Truly. There's some standard approaches like beta-adjusting that can be done to put things on equal footing in different market regimes but it's only one part of what determines a stock's performance.
Testing for the significance of someone's average returns for example often can be rules of thumb. For example, given a return stream of a trading strategy, the Sharpe ratio (mean return/vol) is directly related to the t-stat of the mean return being different than 0 (mean ret / (vol/sqrt(n)). Several simplifying assumptions must be made about the distributions of returns that don't line up with reality to do the above but it's mostly a heuristic that will let you say "ah, the Sharpe is X over the last 5 years? That's significantly different enough than 0 for it to be tradeable."
This is a bit like a popular hedge fund strategy known as alpha capture. In the alpha capture strategy, funds look at the buy/sell recommendations made by analysts (for example those at the major banks). Applying a systematic approach means that they can measure the accuracy of each individual analyst and choose whether or not to follow along by trading the advice from that person. The best analysts eventually get hired into funds for their picks to be exclusive.
I built something (DueDilly) that works in a similar way by monitoring the performance of Reddit user's stock recommendation and tracking how each idea performs and which users do well. You can choose to trade alongside them based on their track records.
I did consider beta adjusting performance and may do this in the future -- it does require more infra set up (ex. computing historical betas for these posts back in time).
Since my target demographic in this case was people on Reddit following trading advice I wanted to keep things straightforward first and slowly over time introduce those educational components to get it closer to how the industry thinks about it.
I've been in the industry for several years and had the same reaction when I first starting seeing wallstreetbets posts. For a while I even thought it stood for "double down" when it seemed like these were encouraging posts. Good point from you and the parent commenter -- I'll make that clearer.
So 99% of DDs are long and I didn't want to get sidetracked trying to build a guessing system since many Reddit posts delete their contents (the titles are not removable which helps me).
For that reason I auto-classify as bullish but there's a report page that anyone can click to report mis-classified posts.
I consider 1 week, 1 month, and 1 year horizons and report all 3. For people pushing things around you expect to see it more on these short/medium term horizons.
So this is a quick hack -- I have a user registration system but I didn't want people parking on reddit usernames. For any user that gets scraped from Reddit I add a "r." in front of their username. The '.' character is invalid in usernames that are user-created. If you sign in with Reddit, you can change the username to whatever.
Thanks! I'm unable to post it there unfortunately (or really anywhere on stocks subreddits) due to moderation rules around promoting webapps but I've seen a lot of those posts.
Thanks! Yeah, I don't want to put anyone down with the worst traders page but worst stocks could be fun, especially if it's done as a way to bring people together around their losses and better inform themselves on what to look out for.
I've noticed something similar with mentions being a poor leading indicator. Anyone can submit any reddit post to the site so hoping with exposure these older traders can submit those older posts as well, as it helps corroborate their track record.
Thanks! I know exactly what you mean. I've experimented with using medians instead of means for aggregating performance as well to lessen the impact of that sort of thing.
Currently, it's pretty difficult to crack the top without something like that. At the same time, I think (as another commenter pointed out) the distribution of returns for the way these traders pick stocks is pretty skewed/lottery-like. One hesitation I have about the median is that although it's more robust to outliers its less reflective of the true performance achieved.
Maybe putting some stocks in a "meme tier" and then excluding those might capture what your describing, thanks for the feedback!
I would like to open it up more generically to allow for sharing any ideas (like from not Reddit) -- the platform isn't tied to it.
The big issue for me is accountability. People make public claims and delete the things that didn't work out, so a lot of platforms can be gamed. On Reddit, the titles of posts can never be deleted so that provides one way to verify it. I think that's valuable if you can start auto-adding tweets/blog posts/reddit so people don't mess with their records.
Built my first "real" webapp to help retail traders figure out whether to trust the people posting trading ideas on Reddit. It links each post to its performance and lets you see the prior trading ideas suggested by each person.
Not much of a developer but I've got several years of experience in the quant finance industry. Seeing people talk about plowing their rent money into an idea when the ideas are hardly financially sound was really frustrating. At the end of the day, it was my non-technical friends and people like them being sucked into these pump and dump schemes.
Happy to get any feedback. Here's an example page if you're not sure what to click:
Testing for the significance of someone's average returns for example often can be rules of thumb. For example, given a return stream of a trading strategy, the Sharpe ratio (mean return/vol) is directly related to the t-stat of the mean return being different than 0 (mean ret / (vol/sqrt(n)). Several simplifying assumptions must be made about the distributions of returns that don't line up with reality to do the above but it's mostly a heuristic that will let you say "ah, the Sharpe is X over the last 5 years? That's significantly different enough than 0 for it to be tradeable."