Data, Analytics, Automation and supporting decision making in the community

As I spend more and more time looking at different aspects of trading, looking at charts and most importantly listening in to VC each day, I can’t help but think about perhaps unrealistic ideas to improve different type of plays the community does.

I am not an experienced trader by any measure, I most likely will never come up with a DD from scratch, but what I do have, is a sometimes unhealthy level of interest in data, data driven decision making and coming up with stories based on data.

Why I’m saying this is because listening to VC during plays, there is always speculation of what is happening, and occasionally about things that I suspect could be better explaned, or confirmed through data analysis. Same goes to getting into plays, but more importantly to dynamically planning exit strategies based on live data. I know that Gods have already done a lot of work in this regard to generate signals, which understandibly is not public generally, but listening to conversations about it in the past months, it`s not necessarily complete and may see some value from brainstorming in general about the topic, even from people without knowledge of the current works.

Now my proposal for discussion is very vague, maybe worth opening seperate conversations in the future if there is enough interest. What I lack is knowledge on what kind of data is potentially available, on what price levels, what is currently available to our community currently, should there be interest in looking into data more closely. The reason why I ask this here instead of googling it to figure out different data subscriptions is that if anyone has experience on the granularity we can possibly get, that`s likely more specific then what I would find.

Some topics/ideas to kick off the discussion:

  • working on models to better predict OI change intraday
    - I wrote some high level ideas in the ESSC thread, I can copy them over here if needed
  • building a scanner of SEC documents to help identifying different aspects of what we consider a potential play
    - getting redemption rates from filings, potentially automatically calculating the available free float after
    redemptions (preliminary calc, to generate at least a signal that it is worth checking)
    - scan filings to see new dilutions/changes in float as soon as possible
  • currently available data to the community, available data to work with to build these even if it is not neceserrily public to the community currently
  • potential future data investments if there is interest

I know some of these topics have probably been mentioned before as plans, maybe some of these things already exist and I just don`t know about it, feel free to say so if that is the case. If you have any ideas in the topic, please add to it, this could be the place to discuss.

1 Like

I’m going to necro this thread because it honestly is a fantastic freeze-frame of the things we were thinking about “way back when”. For those that haven’t been following over the past many months I’ve been building out Mimir in totally random in line with my “manifesto” found here:

Mimir development has hit a point where it’s time to start thinking about exactly the things this post puts forward. Mimir is loaded with news headlines, earnings data, economic data, fundamentals, quotes and more now and it’s all in a system that makes it easy to utilize functions written in any language. For those that haven’t seen it, I’ve released a development playground for Mimir and will continue to build upon the available connectors as interest increases. For more on that you can read here: GitHub - AscendedTrading/Mimir: The system that powers Valhalla and all it's features

What we learned in the early stages of building out the community’s tech is that if we just build bots that dump into channels the information goes unused. So we needed a full system capable of not just aggregating information but one that could actively make decisions based on that information and the foundation for that is now there.

Current ideas that I’m aggressively going to be attacking are building out Mimir’s ability to gauge sentiment of news headlines in two distinct ways.

  1. I’m going to use basic models to calculate and display “How the algos are probably reading it”. A display designed not to be accurate in line with how we’ve observed in the market but instead to represent the “baseline”.

  2. I want to work on our own in-house model designed to ingest and score headlines as they come in, initially based on user input (Buybacks = 10 points importance, CEO departure = 10 points, Winning an award = 5 points) but to also track stock performance of those headlines after the fact in an attempt to generate a process that can watch the stream of headlines and select ones that are most likely to have a substantial effect on the underlying.

Another area that I’d like to put heavy development into is earnings. Mimir now has an earnings database and can calculate percentage beats, with some criteria added to that, we can generate signals in the morning for what specific earnings plays to pay attention to. This can be using fundamentals data to bolster the results or by scanning SEC filings, a functionality I’ve almost completed.

