Before you pitch it: what are you really long, and can you get out?
Two questions a pod PM asks before anything else: how much of this stock is the factor you just hedged, and how many days does it take to leave. Both change the size, not the thesis.
- Count the days it takes to exit a position at a set share of volume
- Find the largest position a days-to-exit limit allows
- Split a stock's variance into the factor and the part that is the company
- Explain why a crowded name deserves a smaller size even when the thesis is right
The intuition
Two things make a pitch riskier than its numbers say. The first is that the stock may mostly be something else: a chip designer whose price is 70% the semiconductor index is a bet on the index with a small company attached, and once the index is hedged away the position is far smaller than it looks. The second is the exit. If a dozen pods own the same name and one of them is forced to cut, all of them sell the same stock on the same day, and the price falls for reasons that have nothing to do with the thesis. The measure of that is how many days of normal volume it takes to get out.
Factor share is the fraction of a stock's variance explained by a factor: (beta × the factor's volatility)² over the stock's volatility². The rest is idiosyncratic, the part the platform is paying for. Days to exit is the position divided by what you can sell each day: the daily value traded times the share of it you are willing to be. A desk sets a limit in days, and that limit, times the daily value traded, is the biggest position you may hold.
Days to exit = position ÷ (daily value traded × participation). Largest position = days limit × daily value traded × participation. Factor share of variance = (beta × factor vol)² ÷ stock vol²; idiosyncratic volatility is the square root of what is left. High factor share: hedge it, and expect the position to shrink. Many days to exit: size it smaller than the thesis would like.
Why it works
- The conventions here: one factor; variances add, so the factor's share is a ratio of variances, not of volatilities. Days to exit assumes normal volume and a fixed participation rate, usually 10% to 25% of the day.
- Idiosyncratic risk is the product a pod sells. The platform allocates capital to pods for returns it cannot get from an index. A position whose variance is 70% the sector is, after the hedge, a 30% position. Say what the idiosyncratic volatility is; that is the size of the bet you are really making.
- Crowding is a risk you cannot see in the P&L. Hedge-fund ownership, short interest, and how the stock behaved in the last deleveraging are the tells. A crowded name can keep working for a long time; it just deserves less size than the same idea would get if nobody else had found it.
- Days to exit is the honest liquidity number. Not the market cap, not the average volume, but how long the fund itself needs at a participation rate that does not move the price. On the day the thesis breaks, volume halves and everyone is selling; the number that matters is what days to exit looks like then.
- The limit is a size, not a veto. A great idea that takes three weeks to leave is expressed smaller, or through a more liquid proxy for the same thesis.
- The PM's two questions come first, before the pitch is heard in full. If you cannot get out and it is mostly the factor, the rest is academic.
| Sellable a day: 16 × 25% | $4M |
| Days to exit: 40 ÷ 4 | 10 days, twice the limit |
| Largest position allowed: 5 × 4 | $20M |
| Factor-driven volatility: 1.2 × 20% | 24% |
| Factor share of variance: 24² ÷ 40² | 36% |
| Idiosyncratic volatility: √(40² − 24²) | 32% |
The position has to halve for liquidity alone. About a third of the stock's variance is the factor; hedged, the bet is a 32%-volatility company, not a 40%-volatility stock.
The formulas
Position over what you can sell each day.
The days rule, run backwards.
How much of the stock is the factor.
What is left once the factor is hedged: the bet you are really making.
The size of that bet.
Worked example
What you can sell each day, then the position divided by it. The follow-up says whether the answer is a problem, which depends on the number, not the thesis.
See it move
Same company. Change the position, the value traded a day, the share of each day's volume the fund will be, and the desk's limit in days. "New company and numbers" changes the factor split; the sliders do not touch it.
- Raise the position. Days to exit rise; the largest position allowed does not move, because it depends on the stock and the desk, not on you.
- Raise the value traded a day. Days to exit fall and the largest position allowed rises.
- Raise the share of the day you will be. Days fall and the largest position rises, and the curve shows the price you pay: past 20% to 25% of the day you are the market.
- Change the desk's limit. Only the largest position moves; days to exit are a fact about the stock.
Run it backwards
The desk's rule, run backwards: given the limit in days and what you can sell a day, what is the biggest position you may hold, and is the one you pitched inside it?
Days to exit = position ÷ sellable per day, so the largest position = days limit × sellable per day. The last line compares it with what you asked for.
If you are over, the answer is not to argue the limit. It is a smaller size, or the same thesis through a more liquid name.
Traps
Say it in the interview
“What's the crowding like, and how long would it take you to get out?”
Check yourself
4 fresh questions, with new numbers. Answer each one correctly to finish the chapter. Get one wrong and you will see the full working, then you can try it again with new numbers.
Answers within 1% are marked right. Type the number; $, %, x and M are fine. First tries count toward Learned: the topic is Learned once every chapter is done and 75% of first tries were right.
- Days to exit = position ÷ (daily value traded × participation); largest position = days limit × sellable per day.
- Factor share = (beta × factor vol)² ÷ stock vol²; what is left is the bet.
- Variances add; volatilities do not.
- Crowding and liquidity change the size, not the thesis.