StockFetcher Forums · Filter Exchange · LONGER TERM ANALYSIS OF EMA CROSSOVERS ON THE SPY<< 1 2 3 >>Post Follow-up
3,833 posts
msg #88570
Ignore Kevin_in_GA
2/23/2010 10:24:25 AM

I have been looking at diferent types of long/short trading signals for a while now, trying to set up a clean and unambiguous system for trading SDS/SSO based on signals generated from either the SPY or ^SPX.

Since there are a lot of ways to do this, I started with simple systems that could be tweaked (moving averages, MA crossovers, standard oscillators like the MACD, etc).

One of the easiest systems to implement is the classic moving average crossover. These are typically seen by many traders as "too slow" relative to oscillators, but still do a good job of keeping you on the right side of the trade, and avoiding whipsaws and overtrading.

Looking at the last 250-300 days provides a very skewed perspective, as there was almost no downside during this time. ANY indicator system was most likely profitable. But now extend that back to 2005, 2000 or 1995 and see if you beat the market with your timing. Probably not.

What I wanted to do was to look at longer timeframes than what we can typically do here at SF or other web sites. I wanted to be sure that any system I was considering had "stood the test of time" and been analyzed against a wide range of market conditions.

I went and downloaded the SPY historical data back to 1995 and looked at 10 day ema crossing a series of longer emas (30,40,50,60, and 70). I felt that this range of EMAs was neither too slow nor too fast. The EMA(10)/EMA(50) cross emerged as the most profitable when played both long and short.

Was it correct all the time - NO, but it was more profitable when it was right than it was costly when it was wrong.

Did it catch the market tops and bottoms - NO, but it caught and rode the dominant trends.

Did I avoid whipsaws - NO, but neither does any other system

Did I beat a simple buy-and-hold over the same timeframe? YES.

Here is the full comparison over a series of timeframes going back 15 years:


SINCE 1/3/1995 - $10,000 GREW INTO $45,953 (CAGR OF 10.7%) COMPARED TO $29,792 FOR B&H (CAGR OF 7.6%)

SINCE 1/3/2000 - $10,000 GREW INTO $22,392 (CAGR OF 8.4%) COMPARED TO $8,699 FOR B&H (CAGR OF -1.4%)

SINCE 1/3/2005 - $10,000 GREW INTO $20,239 (CAGR OF 15.1%) COMPARED TO $9,803 FOR B&H (CAGR OF -0.4%)

SINCE 1/3/2007 - $10,000 GREW INTO $18,368 (CAGR OF 22.5%) COMPARED TO $8,093 FOR B&H (CAGR OF -6.8%)

SINCE 1/2/2008 - $10,000 GREW INTO $20,779 (CAGR OF 44.2%) COMPARED TO $7,752 FOR B&H (CAGR OF -11.9%)

SINCE 1/3/2009 - $10,000 GREW INTO $15,247 (CAGR OF 52.5%) COMPARED TO $11,801 FOR B&H (CAGR OF 18.0%)

Now, between 1995 and 2000 it was nowhere near as good as buy and hold, but that period was an astounding run for the markets that would be essentially impossible to beat with any timing system. Truth is that the system really only kicked into high gear since 2005 which is why the recent data looks so strong.

And remember, those CAGRs are based on the SPY only, not SSO/SDS ... and you have to play both long and short for this to work as well as it did.

Here is a quick filter that shows you the trades over the past two years:

set{E3, ema(10) - ema(50)}

set{Longposition, count(E3 > 0,1)}
set{Shortposition, count(E3 < 0,1)}

draw shortposition on plot trigger
draw longposition on plot trigger

add column ema(10)
add column ema(50)
add column longposition
add column shortposition

draw ema(10)
draw ema(50)


chart-time 2 years

I wanted to post this to remind people that simple indicator systems often work better than complex ones, and that moving averages (essentially a direct reflection of recent price action) stil are among the most reliable systems one can use.

Hopefully this stimulates some discussion.


179 posts
msg #88602
Ignore fortyfour
2/23/2010 6:47:02 PM


I am way too busy trading news and gut feel.

This 7 or so day "rally" has ranckled me to no end because it shouldnt have happenned you know....
I'd probably do better thowing ma(7) on the chart and trading that.

Also, oil is the most manipulated market in the world... classic commodity fundamentals can never
be applied end up looking foolish...

But, with that being said.....this recent rally in oil should never have way... oil shold be at $70.
not $80 a barrell.... Id probably do better throwing ma(10) on the chart and trading that...

That March 09 rally...Forgettaboutit !

And finally,,, this market should go down some should you know.....

Who knows....2 or 3 more years and I may start looking at these things you are talking about. here....Maybe...

But......serioulsly, watching something and .......WAITING for price to cross a line to both BUY and SELL
is very hard to do consistently and that is the crux of the problem........not the period of the line IMHO.

Good Luck...

3,833 posts
msg #88604
Ignore Kevin_in_GA
2/23/2010 7:20:47 PM

Admittedly, this type of system, while effective in the long term, requires patience.

