# Holding period and the chance of a loss

> How often 1-, 5-, 10- and 20-year periods ended with a real loss, including a market that went nowhere for 13 years.

- Course: [Long-Term Investing: Index Funds, Risk and Your Own Behavior](https://productovi.com/courses/long-term-investing.md)
- Module: How much can you lose, and for how long?
- Lesson 7/32, ~22 min
- Source: https://productovi.com/courses/long-term-investing/holding-period

## How long is long enough?

**Grace:** Another March day at Zapas. The onboarding team asks you, as the PM, one question: what minimum horizon should the app ask for before it suggests a stock-heavy portfolio?

**Ethan:** Easy. I want a big line on the first screen: "Stay 10 years and you can't lose." People love certainty.

**Grace:** Let's test it instead of guessing. We take every possible start year in our 40 synthetic years and ask: if you held for 1, 5, 10 or 20 years, did you end with less buying power than you started with?

**Sofia:** So a 10-year window can start in year 1, 2, 3… up to year 31.

**Grace:** Right: 31 windows of 10 years, 36 of 5 years, 21 of 20 years. These are **rolling periods**. They show the whole range of luck a new user could have had, not just one lucky start.

> **Exercise:** With 40 years of data, how many rolling 10-year windows are there (start years 1, 2, 3 … as long as the window fits)?

## World vs one market

**Grace:** Here's the result for Tessaro World and, for comparison, Tessaro Home, one small domestic market. Each bar is the share of windows that ended with a real loss.

**Share of rolling windows with a real loss**

| Holding period, years | Tessaro World | Tessaro Home |
| --- | --- | --- |
| 1 | 30 | 42.5 |
| 5 | 19.4 | 25 |
| 10 | 6.5 | 35.5 |
| 20 | 0 | 38.1 |

*Real terms, after inflation; 40 synthetic years. Windows: 40 (1 year), 36 (5), 31 (10), 21 (20). Not a forecast.*

**Dataset (CSV):** [Rolling windows with a real loss, world vs home](https://productovi.com/datasets/long-term-investing/holding-period/rolling_real_loss.csv)

**Sofia:** For the world the bars shrink fast. For the home market they don't.

> **Exercise:** In the Tessaro history, what share of rolling 10-year windows of world stocks ended with a real loss?

**Ethan:** Six and a half percent over ten years. That's close enough to "can't lose", no?

**Grace:** Tell that to the users in those two windows. And five years, the horizon most of our beta users pick, looks much worse. Work out the five-year share yourself.

> **Exercise:** Of 36 rolling 5-year windows, 7 ended with a real loss for world stocks. What share is that, in percent, to one decimal?

**Sofia:** Almost one in five. And a five-year window can hold the whole crash from yesterday.

**Grace:** Exactly. Twenty-four months down and forty-eight back fit inside five years with almost nothing to spare. That's why the worst five-year window lost 42.2% in real terms.

## Thirteen flat years

**Ethan:** Why does the home market look so bad over 20 years? Its average is positive.

**Grace:** Because it went nowhere for a long time. From year 8 to year 20, Tessaro Home rose 41.0% in dollars, but after inflation it ended just 0.5% up. Thirteen years for nothing. World stocks gained 179.4% in real terms over the same years.

**Grace:** One market can stall for a decade or more. The long-run average still looks fine, so nobody notices until they live through it. As a result 8 of the 21 twenty-year windows of the home market lost buying power.

![A long horizon cut the chance of a real loss for world stocks from 30.0% to 0.0%, but did not stop a 42.2% fall within five years or save one home market (38.1% of 20-year windows lost).](https://productovi.com/courses/long-term-investing/diagrams/m2-l3-what-time-does.png)

*Time helps a broad world portfolio most.*

> **Exercise:** In the Tessaro history, holding for 20 years removed the risk of a real loss, whatever the investor held.

> **Exercise:** Match each Tessaro figure with what it describes.

**Ethan:** Our beta has users holding only the home fund. Should I worry about them?

**Grace:** Yes, and that's April's topic: why one fund should hold thousands of companies across many countries. For now, note that a long horizon doesn't fix a narrow portfolio.

## The onboarding question

**Ethan:** So what do I put on the first screen instead of "you can't lose"?

**Grace:** A question, not a promise: "When will you need this money?" Then show what that horizon meant in our history, both how often and how bad.

> **Exercise:** A user says they need the money in 3 years for a flat deposit. Which onboarding reply fits the data?

> **Exercise:** Rewrite Ethan's first-screen line "Stay 10 years and you can't lose" into an honest version in 2–3 sentences, using the rolling-window numbers.

> **Key ideas**
>
> - **Rolling periods** test every start year: 40 − h + 1 windows of h years in 40 years of data.
> - World stocks lost buying power in 30.0% of 1-year, 19.4% of 5-year, 6.5% of 10-year and 0.0% of 20-year windows in our synthetic history.
> - Five years is a minimum, not a safe zone: the worst 5-year window lost 42.2% in real terms.
> - One market can stall: Tessaro Home was flat in real terms for 13 years (years 8–20) and lost in 38.1% of 20-year windows.
> - Ask "when will you need this money?", show how often and how bad, and never promise that time removes risk.

> **Educational, not advice**
>
> This course explains how long-term investing works; it never tells you what to buy. Investments can fall, and you can get back less than you put in. Check the current rules where you live or ask a licensed professional.

**Grace:** Tomorrow we put all of March together: you'll write the message users see when the market is down 30%, and decide when it fires.

- Previous lesson: [Drawdown, recovery and time under water](https://productovi.com/courses/long-term-investing/drawdown-recovery.md)
- Next lesson: [Case: the "market is down 30%" message](https://productovi.com/courses/long-term-investing/case-down-30.md)

---
Productovi Learning. Free, no sign-up needed to read.