I don't trade ES, but what if you had sold ES last Friday at around 2073 at 09:50 based on the price movement on this 5min chart? Just one very basic price movement that you can find any day, now or 100 years ago, and in any time frame, in futures, stocks, currencies etc.
What kind of Edge you have?
- Thread starter carrer
- Start date
I don't trade ES, but what if you had sold ES last Friday at around 2073 at 09:50 based on the price movement on this 5min chart? Just one very basic price movement that you can find any day, now or 100 years ago, and in any time frame, in futures, stocks, currencies etc.
Something like ewmav with span 16 - ewmav with span 64 on daily data, volatility normalised. Decay parameter A=2/(S+1)
https://docs.google.com/spreadsheets/d/19YNLkKd9fKHf8YgiqYOfFIzI1TOBB6kxVSLHNHpS_y4/edit?usp=sharing
I don't use a seperate stop for exits, just a continous forecast, but it wouldn't make much difference if the stop was sized appropriately to match the speed of the crossover.
I don't optimise them. Instead I use a number of crossovers, dropping only those that are too expensive for a given instrument, or too slow (say more than a 6 months holding period), or are too correlated (say 95%). You don't need to do any optimisation for this. I then allocate forecast weights to the group of crossovers, but in an entire out of sample robust way. A simple average of the crossovers gives pretty similar results however.
When you say the parameters 'had' to be optimised, was that because the performance was disappointing? If so why? I can think of several reasons:
- Speed. These systems work pretty well for holding periods of about a week up to a few months. They used to work at faster frequencies of a few days (pre cost), but that seems to have flattened out in the last 20 years or so. Most of the people on this thread seem to be looking at intra day data. I've no idea if they work at that speed, although if I mined the data for long enough I'm sure I'd find some that did. I'd be surprised if you could overcome the execution slippage, since you'd generally have to be crossing the spread.
- Diversification. If you're looking to get a backtested Sharpe Ratio (SR) of 1.0 trading just one instrument, then yes you're going to have to optimise the hell out of it, and you're not going to make any money in real trading. A diversified but unoptimised set of EWMA crossovers probably comes in at an average SR of about 0.40 per instrument. But if you trade 20 them of them, across multiple asset classes, you can double that. Add a few other indicators and you're up to the SR of 1.0.
- Greed / lack of realism. If even a SR of 1.0 isn't enough for you, and you need / want a much higher return, perhaps because you haven't got much capital and you're targeting very high numbers, then again you're going to need to optimise to see a backtested SR that suits you. And again that's not a great idea.
GAT
And where would you exit?
As I said I don't trade ES, so I don't have access to a chart that would show the volume, which may add to the signal to exit. And further, I trade using various tick charts which will give more accurate entry and exit points. My point was to show that using the 5 or 1min charts that you referred to is very possible as it is done by those who can read price movement. But to try and answer your question, if I were trading off that 5 min chart, I would exit on the 10:15 or 10:20 bar or the 10:40 at the latest. But that of course is a question of tactics rather than the topic at hand being basic price movement on 5 and 1 min charts.
Last edited:
If you don't have a quantifiable edge, you are an edge provider not an edge taker. surf 2015
Can you give an example of what you meant by 'an edge in finding a new edge' ?
I believe that some edges do work for a very long period of time, and possibly won't fade away. Any thoughts?
Hi surf,
Would like to hear an opinion from you. Is a 100% discretionary approach (profitable) considered an edge to you?
Markets can only change so much and still be markets. There will always be greed and panic. Price action, IMHO,...especially when coupled with volume...is just a way to spot where the fear and greed start to take over. I honestly believe they can even spot insider trading too, but that's another discussion. My point is that Price Action T.A. can spot the beginning of a larger movement, and as long as you have fear, greed, and movement, it will work.
Agree, that you can make money providing a service to the market. Can't argue with that, and it is worth looking into. It's much more comfortable to be doing that than selling hotdogs out of a cart on Wall Street. All that being said, that does not mean that T.A. doesn't work. While you have identified a way to make money from the market, that doesn't mean that other methods will not work. Mutual exclusions.
I can't make money off 1 minute candles because my transaction costs are too high and I think Price Action needs some degree of movement to even get a fix on market sentiment to begin with. (By the time I'm ready to make my move, there isn't enough profit in it to be worth my risk). I might be able to make money off 5 minute candles, but even then, that would be hard for the same reasons outlined. That's why I like swing trading personally. Do you deny swing trading can work off of T.A.? There are big funds out there who need to buy lots and lots of shares...Can't those periods of accumulation be spotted and exploited?
Last edited:
It depends for how long its profitable and if you can articulate what you are doing. Otherwise its a streak of good luck.