To be clear, this thread is to discussing creating processes in Mimir to help identify plays. So any idea is welcome.


And just for a clarification, this thread isn’t just for people that are able to code. Most things I can create pretty easily on my own thanks to the way Mimir is built, it’s really just identifying the criteria that we should be looking for. @jjcox82 @BBarna @Navi, etc.


I’d be interested in contributing. A big part of my job (which happens to be my favourite part) is defining logic in data models, so if there is any brainstorming going on in VC or here, I’ll join if I have time.


Phew I thought I was about to have to really disappoint you with some coding.

For one this is awesome and I look forward to seeing it come to fruition and appreciate all your efforts to streamline the server.

IE earnings it has seemed to me that guidance has been more heavily waited in our current economic situation. Many are meeting or beating EPS and revenues but saying future might not be so rosy and even those are drilling. Some miss and guide to upside and they seem to go up. I’d liken this to most funds being largely well positioned in many of these. That if the upside is even a bit cloudy they lock in the profits.

Buybacks seem to contribute to multi day runners.

PT upgrades/downgrades always seem to have big effect and most come day after earnings. I know tweets come out on Tf but seems like they don’t pick them all up as many have 4-5 PT changes at different times of day.

Not sure how hard it is to do but if there was a way to correlate earnings tickers premarket movements to SPY and SPX some kind of percentage divergence some times I feel like that’s my biggest watch. For example APPL is initially down on ER but premarket has moved up 1 percent where as SPY has moved down 1 percent. This may be way hard but see intial statement I can work my iPhone and that’s about it. This may be better left to conversation in Tf or morning forum posts.

Can’t wait to see this. And happy to help any way I can.


Yeah, I think potentially something that generates a chart of Relative strength/Weakness to SPY over different timeframes could be useful.


Any help would be greatly appreciated. We get pretty far out of my area of expertise when dealing with data models. I’ll definitely @ you if we have any VC sessions and feel free to chime in on this thread if you see anything you’d like to participate in or if you have any ideas for things we can attack.

I’d like to get more into guidance analysis as it could lead to super profitable post drop plays. Guidance is often delayed in the news streams so if we can beat them to the punch on our side of things it might produce an advantage (for example if a stock drops off the initial release and we’re able to determine guidance is good or bad programmatically we could frontrun a reversal).

So the question is “where is guidance released” and is that accessible? Is it in the SEC filing, or is it a separate release?

The data that Mimir has is pretty comprehensive on this topic, it just needs to be parsed out of the news headlines feed which is something that a lot of these projects are going to rely on I think. An example is:


There is a bunch of fields in the data we get that generates this, one of which is the ticker. We can easily start storing and cataloguing all the PTs that come through the pipeline and attach them to “symbols” pages on the forum and the yet-to-be-created earnings page with will look similar to this which has the results, a “beat” calculation on the far right and then we can include an X for if the system detects a buyback and a total of up/downgrades:

A strength vs SPY figure is relatively easy to create and is something that I would definitely be interested in as well as SPY divergence was easily one of the largest cornerstones of my personal strategy. Would probably look more like a number than a full fledged chart I think, but is definitely something I would like to create to use in the various signals we end up generating.


EW posts guidance of course but is sort of elementary arrow up and arrow down or the mixed. I typically read mine in the PR that tend to come out even before their calls take place from individual ticker investor relations pages. This would be ridiculously hard to track I’d think. I like to read them because sometimes they adjust guidance but it’s really minuscule such as INTU which didn’t out weigh their beats and reversed intial reaction. I’ll look more into the guidance this weekend and try to see what might be the quickest way to attain it might be.

PTs could maybe easily be pulled from TOS I’d think since you are already extracting from there. And they seem to be as quick as any. The chart with the X would be easily readable and should work great.

I’ll watch few posts tomorrow and see where guidance comes from first or quickest Algos are definitely snatching it somehow in short amount of time.

1 Like