Not to mention that only in textbooks can one find perfect risk-free arbitrage opportunities. Unfortunately, the real-world offers only quasi-arbitrage opportunities. Fortunately, contrary to what some academics believe, such opportunities do exist.
Hmm, this is technically true, but life does provide some actual very close to ideal arbitrages.
For example, back in the day, I used to buy and sell a popular collectible card game on eBay. Right when eBay came out, there was a massive arbitrage on buying cards from people and selling them online, because there were huge geographic price discrepancies. There was no risk in buying a card for .50 and selling it for $8, etc.
Eventually, the market corrected itself and prices stabilized, but for a while, there was a lot of money to be made if you just knew the right cards to buy.
Another example of this sort of temporal arbitrage is sports betting right before the country was networked well. It used to be if you had a guy on the coasts and in various major metropoli, you could place bets on games with varying handicaps, and always win.
The famous poker player Howard Lederer made a lot of money this way.
I entirely agree with you. In fact, there are still people doing arbitrage on eBay, although eBay is now an established site visited by millions. An example: some people look for misspelled items being auctioned (an auction where the item being auctioned is not spelled right, attracts a lot fewer bidders), buy them low, and sell them high at another auction. Looks like a truly classic arb! Of course, there's still the risk that the item one has just bought will attract no bidders...
In exchange-traded financial instruments, arbitrage is now incredibly more difficult than what it was 100 years ago. Computers and communications killed most of the arbs. But it's still possible to find arbs in structured products :-)
And, although sports betting sites are shady, I would bet there are quasi-arbitrage opportunities to be found in online sports-bettings these days...
