4 ms·

Am I reading this right? It seems to be advocating “Buy high, sell low”.

otp124 9y ago
This is exactly how I read it. I think the OP flipped the two.. This, and the fact there is zero evidence to back up the claim is enough for us to confidently ignore it and move on.
- frikk 9y ago
  Well, think about it in the context of a rising volatile asset. I think what OP is recommending is to identify "entry points" and "exit points". - Buy as the asset is rising above 10 month MA (and likely will continue to rise, maybe for weeks, months, years) - Sell as the asset is cooling off (falling below 10 month MA). Don't get "back in" until the asset is heating back up. With something like bitcoin, this idea would be to ride up a large market bull, but then get out until it looks like it is happening again. If you play with the Moving Average[0], you can see that this strategy would basically have you enter in late 2013, sell in mid 2014, and not enter again until late 2015, then hold all the way. Not exactly a riveting strategy, but OP didn't claim as much. [0] https://bitcoincharts.com/charts/bitstampUSD#rg1460zigDailyztgSza1gEMAzm1g300zm2g25zv https://bitcoincharts.com/charts/bitstampUSD#rg1460zigDailyz...
- gnaritas 9y ago
  No, the OP said it correctly, it's a momentum strategy, you presume the market will keep going in the direction it's already going. This is a simple trend trading strategy. It doesn't actually work anymore, but anything that ever worked once will continue to live and be promoted by people who don't grasp that markets aren't static. What the strategy ignores is that it gets eaten alive in ranges, when the market isn't trending, this strategy buys high and sells low over and over until you're broke. Works great when the market is trending though. Bitcoin for example last crossed its 10 month moving average around $246, it's now trading over 4k; $246 would have been a hell of a price to get in at. So yes, buying the "high" works well in a trending market. It's have to drop to $1827 to trigger that same sell signal so you'd lose well over half the floating profit, and then it'd probably cross back and forth a few times when it's near the average trigger many failed entries and exits eating up all the profit you just made. Moving average strategies don't work except in hindsight when you can see the trend and decide which MA would have worked (something you can't know beforehand), but they're great teaching aids in understanding and learning about trading strategies because they're nearly as simple as a strategy can get.
AndrewBissell 9y ago
Since most markets trend for long periods of time, this winds up being a better strategy than you might imagine. "Buy high, sell low" also isn't quite an accurate description, because moving average crosses in strongly trending markets happen at the beginning of the move, rather than the end. No one regrets buying AAPL high in 2010, or selling LEH low in early 2008.
- Dolores12 9y ago
  >No one regrets buying AAPL high in 2010, or selling LEH low in early 2008. This is clearly survivorship bias.
- matt4077 9y ago
  I just tried that momentum theory on the bitcoin closing data someone posted in a sibling comment. I get 715 days of a trend continuing (down->down or up->up), and 739 days of the "trend" reversing: down->up OR up->down. That pretty much points at a random walk. "Momentum", and any other patterns people see, are imagined.

This is exactly how I read it. I think the OP flipped the two.. This, and the fact there is zero evidence to back up the claim is enough for us to confidently ignore it and move on.

frikk 9y ago
Well, think about it in the context of a rising volatile asset. I think what OP is recommending is to identify "entry points" and "exit points". - Buy as the asset is rising above 10 month MA (and likely will continue to rise, maybe for weeks, months, years) - Sell as the asset is cooling off (falling below 10 month MA). Don't get "back in" until the asset is heating back up. With something like bitcoin, this idea would be to ride up a large market bull, but then get out until it looks like it is happening again. If you play with the Moving Average[0], you can see that this strategy would basically have you enter in late 2013, sell in mid 2014, and not enter again until late 2015, then hold all the way. Not exactly a riveting strategy, but OP didn't claim as much. [0] https://bitcoincharts.com/charts/bitstampUSD#rg1460zigDailyztgSza1gEMAzm1g300zm2g25zv https://bitcoincharts.com/charts/bitstampUSD#rg1460zigDailyz...
gnaritas 9y ago
No, the OP said it correctly, it's a momentum strategy, you presume the market will keep going in the direction it's already going. This is a simple trend trading strategy. It doesn't actually work anymore, but anything that ever worked once will continue to live and be promoted by people who don't grasp that markets aren't static. What the strategy ignores is that it gets eaten alive in ranges, when the market isn't trending, this strategy buys high and sells low over and over until you're broke. Works great when the market is trending though. Bitcoin for example last crossed its 10 month moving average around $246, it's now trading over 4k; $246 would have been a hell of a price to get in at. So yes, buying the "high" works well in a trending market. It's have to drop to $1827 to trigger that same sell signal so you'd lose well over half the floating profit, and then it'd probably cross back and forth a few times when it's near the average trigger many failed entries and exits eating up all the profit you just made. Moving average strategies don't work except in hindsight when you can see the trend and decide which MA would have worked (something you can't know beforehand), but they're great teaching aids in understanding and learning about trading strategies because they're nearly as simple as a strategy can get.

AndrewBissell 9y ago

Since most markets trend for long periods of time, this winds up being a better strategy than you might imagine. "Buy high, sell low" also isn't quite an accurate description, because moving average crosses in strongly trending markets happen at the beginning of the move, rather than the end. No one regrets buying AAPL high in 2010, or selling LEH low in early 2008.

Dolores12 9y ago
>No one regrets buying AAPL high in 2010, or selling LEH low in early 2008. This is clearly survivorship bias.
matt4077 9y ago
I just tried that momentum theory on the bitcoin closing data someone posted in a sibling comment. I get 715 days of a trend continuing (down->down or up->up), and 739 days of the "trend" reversing: down->up OR up->down. That pretty much points at a random walk. "Momentum", and any other patterns people see, are imagined.