You are currently browsing the tag archive for the ‘tldr’ tag.

It seems like discussion of Piketty’s Capital has run its course and much of the commentary has moved on (though not necessarily from the broader topic) so now is as good time as any to peer back and reflect on how the debate around the book ended (if such a thing can be summarized). From my own vantage point, the debate about the book (not necessarily the discussion) stalled out around a single question, so I will do my best to restate and clarify that question so as to focus where more evidence and argument is needed, should this be a conversation anyone wishes to resume. None of this is new, exactly, but it’s worth recanting given the importance of the question and the stakes surrounding it.

Around 1800 AD, living standards in some countries began to rise substantially, and over the past 200 years, that rise (as measured in GDP per capita) has been on the order of a factor of 50. This generally seems to correlate with other indicators of increased living standards to a degree that, with some exceptions (such as thinly-populated resource-rich countries) it is generally, though not universally, accepted practice to use GDP per capita as a good-enough shorthand for broad living standards. Whatever the case, exactly how and why this increase transpired is still a matter of debate, in no small measure because most people would find it desirable to replicate the phenomenon in those areas that have not yet experienced it. Indeed, some countries that did not begin experiencing the phenomenon in its initial emergence have experienced it since, leaving, essentially, three groups of countries – those who have experienced it, those who have not, and those in transition.

Piketty’s book, while not exclusively, overwhelmingly is focused on the first kind of country. A compelling portion of his narrative is documenting that transformation, yet the broader focus of the book is on what has transpired since that transformation was consolidated in the era following the Second World War. There are two key factors to be documented. The first is that the countries that have fully experienced this transformation are themselves not ‘complete’ in this regard – average living standards (recent economic troubles excepted) continue to rise and are generally, though not universally, expected to continue to rise in the absence of extreme calamity on the scale of global catastrophic climate change. The second is the change in the distribution of income – since a moment of ‘peak equality’ in roughly 1970, most of the countries Piketty analyzes have seen a sharp increase in inequality, the specific degree of which dependent on method of measurement but whose general contours is not really disputed. This, Piketty and many other believes, poses a problem for these countries that is not alleviable solely by continuing increases in average living standards or aggregate wealth and income growth.

Piketty devotes a lot of space to developing a simple model of how the aggregate quantity and distribution of capital can drive income inequality. This remarkably simple model requires only three input variables – the growth rate of the economy, the average return to capital, and the savings rate (perhaps better phrased as the rate of capital formation relative to national income) – to generate a long term prediction of two key ratios: the ratio of capital to income, and the capital share of national income. From there, wealth inequality can be used directly to compute a floor on income inequality – for example, if 1% of the population owns 50% of the national wealth and the capital share of income is 30%, then that 1% captures, at a minimum, 15% of national income.

And here we arrive at the crux of the debate. Piketty’s model implicitly assumes a certain exogeneity between those three input variables and the two ratios they converge towards, ie, that they are not inherently correlated with each other. This exogeneity poses a fragility in Piketty’s model and a challenge to mainstream economic theory. The fragility is that, if they are strongly correlated (in the direction such correlation is expected), and especially if there is iterative feedback between them over time, then Piketty’s model no longer produces outcomes in which wealth inequality drives income inequality. The key example here is the average return to capital; were it to fall in proportion to the rise of total capital accumulation, then the capital share of national income would be invariant to the quantity of capital, and thus largely undermine the mechanism by which present wealth inequality drives future income inequality. Furthermore, were this anticipatable decline in the in return to capital to drive a decline in savings, the capital/national income ratio would converge at a substantially smaller value than that projected by extrapolating from the initial period. This further depresses the likelihood of ever-increasing wealth-driven income inequality.

