UsableMarkets

markets, design, usability, research

UsableMarkets header image 1

Sweet sports visualization: what it’s like to face Mariano Rivera

July 31st, 2010 · Information Design

Mariano Rivera of the NY Yankees has one pitch, but it’s an awesome pitch. The NYTimes does a sweet job of explaining its nastiness.

rivera1
rivera2

rivera3

~alex

→ No CommentsTags:

Beware of chart bergs

July 20th, 2010 · Book Reviews, Information Design

Yes, information visualization is heating up. With new wonderful software tools (like Tableau Public), and languages (like protovis) it seems that many can get into the game, (including yours truly, but perhaps to your detriment). And to satisfy this interest there are multiple new books, some of them quite good.

I’ve read a few of them of late (such as Now You See It and The Wall Street Journal Guide to Information Graphics), but I think the best one so far is the one I’m reading now, Picturing an Uncertain World. Perhaps counter-intuitively is does not start off with any charts or graphics. Rather, it zeros in immediately on the hidden part of any chart: the raw data.

Like an ice berg, any good chart is supported by a mass of raw numerical data. There is a fair amount of thinking that goes into selecting the data, gathering the data, aligning the data, and so on. And while this geeky grunt work is none to fun, if you don’t know the rules of the road you can easily end up selecting wrong and misleading data, and therefore creating a misleading chart.

Howard Wainer, the author of the book, starts with a discussion of standard error. He highlights what he calls “De Moivre’s Equation,” which helps us determine sample sizes to protect us against errors of sample size. And if that is confusing to you, it can perhaps be best summed up this way: small sample sizes give way to greater variability of results, making it more likely that your results will be misleading. If you aren’t aware of this issue then you could end up charting things like: boys are smarter than girls; and small schools are better at educating our children than large ones (both of which are untrue, just so we’re clear).

And while I haven’t even read half the book yet, I already highly recommend it to anyone who has anything to do with presenting data. While grabbing any old data set and firing up Tableau Public (or whatever your tool of choice is) can be fun like hellfire and hotcakes, it is the process of thinking through the data, and being aware of its foibles, that really helps make a chart meaningful.

Don’t sink the Titanic!
titanic

~alex

→ No CommentsTags:

The Baltic Dry Index. A misleading economic indicator …?

July 19th, 2010 · Economics

For those of you who follow such things, it has certainly been interesting to note the decline in the Baltic Dry Index (or BDI). I have, for the moment anyway, a chart of the BDI in the right column of this blog. A quick glance will show you its direction. Namely: down.

I wasn’t the only one who was wondering at the decline of the BDI, and stories from The Economist and FT’s Lex discuss the meaning behind it.

The Economist says:

There are growing doubts, however, about what the Baltic Dry is actually signalling. The confusion is whether the index is saying more about the supply of ships than the demand for their cargoes. The index spiked dramatically in 2008 as China’s imports of commodities soared at a time when the supply of ships was constrained and port congestion added to demand for capacity (see chart). The financial crisis soon caused the index to fall back but not before this period of dramatic growth in demand from China had prompted a surge of orders for bulk carriers, especially the very largest ones that are used on the China trade routes.

These ships take around three years to come on-stream. Despite the cancellation of some orders the new ships are now flowing in: in the first half of this year the global fleet increased by 23% as new vessels came into service at the rate of 16 a month. There are now 23 such vessels arriving each month, adding to oversupply.

So, oversupply of ships? The FT’s Lex concurs:

Several analysts have poured cold water on the BDI’s prescience though, given some obvious distortions. It takes about three years to build new capacity and, as the peak of the commodities bubble was two to three years ago, new ships are now depressing charter rates. The global fleet has expanded by about a quarter this year alone. For now, other indicators of commodities demand trump the BDI.

Cross-checking with ostensibly similar global measures such as air freight sheds little light on the matter, but rail traffic compares more reliably. Trains carry the same types of materials as shipping and volumes rather than tariffs are reported while capacity changes only slowly. American carriers provide the most detailed and timely data and they continue to report outstanding year-on-year gains for tonnage of key industrial commodities such as metallic ores, coke and lumber, up 93 per cent, 28 per cent and 11 per cent for the first 27 weeks of the year, respectively.

