Another great analysis video, thanks Evan really enjoying this series. As a follow up i would love to see similar results analysis for same strategy on the main US indexes and ACWX.
Best video about algo trading I've ever seen. As I understand, comparing this strategy (with 15% s\l) with simple buy and hold, gives about the same total return, correct ? However, what risk amount was taken ? 1% or more? If we increase it, will the total return increase proportionally ?
Thank you very much, glad you enjoyed it! The answer is [approximately] yes to your first question on matching buy and hold return, and to the second question, that would assume you were risking 1% of your account on every single trade. You certainly could increase it to 1.5% or 2% and that would increase your total return, however, that means each position is going to require more capital investment and you're going to likely run out of cash and not be able to take all of the signals produced. This goes down a deeper hole of portfolio construction, position sizing, etc. which are all beyond the scope of this video.
Great video! Thank you! I probably missed this but how do the three strategies have different number of trades? I thought the only thing different was the trailing stop....
That is correct. Because of the varying stop loss distances, strategies with larger stops will remain "in trade" longer and have less overall turnover. A smaller stop loss distance will repeatedly get triggered and then re-entered once the stock break out.
Great work here. After holding a trade for a year, im prob fully invested in the company from then on lol data seems to show that the longer you hold, the better the return which is pretty consistent with long-term investing. I'll prob be trying this myself.
Thanks! There's certainly a lot of data and research that suggest the less active traders are the more profits they tend to keep, so I think you're in a good spot.
@@TradeRisk You know of any other good software that I could build strategies and indicators in? I was going to use Tradingview's pine editor but I wanted to keep my options open.
I really enjoyed this! It's something I've been wanting to know. I'd be interested to see if you can rerun this strategy but this time assigning say 1-2% of portfolio allocated to each position and then plotting an equity curve for it. Thanks for the awesome job!
This is the first video i saw on this channel & it immediately made me subscribe 😊 I'd love to also see a backtest including shorts at 52 week low. The turtles did go long & short on similar breakouts
I have a question. Maybe I missed something. Why does the number of trades differ for the different trailing stop percentages if the entry/buy signal is the same? The number of trades decreased with wider trailing stops but shouldn't it be the same if the script is analyzing the same stock and time period, buying at the same time since the 52-wk high didn't change?
Great question! Maybe a little difficult to explain in text, but I'll do my best. You're right, all buy prices will be the same (new 52week highs) so imagine a hypothetical scenario where all of the systems: 5%, 10%, and 15% trailing stops, buy as a stock makes new 52week highs. But then the very next day, there's a 5% pullback in that stock, and so the 5% system has to exit. The following day, the stock rallies back up to new highs, and so the 5% system has to then rebuy that same stock because its back at new highs after being stopped out the day before. 2 trades, while both other systems, only took 1.
I'd love to see what happens if you buy at 52 week lows. This suggests what I think is a great strategy for position traders looking to do a "set it and forget it" approach over multiple months (And if the stock pays dividends, even better!) But what happens if a trader was like: I think stocks that are trading at or near their 52 week lows are due for a run up? Is there any merit to that approach?
If I had to guess, I wouldn't think buying 52-week lows would be a great strategy, but it would certainly depend on factors like how long you plan to hold the stock and general exit criteria. But that's just me guessing, I could be completely wrong. Maybe we'll experiment in a future episode!
Very detailed video, i had one question regarding Cumulative R. Does this mean 280% returns in 20 years (15% SL) or 280 times the initial capital. THANKS.
Cumulative R would be a function of your risk per trade. So yes in the 15% SL example, if you risked 1% of your account on every trade signal, total cumulative R would be a 280% return.
Thanks for stopping by the channel. Afraid we won't be able to create a new summary at this time, but the 5 minute mark to the 12 minute mark is all of the coding portion of this experiment.