This is also precisely the challenge to mainstream economic theory. These correlations and feedbacks are precisely what are predicted by fundamental, strongly-held ideas about economics held by economists; most centrally that investment behavior is driven by that most central economic force, supply and demand. Piketty, however, is not simply laying down an alternative model, but an empirical challenge to this challenge. The most crucial assertion made by his model – that the return to capital fails to decline in proportion to the supply of capital – is not simply a theoretical alternative but one derived from the meticulously researched and calculated estimates in his unprecedented data. As I myself pointed out in my write-up of Piketty’s book, the data show that the return to capital is sufficiently resilient to its accumulation to justify Piketty’s model. At least, that is, without controlling for any additional factors.

And here is where debate stalled, with one side asserting that theory demands these variables be tightly correlated, and the other side responding that empirics demonstrates that they are not. The problem, of course, is that macroeconometric panel empirics is extremely sensitive to model specification, to the point of being perhaps the perfect example of how any decent statistically-versed researcher with strong priors can generate the outcomes from the data they which to receive. Certainly it is more than possible to generate a superfluity of complex models demonstrating the theoretically-predicted correlations, and these models will collectively have zero persuasive power because it is trivially easily to create as many or more equally-plausible equally-complex models that demonstrate the obvious.

Why does this all matter, to the degree it’s worth recounting in such detail to the tune of a thousand words? Because it strikes directly at the heart of the most important argument for tolerating high income inequality.

There are basically three arguments in favor of tolerating high income inequality, which I will attempt to summarize as fairly as I can.

  •  The ‘Just Deserts’ Position: incomes reflect the inherently just outcomes of markets. Beyond a certain threshold to prevent the worst form of miseries, it is therefore a violation of justice to take from the deserving and distribute to the undeserving.
  • The ‘Pink Salt’ Position: income inequality is irrelevant except to the irremediably envious, resentful, or spiteful. What matters is preserving and increasing human happiness, which is largely driven by civil liberties, non-market institutions such as family and community, and the secondary impacts of economic progress.
  • The ‘Golden Egg’ Position: income inequality may be ceteris paribus bad but aggregate economic growth is extremely good to a degree that in most plausible scenarios swamps income inequality. Furthermore, income inequality and economic growth may be conjoined outcomes of our economic system and cannot be modified independently. Therefore, we should be extremely cautious about attempting to alleviate income inequality through policies that slow the rate of economic growth, as this may reduce not just aggregate utility but the utility of those benefiting directly from redistribution.

It will shock nobody to hear that I reject outright the first argument in the strongest possible terms, and the second in quite strong terms as well. Indeed, I believe that the majority of Americans, and certainly the majority of voters in developed countries, disagree with those arguments as well. It is that third argument that gives pause to many – including, to a degree, me (though that pause is still far from convincing in my own case). The average person living in a developed country today as compared to a person living in that same country in 1800 is vastly better off, and it is not impossible to imagine that the average person living in a developed country in 2100 will be vastly better off than that average person today. Impeding our shared progress in that regard could simultaneously defer developments that improve the quality of most lives while simultaneously deferring developments (like innovation in renewable energy sources and storage) that could mitigate or reverse the worst consequences of economic growth to date.

This all converges on something of an ironic surprise. In this debate, it has been the left that has been advocating, implicitly or explicitly, on behalf of the resilience of capitalism (broadly defined) and its ability to deliver human prosperity, whereas it has been the right that has claimed, implicitly or explicitly, that capitalism and the prosperity it delivers is fragile, so much so that even increasing post-market redistribution (as opposed to pre-market regulatory redistribution through minimum wages, stronger protections for unions, and abridging the current rights and privileges of lenders and shareholders) could, to use a tired aphorism, kill the goose that lays the golden eggs. This ideological positioning isn’t wholly novel, and whether it is instrumental and ephemeral or representative of something larger remains to be seen; but it is notable, and worth pondering for what it says about the state of both the contemporary mainstream left and right movements in the United States (if not beyond).


My post last week on the case for homeownership as an investment has received some good feedback (the e-word is hereby banished from this blog), a good chunk of which has been constructively critical. While I responded to specifics in comments, I also wanted to supplement the post by fleshing out the remainder of the argument and adding a couple of points.