However, both articles hedge their bets. With the threat of a double dip recession, that’s probably smart. And if it comes to pass, the BDI may not be so misleading after all.

~alex

→ No CommentsTags:

Whither prediction markets?

July 13th, 2010 · Prediction Markets

It really didn’t seem that long ago that prediction markets were world beaters. Prediction market supporters knew they had the best predictive mechanism around, and they were not shy of saying it. They could predict presidential elections, Oscars, and average yearly rainfall in NYC, among other things. Prediction markets were even floated as a way to make legal and policy decisions. A few years ago there was a sense that prediction markets were about to take off. Have they?

Well, it’s true, they haven’t taken the world by storm, but, based on my (extremely) casual observations, there appear to be two main themes in the prediction markets arena of late.

1. Fewer, but more serious, offerings.
One sees fewer of the random prediction markets that used to pop up all over the place. Yootopia, Washington Stock Exchange (WSX), ProTrade, PicksPal, and so on. Even News Futures seems to have gone the way of the Dodo. But that doesn’t mean that new prediction markets aren’t happening. Witness the Hollywood Stock Exchange finally getting its real money market. And Mr. Pennock’s Predictalot.

In other words, fewer prediction markets are coming from hobbyists and idle entrepreneurs. Now it appears that new prediction markets have real money behind them, or at least a serious agenda.

And if they’re happening in the corporate world, well, we’re just not hearing about them.

2. (Even) more skepticism.
Take this recent quote from Constructive Economics.

I ultimately believe that a real-money HSX is unnecessary. These markets are supposed to provide a way to hedge risk on people seeing films (of course, shouldering such calculated risks should be the whole purpose of a studio in the first place, but regardless…), but if I were a studio with exposure risk on box office returns I would shop that risk around to creative hedge funds with good models willing to take an equity gamble. The two big things a market provides, consolidation and anonymity, are just not necessary in this scenario. On every contract there’s just one movie and one seller of exposure risk.

Certainly there was always intellectual rigor in the prediction market community, but that always seemed to take the form of true believers debating who’s using the best formula for automated market making. Now people are questioning whether a prediction market should even exist! Egads, where have we gone?

Furthermore, some of this sort of talk is coming from our prediction market luminaries (or at least under their guidance). Take the recent paper (pdf download) by Sharad Goel, Daniel Reeves, Duncan Watts, and Dave Pennock. Here they assert that, sure, prediction markets are better are predicting the future than other methods, but not so much that it makes a difference. In other words, why waste you time setting up a prediction market when you can just use a (gasp) poll!?!? The results will pretty much be the same.

Somewhere I feel like I can hear John Maloney talking about industry maturity, but that doesn’t quite seem right. Prediction markets, as an industry, doesn’t feel any further along than three years ago. Which is quite sad, since I think some people had high hopes selling prediction markets to all the companies in the world.

Maybe it’s the recent recession. Maybe it’s that I just don’t care as much as I used to, but prediction markets still seem like they have a ways to go before they are widely considered a useful tool to people other than gamblers, media companies looking for a quick story, and academics.

~alex

→ No CommentsTags:

Predicting recessions

July 12th, 2010 · Economics, Information Design

While leading economic indicators get a lot of the press around predicting the short term future of the economy, there is one method of predicting recessions that has a pretty good track record: an inverted yield curve.

An inverted yield curve happens when long dated US treasury bonds yield less than short dated ones. Typically the yield curve inversion is calculated by using the 10 year US treasury bond and the 3 month US treasury bill.

Why do they invert? Think of it this way. A lower yield on the 10 year bond means that investors are more comfortable tying up their money for a longer period of time than for a shorter period of time. In other words, people fear that bad things might happen to the economy in the short term, and they want to have lots of cash around to protect themselves.

A simple chart illustrates its record of predictions.

10 year treasury bond yield – 3 month treasury bill yield, since 1962, mapped against US recessions
updated recession prediction chart
click for larger size

There’s plenty of economic brainpower to back this all up, and plenty to wade through on the internet if you so wish.

