Militia Capital commentary for the year ended December 31, 2024.
Net of .5%/year management fee and 25% performance fee over S&P 500 returns.
Militia Capital launched February 1st, 2021. I was investing from my personal account before that.
Read more hedge fund letters here
I use Interactive Broker’s portfolio analyst to create the reports that I share in these letters. The Sortino ratio above comes from them. Starting in 2024, right after new years, portfolio analyst has been adding an extra trading day to each week. It’s some weird glitch. This is causing the fund’s Sortino to be understated by around 12% and I’m going to adjust it up until Interactive Brokers fixes this.
January 1st, 2025
I use rough numbers to save time and they don’t include fees. See the administrator’s statement for precise numbers. Here’s a link to my past letters.
I wrote my first letter to investors in 2019. It was brief but I included, “It will be hardest to beat the index during bubble markets.” I knew that fundamental investing during a bubble would be hard and I wanted to preemptively reassure investors for when that day came.
I believe we’re in a bubble market today,[1] with the S&P 500’s PE ratio reaching 1998 levels.
During a bubble it can be hard to predict what stocks will become hyped even if some are an obvious risk. You hit outliers that wouldn’t happen, or at least wouldn’t be so extreme, in a regular market.
On the short side, management of some troubled companies will attempt to lean into the AI hype. In most cases the market hasn’t been reacting to them much or at all. But in rare cases the market really believed. Take Lumen (LUMN) for example, an insolvent telecom company. Management told an absurd AI story and the stock went up 900%. Militia lost 1.75%, our largest loss of 2024. Why did the market believe so strongly on this one? I have no clue. I don’t think there is anything to learn from this, either. This is just a headwind that shorts face in a bubble.
On the long side, fundamentally sound companies don’t falsely hype AI nearly as often. And even when some do, a savvy shareholder base means that there are plenty of incremental sellers as the stock price goes up, preventing such a large spike.
Given that backdrop, I’m happy with this year’s result given that Militia was uncorrelated to the general market but narrowly beat it anyway. Had I taken on market risk this year - had our beta been one - we’d have won 55%. That’s how I was positioned back in 2019, which made it far easier to win a lot that year.
As a reminder, I dropped Militia’s beta because my edge in shorting over the years was so extreme and I wanted to lean into that strength. All I do is buy cheap investments and short bad companies, and with such a large edge in short selling I can run market neutral and get the free luxury of not having to worry about where the general market is going. I will maintain this positioning long term unless the market gets obviously cheap someday.
A few LPs, including friends, asked me what they should do with their index (S&P 500) investment. I told them to either leave it alone or trim it a little bit. I don’t have a good answer. I could be wrong about the S&P 500 being a bubble and they already have enough exposure to my ideas, anyway.
Part 1: Fourth Quarter
- Results for October 1st - December 31, 2024
- S&P 500: +2.5%
- Militia Capital: -.25%
- We made 8% on longs and lost 8% on shorts.
- Currently we’re 200% long and 125% short.
- October 25th Email to Partners
2024 Final Tax Estimate
Short term gains:
Net short term capital gains will be 10% of your cumulative Militia win.
Ordinary dividends will be 2% of your average NAV this year.
Margin interest and short dividend deductions will be 20% of your average NAV this year.
Total is an ~8% shield against short term capital gains.
Last year a couple people asked me how I'm doing this. I don't want to reveal all of my tax strategies. I'll just say that this juggle is really easy for me and there isn't anything controversial about this from a tax perspective.
I could push this even further and create a “loss” for your full tax basis to shield short term capital gains without affecting Militia's bottom line much[2]. However, every partner's tax situation is different and many partners would not benefit from such an extreme tax shield. I'm aiming for a healthy balance. If nothing else, this 8% shield can still offset the long term capital gains that I'm generating.
I should be able to maintain this long term. If someday we have lots of short term gains it would be for a very good reason.
Long term gains:
Net long term capital gains will be 20% of your cumulative Militia win.
Qualified dividends will be 2% of your average NAV this year.
Total is ~22% long term capital gains.
This is an estimate and varies by partner so you should leave a buffer.
PM Update
I ended the deal with Militia's second portfolio manager. His result wasn't that bad. He ended up winning around 10% while the S&P 500 was up 17%. I decided this on a combination of performance vs historic alpha standards, heavy concentration and his correlation to junk stocks being lower than hoped.
It wouldn't shock me if I quit him at the exact wrong time and he kills it in his personal account from here.
We parted ways on good terms, and I'm still going to pay him a $25k bonus out of my own pocket for having tried. He told me that he thinks that what I'm building here is incredible, and agreed that his style of investing probably isn't compatible with running other people's money.
- Multi-Strategy Update
Rodrigo continues to perform well, particularly since he did not have exposure to AI stocks on the long side and he has lots of exposure to global stocks, which have been weak. His full year annualized alpha to the S&P 500 was 15% and much higher to most other indexes. We’ve been uncorrelated.
Militia’s third PM was doing well until he got hit by an extreme outlier event on the short side a couple of weeks ago. A few funds that I respect - that have fantastic long term track records - got hit by the same event but our PM lost considerably less. So while he is down 10% since he started in September, I can understand why this happened beneath the surface. Backing out that one event, his alpha would have been strong. I’m still of course mostly focused on his bottom line alpha longer term, the key and simple variable.
- Broker Report
Part 2: Full Year
- 2024 Results
- S&P 500: +27%
- Militia Capital: +27%. Long +37%, Short -7%
Militia made 450 bets this year that won or lost more than .1%, our most diversified year. This sample size is bigger than many single managers will make in a decade. Over time I’ve gradually reduced Militia’s tail end (blow up) risk like I said I would back in 2022, and today it’s almost non-existent.
- Japan Move Finalized
My wife and I moved to Umeda in the Kita ward of Osaka.
Since moving here, I’ve found a couple of interesting new Japanese investments, and built some conviction in a couple that I had been following.
For example, my landlord will be the Sumitomo Realty Corporation (8830). It trades for 11x earnings, and earnings have been growing steadily. That’s a cheap stock on the numbers alone but qualitatively there is a lot to like, too. When comparing apartment rentals, their La Tour brand was especially attractive in terms of location and luxury. They only have a couple units available in Osaka and in all of central Tokyo they did not have a single high end unit available for rent in multiple buildings. So they’re near 100% occupied in their higher end properties. To me, that makes it screamin’ obvious that they have untapped pricing power. Not bad for high quality assets trading for just 11x earnings. A basket of investments like that, especially owned on borrowed yen, is going to kill the S&P 500 from here.
Thank you for continuing to trust in me. It has been life transforming and I do not take it for granted.
David Orr
[1] I define a bubble market as one with a near zero forward expected real return over the next decade. Expected is an important word. There is a wide range of outcomes and this is just my best guess of the average outcome.
[2] Accomplishing this would be such low effort for me that I could even make this a standalone business someday if there was enough interest. It could work for individuals and funds. I’d only want to do it through a single pooled vehicle. But most people seem to ignore taxes - oddly - so I won’t bother marketing it.