It has been pointed out to me that there are certain costs – mostly taxes, insurance, and maintenance – that weren’t included in my spreadsheet and only implicitly in my analysis. This is – for the most part – true! I did handwave away depreciation, as much for the sake of simplicity as anything, but I only touched on the other two to the extent that they’re wrapped up into the rent counterfactual. Let’s delve into that some more.

Rent – the price of shelter to non-owners – is in the simplest analysis driven by the same things that drive all markets prices: supply and demand. That means rents aren’t directly responsive to the costs of housing, but those costs do impact the supply curve. If the costs of creating and renting new housing can’t be justified by rents, then supply will not rise even if demand does, driving up prices until they are so justified. Therefore, in general we should expect the costs of renting shelter to be similar (though not equivalent) to those incurred by the owner of the same. In fact, I bet if you play around with The Upshot’s ‘Buy vs. Rent’ calculator, you’ll find that housing and rental costs are very similar.

This brings me to my next point; while people have pointed out what costs I didn’t include, fewer have mentioned the benefit I didn’t include in my analysis, even though that benefit is much vaster. I focused solely on the capital gains returns of buying a house to demonstrate the power of leverage, but the huge share of the returns to a house are the rents you receive as an owner. This is central to any complete case in favor of homeownership. It is further worth noting that these imputed rents are, in fact, an enormous share of our economy.

a vampire weekend song entitled "david ricardo"

Net imputed rents, as I pointed out in my Piketty thinkpiece which seriously you must have read this thing by now also tend to be fairly stable, returning between 4-6% of the house’s price over time.

seek those rents seek them hard 

This chart actually understates the stability of imputed rents (as the former chart makes clear) since most of that volatility is driven by volatility in the denominator. For context, here’s the Case-Shiller index, since basically forever (with bonus real interest rate series):

 oh hai the aughties

While volatility has more recently increased (consider that my application for the Understatement of the Year Award), note that houses, at the very worst, tend to be inflation proof (the Case-Shiller is a real, not nominal, index) – an asset whose nominal price grows alongside inflation while consistently returning 4-6% annual net returns is, hey, not too bad, and if you can use tax-privileged leverage to buy it, not too bad at all. Especially since we’re going to pay a bundle for housing no matter what we do:

 what would a cake chart look like

…using housing as a vehicle for savings makes an additional sense.

That leads me to an additional point on volatility; here’s Shiller’s stock price index, also since basically forever:

 oh hai the 20s oh and the 90s hello

That looks a lot more volatile than house prices, huh? Which brings us to a key point – as asset price volatility increases, so does the importance of investment timing. This, as Neil Irwin recently noted, can make long-term averages of returns misleading.

seriously though fuck umberto nobody likes that guy anyway

While his examples are obviously stylized, they clearly-enough make the point that otherwise-identical savings behavior in a volatile market can achieve vastly different outcomes depending on the timing of returns even holding long-term average returns constant. Therefore, the relative stability of housing returns – prices + rents – helps savers reduce long term risks.

I want to conclude, though, by taking a major step back and examining the whole purpose of this exercise. When we’re talking about savings from a consumer perspective (not from an investment perspective) what we’re talking about is retirement; and when we’re talking about retirement, we’re always talking about the same somewhat-odd phenomenon. When a person retires, they cease all economic production through labor, yet continue to demand a share of the economic output of their society. We tend to view these claims as just and deserved because they are made by the elderly, who we feel have earned it/are unable to work/are generally venerable (as opposed to similar claims from the non-elderly poor, which we treat very differently) but that doesn’t change the underlying structural nature of the phenomenon, in which we are trying to ensure that a substantial portion of the adult population is consuming an broadly-equally-substantial portion of present economic output while providing no inputs.

