1Me

My name is David Orr and I’m 39 years old. I have degrees in accounting and economics. I grew up in Oregon, spent a decade living in Thailand and now I’m living in Osaka, Japan so that I can study Japanese investments.

In Thailand I supported myself playing internet poker. By 2016 I had played 10 million hands and was a top 50 player in the world at No Limit Texas Hold’em.¹

I love strategy games, particularly ones of imperfect information, and have neared the top in a couple besides poker. Some examples: in the card game Hearthstone I reached rank 40 in North America in a player pool of millions. When I was 15 years old I got rich in the game Diablo 2 by market making in chat rooms. I play Civilization on the hardest difficulty. The stock market is the best strategy game that I’ve ever played.

I’m married and will start a family this year.

¹ My win rate was 10 big blinds/100 hands over a large sample at anonymous $10/$20 no limit 6 max.

2My Strategy

My strategy exploits two market anomalies. First, less volatile stocks have higher returns. And second, small stocks also have higher returns. Combined, an interesting pattern appears:

I’ve run 200% long and 100% short on average. Net exposure ranged from 0% to 150% and gross exposure from 150% to 400%.²In recent years I’ve run near zero beta but this fluctuates based on available opportunities and market conditions.

Assume:

• 200% long earns 30%/year.
• 100% short earns 5%/year.

That’s a 35%/year return, although in practice I’ve performed much better on the short side.

Limiting portfolio drawdowns is key when using leverage because large drawdowns force the portfolio to rebalance — to return to the original leverage target — which results in a permanent loss of capital. It’s alright if this happens once in a while but if this constantly happens the edge is degraded. Besides running with lower beta, these are the biggest ways that I mitigate drawdowns:

• Running a diversified portfolio with hundreds of positions. High concentration doesn’t mix with high gross leverage — too often a few big bets will randomly go wrong at the same time, causing a large drawdown. Thus, my portfolio is diversified with hundreds of positions. Longs are sized 1–10% while shorts are sized 0.1–2.5%. Most of these bets are sized on the smaller side.
• Monitoring correlation between positions. When correlation gets too high it’s far more likely that more positions will go wrong at once, causing a large drawdown. When this happens I can cut gross leverage or trim individual positions.

²Net exposure = total long − total short positions. Gross exposure = total long + total short positions.

3My Investing Process

I sort through a lotof investments across all regions and asset classes. When one stands out I make loose price targets for both good and bad outcomes and then I weigh probabilities. Ultimately, I’m guessing the expected present value of cash flow, the fair value of the investment. I’m not a “growth”, “quality” or “value” investor — I consider everything and boil ideas down to a key variable or two. For example, a long thesis might be as simple as, “Management is good at capital allocation and the business will continue to produce cash that they can reinvest.” I often make decisions within a few hours; any longer is a tell to skip it. Ideas should seem lopsided and obvious. I never use complex financial models — they produce garbage and waste time.

Longs: Decent companies 20% below fair value are interesting. So are great companies at fair value. The larger the discount, and the lower the downside risk, the bigger I bet. With a concentrated fund I’d demand larger discounts but it’s impossible to find enough ideas with so many holdings. I prefer companies with low debt, a visible growth runway and high return on capital employed.

Shorts: I look for catalysts such as government decisions, cash shortfalls, new competition, bankruptcy filings, delistings, technological advances, sentiment shifts and legal cases. Catalyst shorts are best so I bet bigger on them. Unfortunately, they’re hard to find and they only hit once — there are never enough of them. Thus, most shorts are simply failing businesses where liabilities exceed assets and future cash flow. The price target is usually $0. I see these companies as melting ice cubes.

I skim news, earnings reports, call transcripts and risk factors of companies that I follow. Each new piece of information is like a card revealed in a poker game — the odds change. I pay closer attention to larger, higher conviction positions. I quit or add to bets easily as the odds change, which I consider a strength.

4Reducing Risk by Backing Multiple Portfolio Managers

Militia Capital has two significant risks:

• Leverage risk. It’s possible that longs drop and shorts rise at the same time. If 200% long drops 20% while 100% short rises 20%, the fund would lose 60%. This shouldn’t happen but anything is possible. Similarly, I’m sometimes more than 100% net long, generally during a market panic. If the market continued to decline from −30% to −50%, the fund might perform badly.
• Short risk. Shorts have unlimited upside potential and bad companies sometimes spike hundreds of percent for seemingly no reason. For example, I was short Chesapeake Energy with a 0.5% position and it spiked 700%, causing a 3% loss since I was forced to reduce risk. I could have started with a larger position and/or the spike could have been larger.

Backing uncorrelated, high edge investors mitigates these risks. I had a team like this in poker and we never had a losing quarter.

I’ve read hundreds of small scale investors’ work. Most range from okay to terrible but I’ve found a couple of great ones who don’t run outside money yet. I asked them why. They told me they don’t want to deal with the business side of it. This seems like a value proposition — Militia Capital can back them to run a portfolio. I’m offering them ownership and wouldn’t take middle-man fees from my partners like a fund of funds. This also reduces costs compared to running many independent, small funds. Examples are lower margin interest rates, cheaper commissions and lower accounting/administration expenses.

By picking high edge portfolio managers who are uncorrelated — or even better, negatively correlated — Militia’s risk adjusted return greatly improves. Here is how much Militia’s Sharpe improves by increasing the number of uncorrelated 2 Sharpe strategies:³

Importantly, this doesn’t just reduce volatility superficially — it reduces the real risk that comes from running with high leverage and unhedged shorts.

Currently Militia is backing two portfolio managers besides myself.

³I use the equation Combined Sharpe = Sharpe × Square Root(number of uncorrelated strategies). There is some academic debate over what the exact equation is but this is a close enough approximation.

5FAQ

— David Orr