Thread by @hfreflection

VENDOR SPOTLIGHT: LIGHTKEEPER:

I previously wrote about @Alpha_Theory and our day 1 usage for position sizing. Today I’m going to write about Lightkeeper, our portfolio analytics tool that I’m surprised isn’t as popular as it’s better known peer, Novus. A 🧵:

Most investors spend a lot of time trying to find good ideas and doing good research, and much less time thinking about how to get the most out of those ideas (position sizing) and reflecting on their mistakes and successes.

When we do spend time to reflect, we mostly use our memory, which is faulty, and fall victim to recency bias or other cognitive biases. A tool like lightkeeper allows us to more accurately reflect on our successes and failures with data unencumbered by our emotions.

At its core, Lighhtkeeper is a portfolio analytics tool that allows you to measure high level portfolio stats like alpha, batting average, slugging %, beta, P&l, roic, and other key stats in granular ways - by geo, long/short, sector, etc.

Furthermore, and maybe most powerfully, you can create custom tags that allow you to analyze your performance by any category you like - eg by analyst, idea source, investment style, etc.

There are other tools that do this, I just prefer lightkeeper. Regardless of what tool you use, even if it’s manual, I’m a big proponent of having good data upon which to reflect. I’m excited to test my hypotheses and see the results - curios to hear other ideas of what to track

1) I plan to track idea performance by source of idea - was it internally or externally generated and, if so, by who. I think my internal ideas perform better, and I wonder whether ideas I e source

2) I want to track idea performance by qualitative criteria I use to size positions within alpha theory. I have hypotheses of what makes an idea better or worse and I’d be curious to see what does best empirically after I have a large enough sample size.

3) I plan to eventually track idea performance (namely alpha) for other idea generators on the team, to aid in providing feedback and assessing performance over time. I plan to give idea generators access to their lightkeeper data once a year as part of reviews.

4) I have lots of different types of longs I will invest in and I’m curious how my performance varies by idea type - eg broken growth, deep value, liquidations, GARP, etc. same on short side. I will track performance this way as well.

5) what are other things you think would be interesting to track other than the obvious (eg alpha long vs short)? My thinking is once I have 2-3 years of data I can draw some tentative conclusions and make thoughtful modifications to better maximize performance.