Debates about savings and retirement, therefore, are all about how to structure this phenomenon – specifically, what network of programs, policies, mechanisms, incentives, and behaviors we want to establish to justify to the working and capitalists that a portion of their labor and capital outputs be directed to the non-working old, which we often do by creating mechanisms that somehow tether those portions of redistributed present income to guarantees of future income. All governments in wealth nations do this, and the ways in which they vary are influenced heavily by politics, ideology, and other socioeconomic factors. In the United States, our prevalent ideology around a certain kind economic freedom means we tend to be less generous in direct public redistribution and instead attempt to subsidize private savings through the tax code and public insurance – ergo, 401(k)s, the home mortgage interest deduction, and the Pension Benefit Guaranty Corporation. Indeed, the increasing prevalence of that ideological strain is driving defined benefit plans into extinctions in favor of defined contribution plans.

good thing we don't have unions anymore

This leads us to many debates about the best savings vehicles for middle class Americans, yet those debates are to a decent extent a red herring – the vast majority of retirees receive the majority of their retirement income from Social Security, and for many, it’s all the income they have – though to be consistent, I’m nearly certain the figures in the chart below don’t include imputed rents, though I could be wrong, and this is important because 80% of seniors are homeowners:

society secured 

This is very good evidence for the proposition that a vastly disproportionate share of the private-savings-for-retirement subsidy network flows to those who need it least. And it suggests that questions like “houses v. stocks” are, for many Americans, mostly a red herring – if we want to put more money in the hands of retirees, we should simply make Social Security more generous – or, in a better world, maintain it at its current level of generosity while implementing a Universal Basic Income.

Now that these particularly memorable twenty-second Winter Olympic Games have run their course, it’s time to review with an eye on answering two big questions:

-Who won Sochi 2014?

-What factors were correlative/predictive of success at Sochi 2014? (I am staying far away from making causal claims)

Worth noting before you read further:

-You should take a look at my last three posts on this topic first.

-In addition to many graphs and charts, this post will use choropleth maps, which use color to show differences. I wanted to add cartograms, which use (distorted) land area to show differences, but unfortunately ran into technical challenges I couldn’t really resolve to my satisfaction. There may be a follow-up post with some pretty cartograms if I can get it working right; in the meantime, excuse my choropleth.

-The second half of this post will feature a lot of regression results from Stata. If you’re not sure how to read regression results, I will summarize the key points, but you should also check out some of the resources out there on the internet for interpreting basic multivariate regression. It’s interesting and a useful skill! Seriously!

– Be forewarned that this is going to be a long post, so if this isn’t your thing, you should just skip down to the one about Francisco Franco. That was a good one. You could also check out this awesome interactive map from the Wall Street Journal that makes my maps look pretty lame, frankly, and I really shouldn’t link to them except integrity. Sigh.

And if you just want to skip this whole damn thing and cut to the chase, spoiler alert:

team lannister ftw

OK, let’s dive…er, slalom in:

Who Won Sochi 2014?

This question is less about data manipulation then the nature of the question itself. To determine who won, we have to decide the criteria for winning.

Most Medals

The simplest and most-widely-used criteria is most medals won. By that one, the results are pretty clear (note that all charts in this section exclude nations that failed to medal):

total medals

Winter Olympians 2014 Total Medals

Give this one to the home team. These results aren’t too surprising, but they do point to one reason these Olympics were interesting – nobody dominated quite like the US did at Vancouver, where they won 37 total medals.

Also, as we saw with the number of competitors sent to the Olympics, a small handful of countries really ran the show. Just six countries – Russia, USA, Norway, Canada, the Netherlands, and Germany – won over 52% of the 295 medals awarded at Sochi, and just eleven countries – those six plus Austria, Sweden, France, Switzerland, and China – accounted for over 75% of the medals.

But is that the best criteria? While it is a useful starting point as a raw, unweighted barometer of achievement, there are a lot of limitations to that measure. Let’s start digging into other measures.

Most Gold Medals

The second-most-widely-used criteria is most gold medals. By that one the results are also straightforward:

total gold medals

Winter Olympians 2014 GoldMedals

These results are also unsurprising – Ruskies win again – but they do start to point towards a more complex picture.

In some sense, this is also an intuitive criteria – the point of competing is winning! But that criteria also neglects a lot of considerations, first and foremost, the other two medals.

