Would you agree with the following? Why, why not? Never risking more than 2% of capital on any individual trade is a widely adhered to rule among professional traders. Unfortunately, this rule is commonly understood too simplistically and consequently gets a lot of undeserved flak for being overly conservative and not capable of generating enough profit. But the rule is not to be understood simply as e.g. "take a single vanilla position in an asset and exit if loss equals 2% of capital. It can also go something like the below highly hypothetical and highly simplified example: Capital: $100 000 Identify a setup and go long $10 000 in e.g. SPX with x20 leverage. Set stop loss at 1% of SPX. Thus, if SPX drops 1%, the loss is $2000 (=2% of capital) The stop loss is moving. Day 1: Let's say SPX rose by 1%. The trade is now up $2000. Another identical position is taken. Day 2: Let's say SPX rose another 1%. The first position is now up $4000 and the second $2000. Another identical position is taken. Day 3: Let's say SPX rose another 1%. The first position is now up $6000, the second $4000 and the third $2000. This should be enough to satisfy even the greediest traders, in which case all positions are closed for a very handsome profit. But lets see what happens if the positions are left open. Day 4: Let's say SPX dropped 1%, triggering the stop loss for all three positions. Each of them result in $2000 loss day = $6000, but net profit is still $6000. Again, note that this is a highly hypothetical and highly simplified example. But it should still illustrate the power of this money management rule.
That is similar to what I do a few times a year during relief rallies. However I would buy more at the start and lower the amount each time so the average price doesn't increase as much with each step. In the situation described on day 4 if I was still thinking the market would rise I would only close them out if the initial capital dropped back to 100,000 USD. I think its a good strategy if the opportunity comes along.
How are you determining what the 2% possible loss level is for each allocation? Would you use VAR or some other method?
I'm probably misunderstanding your question, but isn't it clear from the original post? "...long $10 000 in e.g. SPX with x20 leverage. Set stop loss at 1% of SPX. Thus, if SPX drops 1%, the loss is $2000 (=2% of capital) The stop loss is moving."
Interesting. If you don't mind, could you give a more complete description using a similar hypothetical simplified example but with your version?
Not really clear to me, because S**T happens and stops are not always protection. You are saying the most you can lose is the value of the stop. Would you use the same criteria for any investment regardless of how volatile is or how liquid it is? If you take an overnight position in a currency pair, stock index, commodity future or biotech, your risk is different with each allocation. You need to determine an expectancy of possible loss if you want to determine what that 2% is for each investment in a portfolio.
Right. Look, it's a highly hypothetical and highly simplified example that intentionally leaves a lot out. Of course what you bring up (and lots of other things) can't simply be glanced over in actual trading. EDIT: For the record though, I've coded up a non-commercial, open source position calculator for that very purpose: http://positioncalculator.bedobi.com for anyone who is interested. Desired stop loss price per unit can be a fraction of price per unit if volatility is high or whatever. The point is that the anticipated loss should never be more than 2% of capital.
This is a senseless discussion. It only makes sense if it is at least close to reality. Discussions about things that will never happen are a waist of energy. So it does not demonstrate the power of money management at all. And stops are no guarantee that you will not lose more money. What if you get whipsawed 5 times and lose 5 times your stop? With hypothetics you can prove anything.
I disagree. It's impossible to have a meaningful conversation about anything unless you can atomically isolate the topic. I could have written a virtually endless original post bringing up everything that could possibly go wrong but that would simply be beyond the scope of the topic. If you want to talk about the technicals of stop losses and how they may or may not work then by all means do that in a designated thread.