This has been charted before, of course, but I took a shot at redesigning it because the ones put out by the Fed just seemed to leave something to be desired. The one below is a typical example. My goal was mainly to show what adding a bit more visual information could do to help bring out the correlation between yield curve inversion and US recessions.

feds recession prediction chart
click for larger size

~alex

→ No CommentsTags:

Worth quoting: Inequality caused the recession

July 12th, 2010 · Economics

Raghuram Rajan, my new favorite economist, has an interesting piece on the Project Syndicate website, How Inequality Fueled the Crisis.

The crisis in the title of course refers to our recent recession. But then the crisis we’re in can perhaps be perceived as something more long lasting. At UM we’ve talked about the income gap before, which is where Rajan begins.

Since the 1970’s, wages for workers at the 90th percentile of the wage distribution in the US –such as office managers – have grown much faster than wages for the median worker (at the 50th percentile), such as factory workers and office assistants. A number of factors are responsible for the growth in the 90/50 differential.

Perhaps the most important is that technological progress in the US requires the labor force to have ever greater skills. A high school diploma was sufficient for office workers 40 years ago, whereas an undergraduate degree is barely sufficient today. But the education system has been unable to provide enough of the labor force with the necessary education. The reasons range from indifferent nutrition, socialization, and early-childhood learning to dysfunctional primary and secondary schools that leave too many Americans unprepared for college.

The everyday consequence for the middle class is a stagnant paycheck and growing job insecurity.

Rajan also asserts that this equality is not something that people (read politicians) don’t know about. It’s just something they won’t talk about.

Cynical as it might seem, easy credit has been used throughout history as a palliative by governments that are unable to address the deeper anxieties of the middle class directly.

Politicians, however, prefer to couch the objective in more uplifting and persuasive terms than that of crassly increasing consumption. In the US, the expansion of home ownership – a key element of the American dream – to low- and middle-income households was the defensible linchpin for the broader aims of expanding credit and consumption.

Which has led to the result we’re all to familiar with.

The problem, as often is the case with government policies, was not intent. It rarely is. But when lots of easy money pushed by a deep-pocketed government comes into contact with the profit motive of a sophisticated, competitive, and amoral financial sector, matters get taken far beyond the government’s intent.

So, does someone need to break our current education system and put it back together?

~alex

→ No CommentsTags:

NYTimes charting overkill

July 10th, 2010 · Fun, Information Design

The NYTimes has rightfully been making a name for itself in the world of infographics. Many of their economic and sporting graphs have been sublime. But have they gone overboard?

I only ask because this chart which is supposed to outline how the European teams started the World Cup poorly, but have now come on in the end, seems to me a bit gratuitous.

nytimes world cup graphic

While pretty, the chart seems more like a bunch of collected statistics, thrown together, which doesn’t really do much to support the main assertion. Using the small multiples to chart each game in the tournament can be endlessly distracting, but where does it lead us?

To avoid being just negative, though, I should point out that the NYTimes live graphical updates during the World Cup games has been quite powerful. One does get an excellent sense of how a game is progressing (especially if one does not have the benefit of a video feed or TV), in a very small amount of space.

nytimes game graphic

~alex

→ No CommentsTags:

US Monthly (Annualized) Car Sales Data, 1996 to present

July 7th, 2010 · Information Design, Travel Markets

car sales data

click to SUV-ize

Source, California State’s Department of Finance

~alex

→ No CommentsTags:

An ugly graph but a powerful message

July 7th, 2010 · Economics, Information Design

nytimes image

From the NYTImes, Economix.

The chart title — Percent change of constant-dollar median usual weekly earnings by educational attainment and sex — is a mouthful, but the message is simple:
1. Men without a 4 year college degree likely earn less than their fathers
2. Men with a 4 year college degree likely earn more
3. Women are likely earning more than their mothers, regardless of education (except for the most poorly educated)

To wit: get yourself edumacated.

~alex

→ No CommentsTags:

TED Spread vs. US bank failures, 2000 to present

June 28th, 2010 · Economics, Information Design

ted vs bank failures, 2000 to present

The TED spread is the difference between the interbank loan rates and the short term treasury rates (3 month LIBOR – 90 day Treasuries). The higher the spread, the less faith there is in the strength of the banks.

~alex

→ No CommentsTags: