30-all is deuce

June 23, 2011 by Ed Cormany in Uncategorized

if i ever have a tennis blog (please don't let me do that), this is my first choice of title. now, i know what you're thinking: "hang on, i remember learning tennis scoring, and it's definitely 40-all that's deuce." well yes, but it turns out that actually they both are. need proof? from either score, there are four possibilities of what can happen on the next two points. server wins both 30-30 -> 40-30 -> game 40-40 -> ad in -> game

server wins one, returner wins one 30-30 -> 40-30 -> 40-40 40-40 -> ad in -> 40-40

returner wins one, server wins one 30-30 -> 30-40 -> 40-40 40-40 -> ad out -> 40-40

returner wins both 30-30 -> 30-40 -> break 40-40 -> ad out -> break

fine, whatever, it's a quirk of terminology. what's the big deal? well, the big deal is that match commentators treat these two scores very differently, when they are theoretically identical. 30-all is frequently called "an opening", "an opportunity", "a half-chance", among other turns of phrase that are full of optimism for the returner's odds of breaking serve. (less commonly they frame it as trouble for the server, but the direction of bias is still the same.) but 40-all is just deuce, and treated as relatively equal. if anything, the server's general expectation to hold biases the commentary in their favor at 40-all.

i'm really baffled what has created this altered perception of 30-all vs. 40-all deuces. the only objective difference is that at 40-all deuce, one of the two players has necessarily missed converting a game point. someone has already had a real opportunity, so perhaps speculating on half-chances and maybes is thrown away then? but the reality of what could be and what must be done remains unchanged. i wonder how many players even realize the equality of the two scores, and in terms of psychology and strategy, would treating them as the same or different produce better results?

everything without representation

June 18, 2011 by Ed Cormany in Uncategorized

i swear, despite the topics of my first two contentful posts, this is not going to be a politics blog. because, well, for the most part i'm rather apathetic to politics (i believe my political views on facebook used to be listed as "Congress sucks" and is now "all legislatures suck"), and i think this post will spell out one of the reasons why. my friend Michael Cowett tweeted a link to his debut column, which deals with redistricting in Massachusetts. my tweeted response was pretty pithy:

this, of course, got me to going through said rethinking. so what's the big problem with our local Representatives to Congress? well, primarily the fact that they're neither local nor representative. that, after all, is the beauty of gerrymandering. i can only imagine that those who actually serve on redistricting committees are either a) grinning like small children who when told not to leave the room are hanging 99.9% out the door with a toe barely making contact with the threshold, simultaneously mocking their parents or b) not doing that, and therefore completely soulless.

so, since fixing both shortcomings is way too pie-in-the-sky for me to even bother making the clackity noise about, let's look at how either one could be fixed. hypothetically. very hypothetically.

locality

according to the 2010 census data, each congressional district now "represents" almost 710,000 people. the founders (frequently trotted out by the far right as demi-gods, but regarded by me, a liberal [i suppose], as probably fairly smart guys, what with starting a country and all) allegedly argued bitterly between allotting 30,000 and 40,000 per district. of course they never could have guessed that the average congressional district would grow by double that figure over just 10 years at the turn of the 21st century.

so, what to do? there have already been proposals to increase the size of the House — something that occurred regularly before 1929 — although most of them are modest. what if we reapportioned fairly conservatively? at, say, 100,000 constituents per district we would be representing at roughly the level of the Italian parliament…with 3,000 representatives. go back to the founders' plan and we'd be looking at 9,000 representatives.

seeing as we have no plans to build the Galactic Senate (which, i might point out, is on the very short list of legitimately cool things from the Star Wars prequel trilogy), this is pretty much a no go.

representativeness

the opposing alternative is to scrap locality in favor of better representing constituents with roughly the number of representatives we already have. most people couldn't possibly tell you what their congressional district looks like or what it includes. it may well sprawl halfway across a state, like my current home district, NY 22.

so, the easy solution? keep the current number of seats per state, and hand them out on a proportional basis, state-wide. this is easy in theory only, of course, since it would essentially destroy the two-party system in one fell swoop. given that those in control of this situation are some of the strongest adherents to said system…yeah, this isn't happening.

both

ok, i lied. once i'm indulging impossible fantasy scenarios, i might as well push it to the limit, no? what would an ideal solution look like? probably one that increases the number of representatives to a much larger but still manageable number, close to but under 1000. then, districting is done completely blindly by some algorithm that has access to census data but not the biasing parts: party affiliation, race, etc. cook up a few competing algorithms, generate their results, and then vote on the overall plan. the benefits? probably long-term lowered costs for having properly compact districts. a bit of a solution to both problems above. the drawbacks? two-party chaos or dissolution of the two-party system.

