Trends, we can compare search volume between different terms over time, and within states and media markets. What Google Trends provides is indexed, not raw data, so the highest point is “100” for any given comparison. I pulled the Google search index for the presidential candidates in the period from June to September, with data in the table at the end of the post.

As can be expected, George W. Bush in 2004 was a well-known incumbent, and received little convention bounce in search volume compared to his challenger. 2008 was an open-seat race, so both candidates holding back to back conventions traded search leads, with Barack Obama going through September slightly ahead. This year, though, searches for Mitt Romney have continued to far outpace that of Barack Obama even after the conventions are gone.

What gives? The reason behind the coloring of this tea leaf depends on your partisanship. Google search volume is hardly the most indicative tea leaf, but there is a vast trove of interesting data to play with. It is rather hard to write about data like this without putting some spin on it.

As a Democrat, I’d like to believe that all these searches for Mitt Romney are because even Romney has failed to define himself. Incumbent-challenger elections are frequently referendums on the well-known incumbents. Republicans this year sorely wanted this election to be about Barack Obama and his economy. Mitt Romney’s campaign has managed to get this far without being able to define himself: and Democrats have filled in the gap. Strangely enough, this year’s election is increasingly becoming more about how Mitt Romney is a job-destroying, out-of-touch Richie McRich than it is on Barack Obama.

If you’re a Republican, you’d like to believe that interest in Mitt Romney is riding high. Voters are clamoring to see more information about this year’s alternative. Why would anyone search for a candidate you already know about in Barack Obama? Voters already know how the Obama economy has worked for them, and they want to know who Mitt Romney is.

Truth be told, I have no idea what Mitt Romney truly believes in, and you probably don’t either. What do you do if you don’t really know what a candidate stands for these days?

Just Google it.

Pro-democracy Legislative Councillor Tanya Chan (center right, black shirt) breaks up a fight at a protest outside the old Legislative Council building on June 23, 2010. I took this photo while in Hong Kong on a summer exchange program which placed me with the Civic Party for a few weeks. Chan lost her seat on the Legislative Council on Sunday’s election thanks to nominating strategy and a flawed vote allocation system.

In Hong Kong, you win by losing.

In yesterday’s Hong Kong Legislative Council, the pro-democracy Civic Party won the most votes in Hong Kong Island and New Territories West, but managed to win only one seat in each district as the rival pro-Beijing DAB got two each. Another large pro-democracy party, the Democratic Party had enough votes to win a seat in New Territories West, but won zero. How is this possible?

The answer is in the voting system. Much like how structural problems in the voting system gave George W. Bush a victory in the infamous 2000 Presidential election, the Civic Party lost out on two extra seats thanks to a voting and apportionment system flaw. While no voting system is perfect, as proven mathematically by Arrow’s Impossibility Theorem, the selection of the Hare quota and highest-remainders is especially helpful to pro-Beijing parties like the DAB.

Hong Kong voters went to the polls on September 9th to vote  for their 70-member Legislative Council, a body stacked with “functional constituencies” that give pro-Beijing parties a guaranteed majority. However, 35 seats are elected in geographical constituencies ranging from 5 to 9 seats that use closed-list proportional representation. Specifically, Hong Kong uses the Largest Remainder or Hare quota.

From Wikipedia:

The largest remainder method requires the numbers of votes for each party to be divided by a quota representing the number of votes required for a seat […] The result for each party will usually consist of an integer part plus a fractional remainder. Each party is first allocated a number of seats equal to their integer. This will generally leave some seats unallocated: the parties are then ranked on the basis of the fractional remainders, and the parties with the largest remainders are each allocated one additional seat until all the seats have been allocated.

There’s a reason most countries that use proportional representation have switched away from the Hare quota to a highest-averages method like D’Hondt or Sainte-Lague: the Hare quota is extremely vulnerable to strategic voting (配票 pui piu), and strategic nominating. If your party is trying to win two seats in a district, and polling shows that you have enough support, you have two choices.

1. Run one list. The Civic Party chose this strategy. This virtually guarantees the first person on your list will win a seat, but whether your ticketmate wins depends on how high your remainder is–and you’ll need a high remainder. If you are running in a crowded constituency, all of the remaining seats might be gone before it’s your turn again. You do not need to allocate votes to different lists, you simply need to pile as many votes as possible on one list.

2. Run two lists. The Democratic Party and the DAB chose this strategy. You need fewer total votes to win two seats. Ideally, your two lists would win highest remainder seats. However, if you have overestimated your support or misallocate your votes, both your lists might lose, leaving you with no seats.  This requires machine-like precision to avoid total 失敗 (sat baai…failure).

Let’s see how Option A turned out for the Civic Party. Results are below:

The Civic Party won the most votes, highlighted in purple. The DAB, however, ran two lists: and even though the Civic Party beat the DAB by 57 votes, the DAB won two seats! Had the Civic Party split their candidates into two lists, and split their votes perfectly evenly, they would have won two seats, as shown below:

Hypothetically, splitting into two lists would have given the Civic Party two seats: but the real world does not necessarily work like this.

This strategy is not foolproof. Running one list was less risky because it guaranteed Chan Ka-lok’s election even if they didn’t have enough support for a 2nd seat. Had the Civic Party overestimated their support by 8,000 or 9,000 votes, both lists would have lost. Balancing the votes evenly would have been an even more difficult challenge for the Civic Party, because they intentionally gave newcomer Chan Ka-lok the first spot, placing incumbent legislator Tanya Chan second. Splitting votes between a well-known quantity and a political newcomer would have been a very risky proposition.

The one-list strategy also gave the Civic Party only one seat in New Territories West. But it’s more complicated than that….results below:

In New Territories West, the Civic Party placed popular legislator Audrey Eu second on their list behind former LegCo member Kwok Ka-ki. They again finished first in votes, ahead of the DAB, but received only 1 seat where the DAB received two. Total disaster struck the Democratic Party, which ran two lists and saw both just miss the threshold, resulting in no seats: the worst possible outcome.

Had the Civic Party ran two lists, and the Democratic Party one list, this would have been the result:

On the surface, this scenario gives the Civic Party two seats, but this would have been nearly impossible, given how thin the margin between them and the FTU would have been. A mis-allocation of just a few hundred votes would have lost one or both seats! Moreover, the Democratic Party ran two lists: had both the Democratic Party and Civic Party run two lists, aiing for four seats, it’s quite likely that they would have ended up with zero seats between them even as their votes outnumbered the DAB.

In the end, the Democratic Party chose a risky strategy to win extra seats, and was punished. The Civic Party chose a safer strategy, hoping for four seats in these two constituencies, but comfortably elected two instead. The DAB chose the same strategy as the Democratic Party, but won more votes and split them effectively between two tickets for 4 seats. Noted, again that the Civic Party won more votes than the DAB in both constituencies.

These are not strategic blunders by the pan-Democratic camp. The selection of the Hare quota (which, again, has been abandoned in general worldwide) was a deliberate choice by Beijing after the 1997 handover. The DAB is widely known to have the organizing force of the mainland Chinese government behind it, and as such is much better able to execute the machine-like maneuvers these strategies require.

The 1991 and 1995 Legislative Council elections for the geographical constituencies were held using American-style first past the post, single member districts, like our US House of Representatives. Both elections resulted in overwhelming majorities within the geographical seats for the United Democrats of Hong Kong. It is well-known that first-past-the-post encourages the formation of two monolithic parties (Duverger’s Law), apparent in the tight grip that the two-party system has had on the United States for two nearly uninterrupted centuries. After the handover, the current system was put in place, which encourages parties to splinter and rewards tight organization.

American liberals have long lamented infighting in the Democratic Party, and sometimes gripe that conservatives are naturally more likely to fall in line. While this isn’t always true, nobody falls in line like the pro-Beijing DAB supporters. Who else could be more obedient than sympathizers to a one-party Communist China? Even as the party-list/highest-remainders system puts immense pressure on large parties to split up, the DAB has remained intact. Instead of a two-party pro-democracy/pro-Beijing split, Hong Kong has one dominant pro-Beijing party leading a handful of smaller ones opposing a splintered constellation of opposition parties.

A complicated election to a rigged legislature using a voting system designed to help well-organized establishment parties over more fractious pro-democracy forces: a democracy that even the Chinese Communist Party could love.