# Retiring in 8 to 10 years, going 80% S&P 500 / 8% QQQ / 6% SMH / 6% single stocks. Rate it. — retained record

> Retained reddit record from reddit:investing.

Canonical: https://highsignal.app/data/records/5d820fb41acb48a8

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:investing)  published 2026-10-11 15:36 ingested 2026-10-12 02:37

[original source ↗](https://www.reddit.com/r/investing/comments/1x3b96e/retiring_in_8_to_10_years_going_80_sp_500_8_qqq_6/)

## retained text



```
Planning to retire in 8 to 10 years. Household spends about 2.8% of liquid assets a year, so the math works on the index alone. I also need to leave a meaningful amount behind for a dependent who will outlive me by decades, so this money has a 40-year horizon, not a 10-year one.

Current plan, held mostly in IRAs with no tax cost to change:

* 80% S&P 500 index
* 8% QQQ
* 6% SMH
* 6% single names, mostly NVDA, plus a small SpaceX fund position
* Deferred comp (about 16% of the total) sits in bonds and is the only fixed income until I move the traditional IRAs toward bonds about three years before retirement
* Rule: rerun the plan every January; if the tilt trails the S&P over a trailing 10 years, cut it in half. No selling in a crash, decided in advance.

My thinking: the AI build-out is $800B+ a year of real capex and the index already owns the top of it; QQQ and SMH add the chip-making and memory layer the S&P under-owns. I know the 2000 to 2010 numbers for tech and semis, and I’m sizing the tilt so a repeat lengthens recovery but doesn’t break the retirement date.

How would you rate this? If bad, why. If you’d do better, how. Appreciate the guidance.
```



document fetched 2026-10-12 00:26 · canonical https://reddit.com/r/investing/comments/1x3b96e/retiring_in_8_to_10_years_going_80_sp_500_8_qqq_6