Most Quality-Weighted Medals

A compromise between/solution to the issues with those first two is to include all medals, but to weight silver higher than bronze and gold higher than silver. This process is inherently subjective; there is no obvious right answer to which system is best, and many systems can be used, some better than others, but none definitive. After some research, I personally developed a fondness for the system suggested by Robert Banks via Mario Livio, which weights the medals by the relative density of their metals; in this system a bronze medal is assigned one point, a silver medal 1.23 points, a gold medal 2.27 points. If you have another system you’d like to test, my data is included at the bottom of this post; feel free to meddle with the medal/metal formula.

total quality adjusted medals

Winter Olympics 2014 Quality Medals

These results show outcomes that roughly split the difference between the first two measures, which is deeply unsurprising – for the Amero-centric report, the US gets edged by Norway by eight-hundredths of a medal! But now we have to address some bigger issues before we can definitively declare any country “the winner.”

Medal Efficiency

The biggest issue regards the size of each country’s Olympic teams – there is an extremely strong correlation (though far from a perfect one) between the number of Olympians any country sent to the Olympics and the number of medals they won:

medal olympian scatter

Note two things. One, the “R-squared” of the trendline – how much of the variation in the data can be explained by a straight-line projection – is over .8. That means just predicting medals won by using a straight line projection from the number of Olympians a country sends will get you pretty close. The other thing is that one of those data points is orange. Note that data point.

Fortunately, this isn’t too difficult a problem to solve – we can just divide the number of medals won by the number of Olympians sent to devise a measure of “medal efficiency.” And lo and behold, these numbers look substantially different:

medal efficiency total medals

Winter Olympics 2014 Medal Efficiency

Here is where we start to get some real answers. By this measure one team in particularly is startlingly dominant – the Netherlands.

Indeed, if we look at gold medal efficiency:

medal efficiency total medals

Winter Olympics 2014 Gold Medal Efficiency

And quality-weighted medal efficiency:

medal efficiency quality adjusted medals

Winter Olympics 2014 Quality Efficiency

The Netherlands remains the dominant performer at this Olympics (also, Belarus looks pretty good – a little more on that later).

Event-Weighted Medals and Medal Efficiency

But! There is one big problem with medal efficiency, the one identified by James Harradence on Twitter last week:

Traditional medal counts award each country one medal per event, without consideration of the fact that many events feature multiple competitors, each of whom receives a medal. A country that sends more Olympians because it is competing in team events may be disadvantaged in a medal efficiency rating system that does not take team events into account (depending, of course, on what you are trying to measure – more on that later).

To break it down, of the 98 events at Sochi 2014, 25 were team events, and in total those 98 events at Sochi 2014 had a total of 203 competitors and an average of 2.07 competitors-per-event. I’ve weighed each team’s medal counts by the number of competitors in the events – the breakdown can be seen in the attached data set. Let’s see how this affects both the medal count rankings as well as the medal efficiency rankings.

First, here are the raw medal totals adjusted for the number of medalists:

total competitor adjusted medals

Winter Olympics 2014 Total Multi

Once you award Canada 46 gold medals for hawckey instead of just two, they start to look pretty rocking.

Indeed, here are competitor-adjusted gold medals:

total competitor adjusted gold

Winter Olympics 2014 Gold Multi

And competitor-adjusted quality-adjusted medals:

total competitor adjusted quality adjusted medals

Winter Olympics 2014 Quality Multi

Behold: the power of maple.

But what if we look once more at the efficiency of these medals? This again strikes at the question of what we are trying to measure, but for now again we can table that question and look at the results:

competitor adjusted medal efficiency

Winter Olympics 2014 Multi Efficiency

And once again Team Orange is on top. And when we look at competitor-adjusted gold medal efficiency:

competitor adjusted gold medal efficiency

Winter Olympics 2014 Gold Multi Efficiency

And competitor-adjusted quality-adjusted medal efficiency:

competitor adjusted quality adjusted medal efficiency

Winter Olympics 2014 Quality Multi Efficiency

The House of Orange prevails.

