Great question. Here's how to solve the conundrum mathematically. You would look at the risk to try and set the initial position with and override based on percent of equity allocated to the strategy. So if you had a close stop and the position calculated out to say more than 5% of the equity, you would default to 5% of the equity position size. If it came out 4.2% of the equity, you would use that number. ETR!
Sure, I use Interactive Broker's stock 30 RSI on the universe of etrs and see what gets above 65 before drilling down to the buy/sell engine level and putting in orders.
Regarding you timing sector strategy, would you be so kind to tell me how you calculate the amount of money you invest with each signal? I see two alternatives:
1. You invest an identical and fixed amount of money for all the signals of all the ETFs. For example, if you have allocated $300,000 to trade this strategy and you get a signal to go long XBI, you invest $15,000 in XBI ($300,000/20). I see a problem with this method: you do not take into account how far the stop is so sometimes your stop may be very close to the entry price and sometimes your stop may be far away. I think you should take the same risk with all the signals because you never know which ones are going to be winning trades and which ones losing trades.
2. But if you take the same risk with each signal (for example, 0.5% of your trading capital which is $300,000, that is you are risking $1,500 per signal), then you will have to invest a lot money if the stop is very close and very little when your stop is far away. Then, it is imposible to control how much money you will need to trade this strategy (it could happen that you need much more than those $300,000 you have allocated to trade it if many of the stops are very close to the entry price).
Great question. Here's how to solve the conundrum mathematically. You would look at the risk to try and set the initial position with and override based on percent of equity allocated to the strategy. So if you had a close stop and the position calculated out to say more than 5% of the equity, you would default to 5% of the equity position size. If it came out 4.2% of the equity, you would use that number. ETR!
Thanks Tom. I am always reading and learning what you have to say. Talking from Brazil. Big fan. You are a legend. Thanks for sharing knowledge.
Hi Tom, Im buying pullbacks on the daily time frame for sure
Always a good read ☺️
Sure, I use Interactive Broker's stock 30 RSI on the universe of etrs and see what gets above 65 before drilling down to the buy/sell engine level and putting in orders.
Thanks for the kind words, Luiz!
Hi Tom, thanks a lot for your update.
Regarding you timing sector strategy, would you be so kind to tell me how you calculate the amount of money you invest with each signal? I see two alternatives:
1. You invest an identical and fixed amount of money for all the signals of all the ETFs. For example, if you have allocated $300,000 to trade this strategy and you get a signal to go long XBI, you invest $15,000 in XBI ($300,000/20). I see a problem with this method: you do not take into account how far the stop is so sometimes your stop may be very close to the entry price and sometimes your stop may be far away. I think you should take the same risk with all the signals because you never know which ones are going to be winning trades and which ones losing trades.
2. But if you take the same risk with each signal (for example, 0.5% of your trading capital which is $300,000, that is you are risking $1,500 per signal), then you will have to invest a lot money if the stop is very close and very little when your stop is far away. Then, it is imposible to control how much money you will need to trade this strategy (it could happen that you need much more than those $300,000 you have allocated to trade it if many of the stops are very close to the entry price).
Thanks a lot in advance.
Dario Polo
Hi Tom, thanks for another great article!
If you don't mind me asking; regarding your momentum eft strategy, what do you screen for? Is it something like rate of change over X period >X%?
Thanks Tom.
Alex