Is it important to define your edge mathematically? If so, how do you go about finding one?

[Snuskpelle]

1. At a minimum you need to know if your profit expectancy (as a function of win rate and size of wins and losses) is positive, that does not necessarily involve any math but you need to know the equation and understand its implications. It's very simple but some people posting over the years can't grasp it, as exemplified when they're raving about their 96% win rate until suddenly they go quiet. Oh yeah, and that financial returns are not normally distributed.

 

[deaddog]

two questions

1. Is it important to define your edge mathematically?

1. 
   How else are you going to define it?
   
   

   [*]How do you go about trying to find an edge?

You study the market and find something that you can exploit for a probable profit.
If you can find a coin flip that comes up heads 55% of the time you have an edge.
If you could find a black jack game where you got to see both of the dealers cards you have an edge.
If you bet on the direction a stock will go, cut your losses quickly and let your winners run, you have an edge.

 

[monet]

Cue arguments about what qualifies as edge...

 

[Mr Subliminal]

....
If you bet on the direction a stock will go, cut your losses quickly and let your winners run, you have an edge.

If only.

 

[smallfil]

two questions

1. Is it important to define your edge mathematically?
   
   
2. How do you go about trying to find an edge?

It is nothing more than constructing a trading system, backtesting it over a large enough sample then, using the expectation formula to determine if you have a trading edge. That is, you need to have a positive expectation like the Las Vegas casinos. Any game you play in the casinos has an edge for the casinos at the expense of the gamblers. Gamblers are playing those games under negative expectation. To make monies in the stockmarket, you have to have an edge or positive expectation. If not, you are merely gambling. This article on expected value is from Investopedia.com

https://www.investopedia.com/terms/e/expected-value.asp

 

[Sekiyo]

Linda Raschke

 

[JonLivingston]

Become a stat arb trader, collect data, build models, test, iteratively refine.

Most manual discretionary traders use their paper/trades as the basis of their model and system.

Many can get lost in the math of it all, others just focus on profits booked.

You’ll have to build discernment while you explore your path - that only comes through experience.

Beautifully put. Thank you

 

[deaddog]

If only.

Not working for you? What's the problem?

 

[JonLivingston]

Yes but …

You can tell how reliable is your source,
The more reliable the more you should bet.

It’s only important for sizing purposes.
If you don’t want to size optimally then it’s useless.

If you’re not willing to put 20% on the table then stick to your normal size of 1,2 or 5%.

For most traders the optimal bet size is 0%

You find an edge by extracting more money than you put into it. That’s what it means: It has Positive Expected Value. You’re paid more for the risk than what’s implied.

If you find a situation where you can win 60% a 1 to 1 reward to risk then you’ve found an edge.

A profitable, repeatable and scalable trade.

You find a sucker and you take his money.
At the right time + At the right place.
With the minimum skills required.

It’s like poker … Be tight agressive,
In a sea of deep pockets but drunk players.

Sharks smell blood.
Most of the time it’s capital preservation.

But fools are long gone when the real opportunity appears.

Look for easy dollars vs hard pennies environment.

Anyone can mak money in a bull market.
But few can keep it. Remember.

Find opportunities.
Define what it looks like.
& How you would exploit it.
Here you have your edge.

Before being mathematical it’s common sense.
No need to make shit looks scientific.

Sorry this is even better. I'll stop now.

 

[schizo]

two questions

1. Is it important to define your edge mathematically?
   
   
2. How do you go about trying to find an edge?

Isn't this what quants do, day in and day out? They use statistical model for their algos, which is pretty much all math-driven.