Publish-and-download-of-strategies

It would be nice to also have a rating system as to how good these strategies might be. Perhaps a table that compares them using the Sharpe Ratio and Wealth-Lab Score for the Dow 30 or S&P 500 over the last 6 years. Posters should fill in the table when they post a strategy.

Yes, I know performance depends of market climate and such, so these scores may not be comparable. But something is better than nothing.

superticker,

I do believe you are suggesting the Glitch and company revive the old WL3/4 website! 😉

Vince

Just what I was thinking!!

We are making solid progress on this larger-scale development item!

You can now Publish Strategies in our Web Builder, and they'll automaticlly get downloaded into WL7 upon startup.

Are there any plans to allow publish/download for coded (or even compiled) strategies?

Currently no.

So, who will be the first person to publish a Strategy in the new WL7 Web Builder??

I was going to publish a simple one on the SPY but no flexibility on the dataset yet?

No.

Cool - thank you. The web builder unfortunately wouldn't work for any strategy I'd consider publishing. It's still a nice feature though.

QUOTE:
No.

Someone is a grumpypants ;-)

First of all, thank you very much for implementing this change request!

Any chance that we will eventually have a way to share coded strategies and not necessarily limited to a dataset? Even something simple like saving a file to a specific folder that gets published to some git repo, from where we can download it (manually if needed; automatically to a preset folder even better).

Once again thank you for the nice work already done!

Who knows what will happen in the future? Based on user demand we might very well get something like that in place!

I do not expect that this will be overwhelmingly popular. From a decade of experience with the similar service for WL6 downloadable strategies, the demand for it was not high. https://wl6.wealth-lab.com/Extensions/Strategy-Finder/

Out of 363 strategies, 314 were uploaded by yours truly and 33 by Cone. This leaves us with 16 strategies uploaded by WL6 users from 2009 to end of 2020 when we stopped the WL6 activity. Or roughly 1.5 strategies in a year.

While its predecessor service (for WL3/4) had over a thousand downloadable strategies, many hundreds of them were just parameter modifications and tweaks of similar "value" (throwed in an indicator etc).

QUOTE:
Out of 363 strategies, 314 were uploaded by yours truly and 33 by Cone.

These are the upload numbers. I think it is more interesting to look at the download numbers. (Do you have them available, also for the "old days"?)

Most members will download strategies, and only a few seniors will upload.

For me the download area was one of the main reasons to start with WL and join the WL community.

I think the download area (along with the discussion forum) were the main highlights in the old (pre-fidelity) days.

And even today, with WL7, I feel the urge to share (working) code examples, and (coded) ideas.

If I post something in the forum or create a blog entry I'd like to be able to refer to a downloadable strategy which underpins my point, and so forth.

I think it would be a big step forward for WL and its community if there were (uploadable and) downloadable (coded) strategies.

I completely agree with Dr. Koch,

A few years ago, I did an extensive analysis of all the backtesting software available. At the time, one of the main reasons why I decided for WL6 was the available strategy corpora!

Sure, most were uploaded by Eugene. But pretty much all my trading systems were inspired by that base of published strategies; especially the great work done for TASC magazine.

Sometimes, when I wanted to trade a new market (futures contract), I would start by running a strategy ranking of all available strategies; in most cases this would point to 2-3 strategies that adapted (sometimes a lot) could lead to something profitable.

As developers, you might feel tempted to keep adding features to the product; but please don’t neglect the allure that a good corpora of strategies has for people that are just starting or don’t have the time or the skills to start from scratch.

Of course that the new building blocks strategies are a major breakthrough (probably revolutionary in the space of backtesting software) and make it much easier for anyone to come up with a strategy quickly but, still, I believe it would be really good to have a much richer list of public strategies.

Hear hear! Hope to see some new community published strategies soon!

I hope when being able to upload code strategies, traders will still be sure there local strategies are secured - I remember this was one of the top questions from new WL7 users.

But at the same time I agree with being able to upload and download code strategies to be a good idea.

No worries. At any rate, uploaded strategies must go through moderation prior to become published. It would be reasonable to provide an Upload page so your local strategies aren't at risk of disclose.

@alkimist

QUOTE:
Even something simple like saving a file to a specific folder that gets published to some git repo, from where we can download it (manually if needed; automatically to a preset folder even better)

We have a feature request for a loosely related (but not identical) idea:
https://www.wealth-lab.com/Discussion/Version-control-for-strategies-7068

Considering that every participant of this discussion is a developer to some extent... maybe some kind of Git integration would be the way to go.

QUOTE:
Considering that every participant of this discussion is a developer

A collection of downloadable strategies makes the product attractive for new customers....

Maybe. We could bundle more Sample strategies with Wealth-Lab and its Extensions as they get requested. On a related note, I noticed certain lack of strategy coding requests compared to what was before.

It's just because we are ashamed to ask, considering the amount of work that you have already been putting into our various change requests. ;-)

More strategies would be very very much appreciated!

For example, the work that you keep doing for TASC would be nice to have available.

Thank you for the kind words.

At least it's easy to share the TASC work since it's available on the magazine's website:
https://www.wealth-lab.com/Discussion/Code-from-S-amp-C-Traders-39-Tips-6203

QUOTE:
For me the download area was one of the main reasons to start with WL and join the WL community.

I totally agree. When I shop for a software solution, one feature I look for is a support forum for that product, which WL has, and that was the main reason for going with WL in the first place. The active forum is the most valuable asset WL has!

But the library of downloadable strategies is also a big plus for users shopping for a trading product.

What turned me off later after getting into WL was that most of the downloadable strategies didn't make money out of the box. They weren't turnkey solutions as we come to expect in engineering circles with software products. My recommendation is this:

*) The worst thing you can do--especially with new users--is disappoint them! In this spirit, I would separate the turnkey strategies from the indicator demo strategies so there aren't any unrealistic expectations.

*) For the turnkey strategies, performance metrics are important. I would post those for each turnkey strategy.

*) For the indicator demo strategies, I would post which indicator they are demonstrating. Obviously, performance metrics for demo strategies are irrelevant because no one expects them to make money. And without that expectation, there won't be any disappointment.

Those indicator demo strategies aren't always well developed which is kind of expected for new content. It may be reasonable to not bundle such examples with WL7 or upload them to Public Strategies (like I did with WL6 every month).

QUOTE:
... indicator demo strategies aren't always well developed ... It may be reasonable to not bundle ... them to Public Strategies

There's nothing wrong with posting the indicator demo strategies, but they need to be posted in a "different venue" so newbies know their intent is not to make money out of the box.

The assumption all newbies "expect" is that all posted strategies are turnkey and make money out of the box. That assumption brings great disappointment.

To avoid disappointment, classify the groupings as " 'near' production ready" and "simplified indicator demos". The goal here is to dispel unrealistic exceptions.