It does, however, also free one's time up to work or play with other things, rather than be glued to the internet watching minor moves (today being an example - only about 1% on the major markets) that will probably self-correct in a month.

Honestly, are you enjoying the hunt or the profit? If it's the hunt that is satisfying, then this type of approach is definitely not for you.

745 posts
msg #88608
Ignore maxreturn
2/23/2010 8:14:40 PM

How about something a lot simpler. Notice how in March 2003 the SPX made a double bottom on high volume. Then in November, 2007 the SPX made a double top on high volume at the same price levels as the 2000 high. Then in March, 2009 the market again made a double bottom on very high volume at the bull market lows of 2003. KISS

179 posts
msg #88616
Ignore fortyfour
2/23/2010 11:32:42 PM

Much of what I said was tongue in cheek to highlight the inconsistency and sometimes futility of
tying to predict market direction..

IMHO much better to wait for entries rather than jumping the gun or anticipating what does not come to being.

Trend following, or what I know of it, is very hard to follow.

As we speak, I am looking at a MA x-over test results in Covells trend book.

Generally , the fast ma is best when 20 to 90 days with a slow ma that is 40 to 80 days longer than the fast.

The best results came from the areas of 50ma/110ma and the 80ma/120ma. Each of these areas were above
20% anual profit and were surrounded by a large area of good performance. Ie: changing either fast or slow MA by + OR -
20 days did not hurt results badly so it is robust.

His test of 10ma/50ma x-over was specifically in the 6-8 % anual return area.
The nearest , best improvement to the 10ma/50ma (12-14% anually )was by moving to the 20ma/80ma area.

Test markets were from 1/1/90 to 12/31/2004 .
Markets were Currencies, Energies, Metals, Grains, Soft commodities, Financials, Meats......... 20 markets.

If you are interested.......
It is a quick easy read......mostly in the beleif system of trend following.... too bad it is not a bit more.

Author : Michael Covell
Title : Trend Following "How great traders make millions in up or down markets"
Publisher: Financial Times Press.

I dont think todays move was minor.... but, there I go again....

3,833 posts
msg #88629
Ignore Kevin_in_GA
2/24/2010 6:51:45 AM

I dont think todays move was minor.... but, there I go again....

In the context of a 65+% percent gain in the last 12 months or so, a 1% down day is noise. Do you feel the same level of emotional response when the market goes up or down 1% over a week? If not, then it is not the magnitude of the change, but the rate that is troubling you. If it corrects back up 1% today after Bernanke's speech, should you have traded off of it?

I'm just trying to point out that longer term investment/trading should not be focused on intraday or daily movements in the markets, and that there are valid, time-tested and profitable strategies that do not require constant vigilence of the markets.

683 posts
msg #88635
Ignore duke56468
2/24/2010 10:13:20 AM

Kevin thanks for posting your findings. Is there a way statistically to check the effectiveness of the MONTHLY price close relative to the MA(12)? I have done this manually with SP-500, back into the 1980's and it doesn't seem to produce too many whip-saws. I know this is too long a time frame for traders, but may be helpful for longer term portfolios.

3,833 posts
msg #88657
Ignore Kevin_in_GA
2/24/2010 1:42:28 PM

I had to do this analysis by hand as well - I am not aware of any program that will backtest as far back as 1995 or earlier.

I saw your monthly MA(12) post from a while back (at least I think it was yours) - that actually got me thinking along these lines, as a way to "semi-actively" manage investments to avoid larger cyclical downturns (like 2000-2002 and 2007-2008).

These signals, when they come, are quite important and should not be ignored.

Along these lines, another valid approach is the 5 week/20 week EMA crossover, although when I looked at these types of systems (and I did), they were not as profitable as the 10/50 day EMA cross.

179 posts
msg #88661
Ignore fortyfour
2/24/2010 2:06:16 PM

In Covells book they state that testing was done on Fidelity Wealth-Lab Pro .

"All trend following systems in this appendix have been developed and tested with
Wealth-Lab Developer"

In 2006 ish they were able to look back to 1994

745 posts
msg #88677
Ignore maxreturn
2/24/2010 7:27:41 PM

Frankly I'm a little surprised at the lack of response to my post. I guess I shouldn't be. Many people tend to dismiss the simplicity of my approach as too simple. Yet my approach will beat the pants off of any MA crossover system or price crossing above/below an MA EVERY TIME. If you had been paying attention to a simple monthly bar chart with volume and major support and resistance levels you would have seen obvious distribution in volume and price at the 2000 and 2007 highs which coincidentally were at the nearly the same price level. Likewise, in March 2009 the decline was stopped on huge institutional buying at nearly the same price levels of the 2003 bull market. Go ahead and ignore me. I'll continue to profit.

StockFetcher Forums · Filter Exchange · LONGER TERM ANALYSIS OF EMA CROSSOVERS ON THE SPY<< 1 2 3 >>Post Follow-up

*** Disclaimer *** does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. does not recommend particular securities., Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.

Copyright 2016 - Vestyl Software L.L.C.Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus