Talking Trading: New ETFs

There's been a lot of buzz on the street recently about a series of new ETFs from ProFunds Group. These new Proshares allow the investor to take a leveraged bet on the major indecies on either the long or short side. Very interesting stuff. I can tell already that the portfolio theorist in me is going to have an absolute ball with these puppies. Aside the obvious ramifications of being able to be short the market as well as to generate leverage in an IRA for the first time in history, there are several hypotheses I'm hoping to play around with. While there is still nowhere near enough data to work with, I am planning on quite a bit of number-crunching. Some hypotheses include:

• Taking advantage of outlier days - if the UltraQQQ, for example, really produces double the return of the regular QQQ on a consistent basis, can one buy the UltraQQQ and short the regular QQQ in order to take advantage of higher volatility? Should one hedge with 2 QQQ contracts for every one UltraQQQ? Is there a correlation with the VIX?
• If, as many retail brokerages like to claim, the market historically returns 9-10% annually, shouldn't we all just throw 100% of our portfolio into the UltraQQQ and, in the long run, earn a rate of return that is significantly higher than the rate of inflation, allowing us to retire quite comfortably on our own private islands somewhere?
• Are there arbitrage opportunities between the Ultra long and Ultra short ETFs?
• If the market has a huge runup, similar to the Nasdaq circa 1999-2000, does the UltraShortQQQ go to zero?

A lot of these questions would be easy enough to answer via Monte Carlo Engine if we knew the exact methodology that ProFunds uses to price these ETFs. For now, all I've seen is that the pricing is determined through the use of various derivative products that may include futures, swaps, and other instruments, and that there is no guarantee that the actual value of the ETF will really be equal to what it should be. Acting on the assumption that the Ultra ETFs will behave exactly as they should can give me a decent idea of what to expect, but I'd rather be able to back-test with actual data. (I am well aware that backtesting can often turn into "backfitting" the data, but I am also a very firm believer that markets are neither perfectly rational nor perfectly efficient, and these mysterious derivatives that ProFunds is using to reach the expected levels for its Ultra funds sounds exactly like the sort of thing that will make spreads get all sorts of funky because actual values and theoretical values don't quite add up.)

Anyway, these new ProShares look pretty interesting, and not just from an academic standpoint. If business is as slow tomorrow as it has been all month, I should have plenty of time to mess around with these things.