in sum

nobody's going for it. we walked down this ugly path and now American politics is stuck in it. like i said, it sucks no matter who it favors and no matter which side you're on. cheery, i know.

so, to follow up that in terms of implausibility, i think my only possible next topic is to discuss my awesome ideas for college football playoffs. until then.

playing the quorum game

June 16, 2011 by Ed Cormany in Uncategorized

earlier this week, Italians voted on a slate of four legislative referendums. for each the question was to overturn a law passed by the current ruling party, led by Silvio Berlusconi. (if you don't follow Italian politics closely, the last you heard of him he was probably either getting indicted for paying an underage prostitute or getting smashed in the face with a model of the Milan cathedral.) i found out the news because i read the headlines of La Repubblica every morning. however, i didn't really know anything about the referendums until the day of the voting. this is mostly due to the fact that Italian journalistic style makes it as difficult as possible to actually keep track of what's going on. back page stories become front page stories with no fanfare, introduction, or exposition, and if you haven't been reading cover to cover (and how do you even know when it's all online?), then too bad. add in the fact that their politics coverage reads like Variety even when it is the top story, and it can be a bit challenging.

nevertheless, it was obvious that i was looking at election results, and the key bit of news was that there was a quorum. after a crash course on Wikipedia i found out that these votes, in order to be valid, have to have 50% turnout of all eligible voters nationwide — something like 25 million votes!

i checked back later in the day, once they'd started tallying yes and no votes, to see the percentages. all four referendums were passing (where passing means to repeal the law) by about 95% to 5%. this result was billed as schiacciante, 'shocking'.

but was it really? i thought about it and realized, that no, Italians either understand or have been well-coached on how this fairly complicated electoral process works. because of the quorum rule, there is no potential benefit in casting a "no" vote.* once the quorum is met, a simple majority wins the contest. so in perfect conditions, 25%+2 votes are the minimum to pass the referendum (50%+1 of eligible voters cast ballots, and of those ballots 50%+1 marked "sì"). taking the real numbers from this week's election makes the danger all the more evident. if, for example, turnout had been a few percent different and the quorum was only 51%, the "no" votes would have contributed to passing the measures. if they abstained, the quorum would not have been met, and the measure would fail. that didn't happen, since an absolute majority of the electorate voted "sì" and thus guaranteed victory, but it shows just how tricky the situation was.

so was it shocking? from a big picture, political perspective, perhaps. maybe Italy is finally ready for some real change away from the craziness and corruption of the Berlusconi governments. but from a game theory or strategic perspective? not at all. both the ayes and the nays knew exactly what they were doing, and the ayes had it.

*hypothetical exceptions:

there is a historical tendency of very high voter turnout regardless of position (not improbable given Italy's 90% turnout in parliamentary elections – yes, the USA is at the absolute bottom of that table with 48%)
nobody gets it and everyone who wants to vote "no" has to do so just to try and make sure that all of his dumb compatriots don't screw it up for him. this quickly devolves into paranoia and/or a lose-lose situation.

blogging, for real this time

June 16, 2011 by Ed Cormany in Uncategorized

i'll admit it: me starting a blog in 2011 seems a bit daft, at least on the surface. blogs were at least two waves of internet communication ago, right? but i'm not doing this to be fashionable, and as far as i can tell, people do still read words on the internet. here's the simple reason that i'm doing it: i need a venue for medium-format writing. at this stage in my life, i'm writing in two sizes: dauntingly long (read "dissertation") and very short (read "tweets"). everything else falls between the cracks. dozens of interesting ideas have simply gone down the drain after i realize that they're longer than 140 or 280 characters. maybe i'll expound on whatever i'm thinking about to my officemates or friends at the bar, but if nobody's around, i think "oh i'll get back to that", which is like filing my notes in the shredder.

in the past, blogging has been something i've desperately wanted to do, but also something that, without fail, turned into a chore at some point. in retrospect, i've experienced two clear causes for blog-killing malaise.

being on a schedule. nothing kills my desire to see the new post window than requiring daily writing of myself. even worse if it's about time-sensitive information. if i skip a day, suddenly i have a backlog, imperfect coverage, and guilt.
Blogger sucks. hardcore. i don't know how many of my linguablog posts never made it because i got fed up with piece of crap editing software. i'm on WordPress now, which i've liked when i've used it in the past. (all of my WP blogs have died due to reason (1) above.)

that's basically as far as i've planned this. if you like the random assortment of content that comes through my twitter feed, or you like me in person, you'll probably like what i write here. the rest of the internet is welcome to jump on board too, but i have no expectations that they will. let the bloggery begin!