@@TradeRiskI realize most of my winning trade is this strategy bc u entering at the number 1 most volatile level where it guaranteed explosive volatility to either direction. I think it great to run a straddle option trading strategy. Can you back test this strategy? Thanks
@@jasonpermanna308 unfortunately option strategies become a little too difficult to backtest and the data is very expensive, won't be able to help out on that one
Fascinating study with a surprising result (at least to me -- I'm thinking of the "down" years). It would be a nice touch IMO if at the end of a scenario the result was compared w/ a similar buy-and-hold for the relevant ETF such as DIA ... perhaps a comparative, simple tracking chart comparing the trading strategy with the ETF in question.
Thanks, Evan. Really enjoyed this! I'm a dev myself. How easy would it be to download the code and run the same scenario with a different dataset, timeframe, stop-loss %, etc.?
Thanks William! As far as the code, strategy, and environment, it should all be pretty plug and play and painless to get up and running on a local NinjaTrader 8 environment. I've got some instructions on how to do that on the GitHub repo. The biggest hurdle would be getting stock market data imported to NT, but there are a handful of sources (tiingo, yahoo, etc) that can at least get you started.
Would love it if you could do some of this stuff on the UK FTSE100. There is a serious void of public technicians in the UK and I don't have the time or firepower to be able to do some of this type of analysis myself!
Solid data there, appreciate it!
Thanks!
Another great analysis video, thanks Evan really enjoying this series. As a follow up i would love to see similar results analysis for same strategy on the main US indexes and ACWX.
Thanks Stuart, glad to hear! Noted, on the additional performance tables.
Best video about algo trading I've ever seen.
As I understand, comparing this strategy (with 15% s\l) with simple buy and hold, gives about the same total return, correct ?
However, what risk amount was taken ? 1% or more? If we increase it, will the total return increase proportionally ?
Thank you very much, glad you enjoyed it! The answer is [approximately] yes to your first question on matching buy and hold return, and to the second question, that would assume you were risking 1% of your account on every single trade. You certainly could increase it to 1.5% or 2% and that would increase your total return, however, that means each position is going to require more capital investment and you're going to likely run out of cash and not be able to take all of the signals produced. This goes down a deeper hole of portfolio construction, position sizing, etc. which are all beyond the scope of this video.
Great video! Thank you! I probably missed this but how do the three strategies have different number of trades? I thought the only thing different was the trailing stop....
That is correct. Because of the varying stop loss distances, strategies with larger stops will remain "in trade" longer and have less overall turnover. A smaller stop loss distance will repeatedly get triggered and then re-entered once the stock break out.
Great work here. After holding a trade for a year, im prob fully invested in the company from then on lol data seems to show that the longer you hold, the better the return which is pretty consistent with long-term investing. I'll prob be trying this myself.
Thanks! There's certainly a lot of data and research that suggest the less active traders are the more profits they tend to keep, so I think you're in a good spot.
@@TradeRisk You know of any other good software that I could build strategies and indicators in? I was going to use Tradingview's pine editor but I wanted to keep my options open.
Yes, I would look into NinjaTrader, Amibroker, and TuringTrader.
I really enjoyed this! It's something I've been wanting to know. I'd be interested to see if you can rerun this strategy but this time assigning say 1-2% of portfolio allocated to each position and then plotting an equity curve for it. Thanks for the awesome job!
This is the first video i saw on this channel & it immediately made me subscribe 😊
I'd love to also see a backtest including shorts at 52 week low.
The turtles did go long & short on similar breakouts
Thank you! Glad you enjoyed. Noted on the 52-week lows backtest
I have a question. Maybe I missed something. Why does the number of trades differ for the different trailing stop percentages if the entry/buy signal is the same? The number of trades decreased with wider trailing stops but shouldn't it be the same if the script is analyzing the same stock and time period, buying at the same time since the 52-wk high didn't change?
Great question! Maybe a little difficult to explain in text, but I'll do my best. You're right, all buy prices will be the same (new 52week highs) so imagine a hypothetical scenario where all of the systems: 5%, 10%, and 15% trailing stops, buy as a stock makes new 52week highs.
But then the very next day, there's a 5% pullback in that stock, and so the 5% system has to exit. The following day, the stock rallies back up to new highs, and so the 5% system has to then rebuy that same stock because its back at new highs after being stopped out the day before.
2 trades, while both other systems, only took 1.
How can you choose specific stocks to run this on? Like specific criteria rather than the stocks in an index?
Yeah it's real easy in NinjaTrader you can select individual stocks to run the strategy on or you can create your own custom watchlist of stocks.
I'd love to see what happens if you buy at 52 week lows. This suggests what I think is a great strategy for position traders looking to do a "set it and forget it" approach over multiple months (And if the stock pays dividends, even better!)
But what happens if a trader was like: I think stocks that are trading at or near their 52 week lows are due for a run up? Is there any merit to that approach?
If I had to guess, I wouldn't think buying 52-week lows would be a great strategy, but it would certainly depend on factors like how long you plan to hold the stock and general exit criteria. But that's just me guessing, I could be completely wrong. Maybe we'll experiment in a future episode!
@@TradeRisk Would love to see it!
Very detailed video, i had one question regarding Cumulative R. Does this mean 280% returns in 20 years (15% SL) or 280 times the initial capital. THANKS.
Cumulative R would be a function of your risk per trade. So yes in the 15% SL example, if you risked 1% of your account on every trade signal, total cumulative R would be a 280% return.
@@TradeRisk could you give an example explaining this? Thanks.
Can you make a 5-minute summary of how you do this strategy
Thanks for stopping by the channel. Afraid we won't be able to create a new summary at this time, but the 5 minute mark to the 12 minute mark is all of the coding portion of this experiment.
I was thinking of breaking 52 weeks high as a new discovery trading strategy but I didn't know it already existed lol
Haha, yeah it can be hard to find truly unique strategy ideas nowadays, but sometimes the oldies are still the good ones
@@TradeRiskI realize most of my winning trade is this strategy bc u entering at the number 1 most volatile level where it guaranteed explosive volatility to either direction. I think it great to run a straddle option trading strategy. Can you back test this strategy? Thanks
@@jasonpermanna308 unfortunately option strategies become a little too difficult to backtest and the data is very expensive, won't be able to help out on that one
Fascinating study with a surprising result (at least to me -- I'm thinking of the "down" years). It would be a nice touch IMO if at the end of a scenario the result was compared w/ a similar buy-and-hold for the relevant ETF such as DIA ... perhaps a comparative, simple tracking chart comparing the trading strategy with the ETF in question.
Thanks and yes exactly that would be next the step in this process building out a portfolio, position sizing, etc.
Great Video . Very informative
Thanks! Glad you liked it
I'd be interested to see how this looks with a trailling ATR based stop loss (1x, 2x & 3x)
Yes that certainly impacts returns. I don't have the data in front of me on that but maybe in a future video.
I'll be interesting to see how this strategy performs if we trade 52 week lows as well.
Maybe in a future video we can explore
Thanks, Evan. Really enjoyed this! I'm a dev myself. How easy would it be to download the code and run the same scenario with a different dataset, timeframe, stop-loss %, etc.?
Thanks William! As far as the code, strategy, and environment, it should all be pretty plug and play and painless to get up and running on a local NinjaTrader 8 environment. I've got some instructions on how to do that on the GitHub repo. The biggest hurdle would be getting stock market data imported to NT, but there are a handful of sources (tiingo, yahoo, etc) that can at least get you started.
Very interesting
Cool!
Is this Python ?
C#
Would love it if you could do some of this stuff on the UK FTSE100. There is a serious void of public technicians in the UK and I don't have the time or firepower to be able to do some of this type of analysis myself!
Completely understand Michael and the problem is the data. I don't subscribe or have any UK stock data except the top level ETFs.
Uk is a dud. Brexit has killed inward investment and UK equities reflect that.
nyc