Why do moving averages SUCK so much?

S

Saltynuts

So I ran some simple moving averages on Excel. Just simple moving averages. Like on SPY. But I tested everything from like 2 day moving averages (lulz) through 300 day moving averages.

All the numbers sucked versus just buy and hold. Way worse upside, and the max drawdown was barely reduced. It didn't matter the period.

HOWEVER, then I found a flaw in my equations. I had a < when it should have been a >. So... AHA! I had been testing like the REVERSE of moving averages, by when spot dropped below it, sell when it went above it. THAT'S why the numbers sucked! I was going to fix it and the numbers would truly be great!

Only, I fixed it, and the numbers still sucked. Which perplexes me much. If markets trend, how can moving averages (and their reverse) suck so bad?

Here is the best I can come up with: Stock prices, on average, move up over time. Thus, despite whatever the trend is, the stock is likely to go up the next day. For every day you are out of the market that is, on average, upside you are losing out on.

Is that the answer? Or why are they so bad?

Thanks!!!
 
Saltynuts said:
So I ran some simple moving averages on Excel. Just simple moving averages. Like on SPY. But I tested everything from like 2 day moving averages (lulz) through 300 day moving averages.

All the numbers sucked versus just buy and hold. Way worse upside, and the max drawdown was barely reduced. It didn't matter the period.

HOWEVER, then I found a flaw in my equations. I had a < when it should have been a >. So... AHA! I had been testing like the REVERSE of moving averages, by when spot dropped below it, sell when it went above it. THAT'S why the numbers sucked! I was going to fix it and the numbers would truly be great!

Only, I fixed it, and the numbers still sucked. Which perplexes me much. If markets trend, how can moving averages (and their reverse) suck so bad?

Here is the best I can come up with: Stock prices, on average, move up over time. Thus, despite whatever the trend is, the stock is likely to go up the next day. For every day you are out of the market that is, on average, upside you are losing out on.

Is that the answer? Or why are they so bad?

Thanks!!!
More...
There's a guy who has a MA system on a MONTHLY Index Chart going back to 1920 something.
BH in his Simple Method was not as profitable as In / Out. Def not as profitable as Long / Short, by $Millions on a $10K inital outlay.
You want the info on that, say the word, you got it.
 
Last edited:
Most of us learn to trade using tons of indicators
because there are tons of information about how great the indicators are.

Later we discover that we need to learn to trade using no indicator at all.
Because the market moves in various fashions / manners / speed /
pattern, acceleration/decceleration rate etc

There is no such thing as a ONE-SIZE-FITS-ALL indicator.

I just have candlesticks (the simple type and not the Heikin Ashi, Renko ...)
and perhaps zig zag line.

There are traders who have been using indicators with great success.
Anyway, have a good day
 
Last edited:
  • Like
Reactions: Gaw
This has been debated ad nauseum, however a few points:

1. Lots of algos scan price series in real time to detect exploitable patterns, ensuring that something so trivial as a price > MA rule probably doesn't have much of an edge alone. It would be found and arbitraged if had been that simple - longer time frames are less crowded though.

2. Moving averages are laggy by definition. They show a bit of the past. Will that past say anything about the future? Why should it? [I'm not saying this is always the case, just ask for your specific situation.]

3. It helps to think about why "technical analysis" in general sucks so much. That said, moving averages are among the few usable things in the TA toolbox... actually a form of lowpass filters that filter out higher frequencies. Treat them as such rather than magic predictors of the future.
 
Moving averages are lagging indicators. Indicators for the most part are useless. What data do you feed indicators to get your readings? Price. That is the only thing that matters. Stockcharts are the purest form of data out there, without any indicator. Think about it. It is a summation of all the trades put on by market participants. You know where most investors or traders are committing their monies and where they are bailing out too. Despite all the lies and disinformation peddled by CNBC, Bloomberg, ET trolls, Reddit and other social media hacks, they cannot lie about where a stock is headed, or the trend. Remember ET trolls for months on back in 2020 were telling us to prepare for a market crash, some of them still saying the same thing in 2021? Yet, the stockcharts show a very bullish trend so, who do you believe? The know nothing ET trolls or the stockcharts?
 
Last edited:
MNQ 5min bars, Stochastic fast, 53,53,50,50 wilders moving avg
MNQ 5min bars, Moving avg two lines 11,50 wilders mov avg cross over
Rabbit hole trying to match price waves which are caused by fundamental decisions made at certain times of the day and night.
 
notagain said:
MNQ 5min bars, Stochastic fast, 53,53,50,50 wilders moving avg
MNQ 5min bars, Moving avg two lines 11,50 wilders mov avg cross over
Rabbit hole trying to match price waves which are caused by fundamental decisions made at certain times of the day and night.
More...
We been missing your screenshots, holmes.
 
Um, they work just fine. But they are not literal and universally all the same. Try looking at them with an dynamic offset based on the market action..
 
You must log in or register to reply here.
Back
Top