How Did The Netherlands Do It?

Two words: speed skating. Of the twelve speed skating events at Sochi 2014, the Dutch won eight of them outright and won 23 of the 36 available medals. Speed skating accounted for all but one of the Dutch medals at Sochi, the only other being Sjinkie Knegt’s bronze in…short track speed skating. I’m not an expert in Olympic history, but the Dutch dominance in the event this year was pretty remarkable. It was certainly an exceptional performance for the Dutch, who had never won more than 11 medals at any prior Winter Olympics; their haul at Sochi is ~22% of all Winter Olympics medals ever won by the Netherlands.

While the Netherlands have historically been excellent at speed skating (they’re the all-time medal and gold-medal leaders in the category, the former by a vast margin; conversely, they’ve only ever won five Winter medals in non-speed-skating events, ever) they’ve never been this dominant. It’s also not a function of new events; the team pursuits were first instituted at Torino and beyond that the same suite of speed skating events has been in place since Calgary. Whether it’s a level-up of the Dutch or a collapse of traditional competitors (USA, South Korea,, Canada), other analysts have put forward the case that the Dutch speed skating performance at Sochi was the most dominant national performance in any event category ever, or at least of the modern era. Just to cement the case, no country had ever “swept the podium” more than twice in a single Winter Games; the Dutch did it four times at Sochi.

So, yeah – go the Dutch. Keep on skating.

A Digression on Medal Efficiency

I think the focus on medal efficiency and competitor adjustment is useful, but it’s worth doubling back to a more foundation question – what is a country’s goal in the Olympics? Is it to:

1) Win the most events?

2) Win the most medals?

3) Do (1) or (2) within a resource constraint?

4) Some combination of all of those?

5) Other intangible goals?

Depending on how you interpret a country’s goal at the Olympics, some of these metrics do or do not make more or less sense. Depending on your resource constraints, it may not make sense, for example, to field a hockey team at all. This year the Dutch sent their largest-ever Olympic contingent, 41 athletes; but just ten of them won 23 of the 24 medals they accumulated. So how you judge who “won” the Olympics is largely dependent on not just the numbers but what you think the purpose of the Olympics is. You know, other than international harmony and sportsmanship.

What About Belarus?

So Belarus did pretty well, too – not as well as they looked during my mid-Olympic review but well enough to still stand out. How did they do it?

Answer: biathlete Darya Domracheva, who won a rockin’ three gold medals, half of Belarus’ total medal haul at Sochi. Her performance, while not unprecedented in Olympic history, was enough to render Belarus’ small contingent highly efficient, especially if the goal was to win events outright.

What factors were correlative/predictive of success at Sochi 2014?

Let’s zoom out and see if we can look at more generally what factors were associated with achievement at Sochi 2014.

As we saw above, the best way to win a lot of medals is to send a lot of Olympians:

medal olympian scatter

This finding is resilient to pretty much every measure of medals you want to use (and if all this Stata output makes your eyes glaze over, you can skip it, I promise)…

Screen Shot 2014-02-23 at 12.21.41 PM

Screen Shot 2014-02-23 at 12.21.31 PM

Screen Shot 2014-02-23 at 12.21.13 PM
Screen Shot 2014-02-23 at 12.21.08 PMScreen Shot 2014-02-23 at 12.21.18 PM Screen Shot 2014-02-23 at 12.21.26 PM

And even explains a pretty large share of the variation among just those countries that won at least one medal:

Screen Shot 2014-02-23 at 12.21.03 PM

Screen Shot 2014-02-23 at 12.20.57 PM

Screen Shot 2014-02-23 at 12.20.53 PM

Screen Shot 2014-02-23 at 12.20.13 PM

Screen Shot 2014-02-23 at 12.20.08 PM

Screen Shot 2014-02-23 at 12.20.02 PM

Note, though, that the association is consistently weaker when you look only at gold medals, a point we will return to later.

