Traders watching the S&P 500 have at least four possibilities for market proxies (and most other major indexes have similar issues), but they’re not always created equal. Some instruments do some things better than others and some are completely inappropriate for certain uses. Let’s start with looking at the pros and cons for each SPY proxy:

SPY: An ETF that basically allows a stock trader to trade the entire S&P 500 in one instrument. Important to understand how SPY is priced and pays dividend.  $0.10 in SPY = 1.00 Cash S&P point. SPY accrues dividends and pays them out quarterly, which means at any time SPY = (Cash / 10) + accrued dividends. This is extremely significant.
Pros: stock daytraders can easily punch it up on their boxes (doesn’t seem like a good one, huh?)
Cons: noisy with spikes at turning points.
dividends distort previous levels (this is HUGE.  read on…)
dividends distort spreads with other instruments that pay dividends on different schedules.

Futures: It is important to understand how futures are priced. Futures = Cash S&P + amortized risk free rate – amortized dividends. In practice, this means that futures will usually trade at a small premium to Cash, converging on cash at expiration. Futures expire and roll quarterly.
Pros: this is, in many ways, “the real market”. Most people actually trading the market are trading futures.
Favorable tax treatment for short term traders.
Cons: Eminis have fairly large ticks so there is some loss of resolution in small bars.
There can be some taxation issues if used in pairs against equities.
These do expire, which may be confusing to some traders who have not dealt with futures, but it’s really a very simple and cheap process to roll if you’re holding longer term positions.

Cash: This index is published every few seconds and is a calculated index based on the actual price of the S&P 500 stocks.
Pros: Cash is king. You want to know what the level actually is, free of any issues of dividends or premiums? Then use cash.
Cons: It’s a cash index so you can’t actually trade it. You could trade it with a program that trades the constituent stocks in the correct size for the index weighting, but that is beyond the means of most of our readers. (This is, in fact, how the SPY and Futures are so well arbed.)
This index is calculated so it lags the market by a few seconds. Difficult to use this as a timing tool for short-term trading.
Opening print on Opex day is subject to distortion since SPX options are marked from Friday’s open.

Inverses/Ultras/etc: Not too much to say about these that hasn’t already been written.
Pros: Good bang for the buck for short-term traders. (Don’t underestimate this.)
Cons: Can be thin at times.
The math makes them completely inappropriate for long-term holdings in almost all situations. (I’ll probably punch the next person who tells me they are “good enough”. Learn to multiply and divide and think…)
The same math distorts levels beyond the current trading day.

Whew. Have a headache yet? Let’s boil this down to “why do we care?” Imagine you were watching the market this morning (Monday, September 20, 2010) for a breakout of a significant technical level. If you were watching the cash S&P, notice what you saw at 10:00 AM:

(The chart on the left is a daily chart so you can see where the level comes from. Focus on the one-minute chart on the right.) Well, notice first of all one-minute cash S&P charts are not pretty, because of the way the data is published. However, the key fact is that the market broke resistance at 10:00, consolidated basically at the level and then traded higher. Now, look what you would have seen if you were watching SPY:

Ok, now SPY paid a $0.60 dividend on 9/17/10, so think for a moment what this does to the SPY price. On that day, the price (theoretically) drops $0.60 relative to the previous and all previous trading days. Furthermore, those days were distorted as well because dividends were building up, pushing the price higher. So this is the SPY dividend roller coaster:  Each quarter there is a dramatic one day drop, followed by a slow climb (this is relative to Cash S&P) to the next ex-date and accompanying drop. This means if you were using previous SPY days for reference, you would have seen the level potentially as much as 0.5%  higher, in this case maybe around 113.25. Do you think that might have had an impact on how aggressively you traded the bull flag at 10:00? You betcha.

Just to be complete, here is the same level you would have drawn on the December futures contract. (If you were using a continuous contract the level might have been slightly lower.) As you can see, also some distortion here.

In summary, cash is king. The “real” levels are most visible on the Cash index, though the way they are published makes them unsuitable for short-term purposes. You can certainly make a compelling argument for watching certainly levels on the futures, especially since an army of futures traders and futures trading algos often drive the market in the short-term. For short-term levels (same day), most of these instruments will be almost exactly the same with a little additional noise in some of the EFT products, so it does not matter what your very short term reference is. However, and this is important, what you cannot do is watch long-term levels on SPY unless you adjust the chart for dividends, especially around ex-dividend dates.

Follow me on twitter AdamG_SMB