Ok, so think about the situation when the calls would get early-Xed. Calls get Xed and your stock goes with them - that would mean that the calls are ITM (i.e. the stock went up after you put on a position). What do you think happens to the price of the put when the stock goes up?
Valid concern, but when the stock is going up, the calls are LESS likely to be exercised because the position is getting out of the money... You can set this up in the last hour (or in the last 10 minutes) before the close, how much stock movement is expected in 1 hour or 10 minutes? Check the movements of the above mentioned 4 stocks in the last 2-3 hours on Thursday. They were all flat or down, when the market was consolidating after the morning rally.
Are you sure about that? Cause I am saving this quote for posterity Anyway, I hope you can see where the risk come from now. This does not mean that it's a bad strategy for you, just that it's a riskless arbitrage. Now, at the risk of contradicting the OP, I have to note that there are some opportunities that are "shoe leather arbitrage", in financial markets as well as outside. There are few reasons why these persist, but the main one are extreme capacity constraints.
to the OP - you reject the riskless strategy of selling futures (in contango) and buying spot, then waiting for convergence at expiration. Where is the risk in such a trade?
No, he doesn't. He just says that arbitrage strategies like this are not probable for a retail player. Furthermore, even for a commercial operator, the strategy above has a bunch of risks. Let's say you are running an index arbitrage book - buying stocks and selling futures against it (most of the time that's the direction the strategy works, for various reasons). Yes, the book is neutral with respect to delta but there is financing risk and dividend risk. A lot of times these opportunities are particular to players with a specific edge (which is why they exist). For example, to do cash-and-carry in pork bellies you'd need a commercial meat storage facility. At other times, it's there because one leg of arbitrage can not be sourced. Sometimes is because there is a hidden risk that is embedded in the structure.
Even the "risk free" treasury rate ... how about the risk of losing to inflation! http://www.marketwatch.com/investing/index/irx?countrycode=xx
You're basically trading interest rate (in equities)... you've locked in a certain rate by long spot vs short future. The convergence is just the release of the interest over time. While it's virtually riskless in the sense you're hedged on market movement... the only reason you would do a strategy like this is if you think interest rate is too high. Or, if you think a dividend will be paid before expiry, which isn't included in the future pricing at this time. So there is financing and dividend exposure. Also, I think (like SLE says) for retail this involves a decent amount of financing and financing risk. If leveraged, the financing rate will be a lot higher than what you lock in with the (risk-free) future rate. Even if not leveraged... having to buy multiple stocks in certain quantities will mean a decent cost, I think you're better off just buying T-bills for the same or better return. Just to clarify... a risk-free strategy which pays a less return as inferior when something better is available. So, you could argue... hey, T-bills are risk-free!!! But that's not what I meant by this thread... (and, even T-bills are not really risk-free, since blablabla) If you're talking about commodity futures... than there's a whole lot more risk involved.