Serious Question

[Peter8519]

60 lines of code will definitely not be an edge. 6,000 lines maybe. Of course, 60,000 lines would fair better.

p.s. I have around 3200 lines of perl script and could hardly keep up with the market beast.

 

[Metamega]

I find myself very humble with possible expectations. The markets are highly competitive and to think theirs any massive edges out there seems absurd.

I like to think more in risk/return metrics not simply returns.

Like the threads about selling option premium as some debit machine, theirs risks the poster doesn’t see.

 

[rb7]

Let's say I have an edge and it is as simple as applying a very clever algorithm to linear combinations of futures and stock index data feeds.

Let's say it's 60 lines of code. A script then generates a "signal" that gives you early warnings of reversals, tops, bottoms, trends and immediate volatility.

Using this and then comparing it to price action easily enables one to profit almost continuously during RTH.

Assume probability of profit > 85% and 10% on account equity daily is relatively easy to do with standard exchange margin.

What would you do with this thing?

1) Keep it a secret, trade the edge and run up your account XXX%, then repeat.
2) Approach a prop firm?
3) Trade it, make a track record and then approach a prop firm?
4) Sell it to people?
5) Something else?

What I have learned is that people won't believe this. Also, fake educators and snake oil salesmen peddling pitchforks, crossovers, and the rest have poisoned the well. It's almost like this sort of thing would have to be marketed with a website, videos, whatever. And even if I did all that, most people wouldn't even understand the concept, let alone the actual code and script, even though it would be obvious it worked.

I'm leaning toward doing (1) a lot and then maybe (3) later on.

I'm interested in your opinions. Maybe one of you even has a story about what you did with something similar.

Thanks.

You don't know how good (or bad!) a system (or edge) before trying it (for real!), in different market conditions.
Start trading 'your edge' with real money, and in a year or so, you'll be able to answer your question.

 

[tommcginnis]

Not really.
If it is an intraday strategy it doesn't need one year to verify.
Because there can be many trades in one day and different day the intraday chart pattern just repeat in same way. Also different market conditions produce the same intraday chart pattern. If you put up an intraday chart and roll back to like one month, it will include almost all condition that an intraday strategy could experience.So I would say one month intraday trading record is enough to verify.

No way, Ray.
There are single, whole months that have high flops and flips, whole months where volatility approaches single digits, whole months where the air leaks slowly out of the market balloon. Their intraday behavior reflects that variety very much, and whether your (or the OP's) intraday trading strategy could operate well in those different environments is an open question.

[FWIW, an ATR with a MA set up to mimic market implied vol makes a great temperature gauge. That's another 2 lines of code for the OP -- as a Go/NoGo starting point. ]

 

[bbpp]

No way, Ray.
There are single, whole months that have high flops and flips, whole months where volatility approaches single digits, whole months where the air leaks slowly out of the market balloon. Their intraday behavior reflects that variety very much, and whether your (or the OP's) intraday trading strategy could operate well in those different environments is an open question.

[FWIW, an ATR with a MA set up to mimic market implied vol makes a great temperature gauge. That's another 2 lines of code for the OP -- as a Go/NoGo starting point. ]

So did you find a strategy that can make daily 10%? No, so your strategy is not good enough, right? That's why you see market varies big day to day. Because your strategy is not robust enough to deal with market situation change.
A strategy that can make 10% daily is a strategy that is robust enough to deal with (in your eyes) different situation. In another word,his strategy view those situations as no different day to day.

 

[tommcginnis]

So did you find a strategy that can make daily 10%? No, so your strategy is not good enough, right? That's why you see market varies big day to day. Because your strategy is not robust enough to deal with market situation change.
A strategy that can make 10% daily is a strategy that is robust enough to deal with (in your eyes) different situation. In another word,his strategy view those situations as no different day to day.

You can't have it both ways, Charlie. Either robust performance adds value to the strategy, or testing it is irrelevant. Your lack of cohesive logic is rather glaring. Fire your risk manager.

 

[bbpp]

Don't make conclusion on things you don't understand.

For example , a strategy based on shorter time SMA cross longer time SMA will not be affected by volatility change. I am not saying this strategy will make 10% daily but at least it will work in different volatility situation .

 

[tommcginnis]

For example , a strategy based on shorter time SMA cross longer time SMA will not be affected by volatility change.

Errrrr, whut? I think you must have a fold in your space-time blanket, but in *this* reality, whether you get shaken out or are able to maintain a signaled trade is very much dependent on pricing's true range, by construction: either ATR or volatility will tell the tale, AND will vary from month-to-month, day-to-day, and since calendar 2018, hellfire-sure, may vary from hour to hour. Again, fire your risk manager.

 

[bbpp]

I will tell you what my strategy is doing.

My strategy just gave a buy signal at ES 2923.
And my calculated stop is 2916.
So when my entry is made, a stop is placed at the same time.
Since this stop is not too far from my entry, so I used normal amount of contracts.
Now supposed today volatility is 3 time as usual, I would find my calculated stop is 3 times as far from my entry as usual. So my entry amount of contracts will be reduced to 1/3 of usual amount.
Now tell me how can I be affected by unusually high volatility today?

 

[fan27]

Trade with real money (get a track record), start a journal on this site and make sure you have a tight LinkedIn profile. I have done those three but also have the advantage of having a substantial career in tech in addition to having a quant related product offering and have been contacted by a couple of hedge funds via LinkedIn.

Also, in the event that any opportunities with third parties arise, make sure you have a good lawyer who can help with contract negotiation to help prevent getting ripped off.