Finding an Edge

[tomorton]

most of them do trend-follow.. when volatility is gone, they don't make money.

first there were turtles;
then came the turtle faders;
then came the turtle fader faders;

and so on - the market is saturated with technicians and nobody is making money.

CTA's I'm not at all familiar with so I can't automatically draw what my performance will be from what their performance has been. I'm just a private retail trader but certainly not subject to the influences and pressures affecting CTA performance. I certainly will never have enough capital to move prices as I enter/exit a position.

Obviously, trends don't persist for ever, and an uptrend is rarely instantly replaced by a downtrend, but within clear constraints like these, why wouldn't trend-following work for a private retail trader?

 

[digitalnomad]

This whole “reading price action” thing is a fallacy. Don’t buy into it, or you’ll go broke.

You make money by numerically measuring price moves within a structured model. Only then can you have the confidence to push leverage, and know what should, or shouldn’t happen. No need for complex theories, but you need to be some form of a quant.

 

[tomorton]

This whole “reading price action” is a fallacy. Don’t buy into it, or you’ll go broke.

You make money by numerically measuring price moves within a structured model. Only then can you have the confidence to push leverage, and know what should, or shouldn’t happen. No need for complex theories, but you need to be some form of a quant.

Absolutely everybody in trading needs to be some sort of a quant?

(I'm not a price action reader either, I just see trends and ride them)

 

[Diskreet]

Many are saying that finding an edge is crucial in becoming a profitable trader. What are some important steps and stages in finding an edge? If you have an edge, what advice would you give those who are still looking?

I'd say that finding an edge is all about finding something that is repeatable. That's a layman's way of saying what others may have already suggested, which is finding positive expected value over time.

People often confine edge discovery to things like finding optimal chart patterns or correlation to say sentiment on Twitter or other such data sets. But what we fail to appreciate is that our real edge is often in the execution of a methodology or strategy that relies upon data, not the observation of interesting patterns in the data itself.

As Warren Buffet said, he could tell people about his strategy all day, but very few would benefit from it due to their inability to stick to it. Automated trading attempts to resolve this dilemma, but good luck with that...

 

[digitalnomad]

Absolutely everybody in trading needs to be some sort of a quant

If you want to make a serious business out of trading, with large stake investment, absolutely yes. If you’re hobbying around markets from your cell phone at work with a piker Robinhood account, than no.

 

[digitalnomad]

I'd say that finding an edge is all about finding something that is repeatable. That's a layman's way of saying what others may have already suggested, which is finding positive expected value over time.

People often confine edge discovery to things like finding optimal chart patterns or correlation to say sentiment on Twitter or other such data sets. But what we fail to appreciate is that our real edge is often in the execution of a methodology or strategy that relies upon data, not the observation of interesting patterns in the data itself.

As Warren Buffet said, he could tell people about his strategy all day, but very few would benefit from it due to their inability to stick to it. Automated trading attempts to resolve this dilemma, but good luck with that...

Probably the most logical first post on ET. Automating execution is a monumental task by itself, but doing the proper analysis in terms of data quality and accuracy is where most fail. It’s extremely difficult to do good walk-forward analysis, without sufficient time, and capital investment.

 

[Diskreet]

Probably the most logical first post on ET. Automating execution is a monumental task by itself, but doing the proper analysis in terms of data quality and accuracy is where most fail. It’s extremely difficult to do good walk-forward analysis, without sufficient time, and capital investment.

Thanks; my post was really a reflection of my own experience in trading. You're probably right in saying that data analysis is the bigger bottleneck for most people, but for me it's been more about fighting the urge to deviate from what I've already found to work. My immediate goal is to allow my strategy to pan itself out for the next couple of months so that the equity curve itself will justify sticking to the plan. I do appreciate that finding any edge whatsoever is extremely difficult, but we need to put as much if not more effort into being disciplined in seeing it blossom.

 

[tomorton]

If you want to make a serious business out of trading, with large stake investment, absolutely yes. If you’re hobbying around markets from your cell phone at work with a piker Robinhood account, than no.

I'm between the two.

 

[tomorton]

.....As Warren Buffet said, he could tell people about his strategy all day, but very few would benefit from it due to their inability to stick to it....

Richard Dennis I think.

 

[sle]

Supply/‘demand imbalances? Please tell me more!

What's with the sarcasm in this place? He is suggesting a valid approach, even if it's only a small part of alphas in the market.

There are a lot of situations where a certain product is overpriced or underpriced due to (drumroll!) supply/demand imbalances. Some of these imbalances are due to the various forms of risks premia such as liquidity or PnL variance. Some are structural due to things like the regulatory or compliance reasons, lack of market access and so on. Here is a simple example - in the markets where shorting in the EFP is difficult, futures basis would be very cheap. If one can find ways to trade that, it's a an alpha based on the supply/demand imbalance.