If you have programming skill and a good understanding of statistics, do the following:
1. Identify a subset of equities in the total market which a) have fairly one dimensional revenue streams, b) have a market capitalization of at least ~$1-2B, and c) are not prone to extraordinary hype or tech-centric accounting, such that e.g. a "win" or a "loss" in an earnings announcement is fairly straightforward to understand (and therefore you can more easily, if not perfectly predict how the market will react).
2. Identify a strong, legal source of alternative data that maps directly to the revenue stream of one of these companies. The more difficult to find and collect, the better. Use your programming skills to automate the collection and curation of this dataset.
3. Incubate your dataset for a period of several months, then build it into a timeseries. Using the timeseries, build a model that forecasts the expected revenue of each particular company using historical 10-K and 10-Q documents.
4. For the companies whose data imply a jump in either direction that is very unexpected (according to e.g. the aggregate analyst consensus), take a contrarian position in the equity. If you're feeling very confident and have a higher risk tolerance, study options and take the corresponding derivative position.
5. In particular, establish a target win rate overall, a target tolerable drawdown period overall, and a target exit price (sufficient win or bearable loss) for each position, then follow it.
If you do this correctly and consistently, you will profit significantly and consistently enough that your system will be fully distinguishable from uninformed gambling. To equip you with a bit of meta-analysis here, this outline works because a) all trading strategies profit from finding opportunities to exploit pricing inefficiencies in various securities (or groups thereof), and b) the only way to deliberately identify those opportunities is by having information, access, or techniques that the broader market does not have yet (or else the price would reflect that information).
The great difficulty in this process is finding and analyzing the alternative data in the first place. As a fallback, if you're not confident you can build a trading strategy with this data you can also sell it to hedge funds, who will be very happy to buy it if it actually maps to revenue and is otherwise unknown.
Years ago I traded energy futures and we hired someone away from a rival. He knew a trick for seeing crude oil price changes on CME a few milliseconds before they sent the updates in their data feed. For a brief period we were printing money across the energy complex, but eventually it stopped working well. The guy who taught us the trick jumped from firm to firm, many others independently learned it, and it became the worst kept secret in trading: https://outline.com/MHp6Yu
If you've done this earnings forecasting yourself, how much size can you trade before the market moves enough to make the risk/reward unfavorable? For a retail guy it's probably not an issue, but curious if major funds are doing it at scale.
How does selling to hedge funds work when the information is valuable only insofar as few others have it? I suppose you could have an exclusivity contract but there's a strong incentive to sell to multiple buyers. Is it more of a relationship/reputation type setup? Are there brokers of some sort that help filter disreputable sellers out?
From my casual observations, usually about 8 figures or so. It was not unusual for us to deliver a particularly impactful report, then have institutional capital rush into it and push up the price fairly quickly. Then they’d hold that position for a month or so before earnings. Typically “word on the street” would be that the big movements signalled smart money, but usually there was still premium left over to capture during the actual earnings announcement. Humorously, it was sometimes frustrating when institutional money would make large bets in the direction opposite to our forecast, because despite being eventually vindicated, clients would get restless about it.
> How does selling to hedge funds work when the information is valuable only insofar as few others have it? I suppose you could have an exclusivity contract but there's a strong incentive to sell to multiple buyers. Is it more of a relationship/reputation type setup? Are there brokers of some sort that help filter disreputable sellers out?
It’s not typically the set of all hedge funds purchasing the data for any given equity (that is, unless it’s groundbreaking), it’s the subset of hedge funds which already have an interest in the particular equity. That limits the competition and information diffusion somewhat. For certain exceptionally high value data we did try exclusivity contracts so that its value would last longer. For example, before I left the research firm I was working at last year I developed (to my knowledge) the only non-drone, purely web-based method for forecasting the exact number and type of all vehicles Tesla produced with a <1% margin of error - to the point we knew well ahead of time that the Model 3s would completely miss. We were marketing that data to particularly long term, well known clients who could be trusted not to burn the utility of the information.
From there, it’s as you say: there is absolutely a reputation system with a strong basis in long term relationships. It was common to have to jump on calls directly with analysts at funds to talk through the forecast. And yes, there are brokers to help facilitate this entire process.
If you’d like to chat about this more, you’re welcome to email me. I can’t go into deeper detail for much of the proprietary workings, but I’m happy to talk about things that don’t have an NDA covering them.