Now, what if we try to control for other factors? In my previous post I showed that population, GDP-per-capita, distance from the equator, and to some extent former Soviet Union status (as well as interaction terms) had decent explanatory power in terms of how many Olympians countries sent to the Olympics. And clearly the number of Olympians is highly predictive of winning medals. But do those factors, above and beyond their association with the number of Olympians, have any predictive power of medal winning?

As it turns out, not really:

Screen Shot 2014-02-23 at 6.08.18 PM Screen Shot 2014-02-23 at 6.08.13 PM Screen Shot 2014-02-23 at 6.08.05 PM Screen Shot 2014-02-23 at 6.08.00 PM Screen Shot 2014-02-23 at 6.07.54 PM Screen Shot 2014-02-23 at 6.07.47 PM Screen Shot 2014-02-23 at 6.07.42 PM Screen Shot 2014-02-23 at 6.07.34 PM Screen Shot 2014-02-23 at 6.07.27 PM Screen Shot 2014-02-23 at 6.07.21 PM Screen Shot 2014-02-23 at 6.07.15 PM Screen Shot 2014-02-23 at 6.07.05 PM

In almost all those models, we gained almost no additional explanatory power. As further evidence, here’s the F-test on the first model of all the additional variables, which in layman’s terms tests whether or not adding a group of variables to a model improved it:

Screen Shot 2014-02-23 at 6.14.41 PM

So to summarize thus far:

1) Knowing about a country’s income, population, climate, and former Soviet status can help you predict how many Olympians it sends to the Winter Games.

2) Knowing how many Olympians a country sent can help you predict quite well how many medals that country will win.

3) Above and beyond that, knowing all the stuff from (1) won’t help you predict medal outcomes.

What about medal efficiency?

Screen Shot 2014-02-23 at 6.02.50 PM Screen Shot 2014-02-23 at 6.02.56 PM Screen Shot 2014-02-23 at 6.03.02 PM Screen Shot 2014-02-23 at 6.03.08 PM Screen Shot 2014-02-23 at 6.03.14 PM Screen Shot 2014-02-23 at 6.03.19 PM Screen Shot 2014-02-23 at 6.03.25 PM Screen Shot 2014-02-23 at 6.03.30 PM Screen Shot 2014-02-23 at 6.03.35 PM Screen Shot 2014-02-23 at 6.03.39 PM Screen Shot 2014-02-23 at 6.03.44 PM Screen Shot 2014-02-23 at 6.03.48 PM

Ignore some of those low p-values; they will lead you astray. The more important thing to note is that the r-squared is generally low, meaning the model explains little of the variation, and the effect size is extremely small, so small as to predict almost no difference between the largest and smallest national teams. It does a little better in explaining the variance in competitor-adjusted models that include all countries, not just the medalling countries, but the effect size is still negligible.

Does this hold even if we include the kitchen sink? You know what; you don’t need all that Stata output. The answer is “yes, it totally holds.” You’re welcome.

Remember what I said above about the difficulty in proving gold medals? I believe that’s related to the difficulty in predicting medal efficiency. Without having much more intensive and extensive data, the models aren’t really able to explain the “x-factor” – an “x-factor” that is likely some combination of luck and national passion for a particular sport unexplainable by the other factors. There are lots of small, cold, rich countries, but none of them have specialized quite as intensively in speed skating as the Dutch. This is where Richard Florida’s conclusion, restated yesterday

The top performers, after adjusting for the size of each country’s population and economy, turn out to be smaller countries in colder climates. These countries tend to specialize in winter sports, and their Olympic programs lack the competition from big-time professional sports leagues that siphon off top athletic talent.

-really comes in in explaining why certain countries excel (or the opposite) beyond what these models would predict.

For those who want to dig more, here’s the data – have fun!

Winter Olympics Full Data Set

Moral of the story: next time my wife asks me a simple question, I will not proceed to make it so complex it requires ArcMap and Stata to answer.

Join 3,848 other followers

Not Even Past


RSS Tumblin’

  • An error has occurred; the feed is probably down. Try again later.