Tonight I was re-watching Usain Bolt's recent world record 100m performance in the 2008 Olympics and it got me thinking about physiological limits and human performance trends. What are the limits to human performance, and are we approaching them asymptotically, or are there jump-shifts in the performance functions, one where new performance levels are reached, and then eventually breached, and so on?

To get people thinking, here are two graphs. The first shows the progression of records in the Boston Marathon for men since 1927. Admittedly, there are a few issues with the data, not least of which is that the distance changed slightly over the period, but the main point is in what has happened over the last decade or so.

Now let's turns to the world record progression in the men's 100m. Here, courtesy of Wikimedia, is a similar graph.

What is interesting, at least to me, is that in the longer event, the Boston Marathon, there has been a clear plateauing. The record time hasn't dropped appreciably in more than 20 years (which is not the case for women, for reasons that Stephen Jay Gould goes into in his book Full House). That is interesting for many reasons, not least of which is that small percentage increases in training, technique or fitness would make a larger difference in that event, and yet we haven't seen it.

The situation is very different in the men's 100m. While there have been various performance plateaus reached, each time that plateau has been breached and a new performance regime entered. Most recently, of course, Usain Bolt's Beijing world record took us below the 9.7 seconds mark for the first time.

What is going on here? Is there a physiological explanation? A pharmaceutical one? And I am painfully familiar with drug history of men's sprinting, so I'm cheerlessly comfortable with the idea that we're seeing better times through pharma, not through karma.

All cynicism aside, however, why is there a continuing progression of performance in a shorter event, but a plateau in the longer event? Shouldn't it be the other way around? Because it looks like we have jump regime shifts in the 100m record, right up to present, but an asymptotic wall looms in human performance in the Boston Marathon.

Some further reading:

• Projected sprint performances are physiologically unreasonable (Medicine & Science in Sports & Exercise)
• World record progression 100 metres men (Wikipedia)
• List of winners of the Boston Marathon (Wikipedia)
• Bolt blazes to gold in 9.69s (NBCOlympics)
• Full House: The spread of excellence from Plato to Darwin, by Stephen Jay Gould (Amazon)

Missed this in yesterday's WashPost, but a piece on how Fannie Mae was desperately seeking subprime mortgages, just as the market went boom, is a must read.

In January 2007, as years of loose mortgage lending were about to send the nation's housing market into devastating decline, Fannie Mae chief executive Daniel H. Mudd wrote a confidential memo to his board.

Discussing the company's successes, Mudd said one of Fannie Mae's achievements in 2006 was expanding its involvement in the market for subprime and other nontraditional mortgages. He called it a step "toward optimizing our business."

A month later, Fannie Mae outlined plans to further expand its activities in the subprime market. The company recognized the already weak performance of subprime loans but predicted that they would get better in 2007, according to another Fannie Mae document.

Internal documents show that even late in the housing bubble, Fannie Mae was drawn to risky loans by a variety of temptations, including the desire to increase its market share and fulfill government quotas for the support of low-income borrowers.

And here all this time I thought the Feds made them do it. Gosh.

I had a surreal moment on CNBC this morning. In talking about proposed changes to the SEC's new-ish short-selling rules, I cited a recent IMD study showing that market quality deteriorated during the last month or so for the 19 stocks covered by the initial SEC ruling.

And what was the response to this from a fellow CNBC guest who was frothing in favor of more rules, especially ones banning the already banned practice of naked short-selling? "I think it's interesting that a columnist for TheStreet.com is citing a European study."

Hello? I guess he lives by that old rule, When you cite a European study, the naked short-sellers win. Who knew market facts were so geographically contingent?

Speaking of those damn foreigners -- you know, the ones who keep propping up Fannie/Freddie by buying their debt -- there is a new Federal Reserve study out seemingly showing that non-U.S. exposure to subprime is a little less than some of us expected.

In a hypothetical scenario with a 20 percent default rate on nonconforming mortgages and a 10 percent default rate on other types of underlying loans (with a 50 percent recovery rate for each), we predict that foreigners would ultimately lose $75 billion on their holdings of ABS backed by U.S. assets.

Oh, that doesn't sound too bad. But wait:

Then again, the mark-to-market losses stemming from a price markdown in all foreign-held ABS can be as much as six times larger using a 20 percent price markdown.

So, it's either $75-billion in foreign losses or a half-trillion -- which is reassuring, or not.

Source:

August 2008
Daniel O. Beltran
Laurie Pounder
Charles Thomas

Some quick links to a few articles that people have sent my way today:

• Laws of population growth (arXiv)
• CalPERS is going heavy into infrastructure and water (EMII)
• The genetic map of Europe (StrangeMaps)
• Why it has become more difficult to predict Nobel Prize winners (arXiv)
• Treasury is backing away from soothing Fannie/Freddie talk (FT)

Inc. magazine has out its 2008 list of the fastest growing private companies in the U.S. How many of the companies on the list last year were in financial services? How many this year?

You ready for the answer? The answer is there was 75 such companies on the Inc. fastest-growing list last year, but 99 this year.

Howzat, you wonder aloud? How can it be that there are more fast-growing private financial services companies in the U.S. this year than last year? An excellent question, and one to which I do not have the answer.

Some remarkable rainfall piling up in Florida during the current stalled Tropical Storm Fay. Check the following 24-hour precipitation figure, and note that there are areas in Florida now well on their way to 20 inches (!!) of rain over the last day or so.