BlogSystems and backtests

Niftybees/Goldbees ratio switching system: 18% CAGR

One trading rule on the Niftybees to Goldbees ratio keeps you fully in whichever asset is leading. Here are the rules and the 18 year results.

Education only, not investment advice. Figures here come from past data and backtests: illustrative, not live market data and not a performance claim.

₹1 lakh invested in Niftybees in January 2008 is worth ₹4.43 lakh today: 8.45% CAGR over 18 years. That same ₹1 lakh, switched between Niftybees and Goldbees using one trading rule, became ₹20.84 lakh: 18% CAGR. Here are the strategy details.

Why equity, gold or a split falls short

Most people either go all in on equity and get destroyed in 2008, or sit in gold and miss the cyclical bull run, or split 50/50 and underperform both during regime shifts, because half their money is always in the wrong asset.

The standard advice is “diversification”. Hold 60 percent equity and 40 percent gold, and rebalance once a year. It sounds smart. It delivers 12.01 percent CAGR. Still not that great.

The deeper problem is that equity and gold do not move together. They trade leadership in multi-year regimes. 2008 to 2012 was gold. 2014 to 2018 was equity. 2020 to 2022 was equity again. 2024 to 2026 is gold. Holding both at all times means you are always half wrong.

The rule: one ratio, two assets

What you actually need is to be fully in whichever asset is winning right now, and to switch when the leadership changes. Calculate the ratio: Niftybees close divided by Goldbees close. Track its 20 day rolling high and low.

When the ratio breaks above its 20 day high, equity is winning: sell Goldbees and deploy full equity into Niftybees. When the ratio breaks below its 20 day low, gold is winning: sell Niftybees and deploy full equity into Goldbees.

Hold whatever you have between signals. Never be in cash after the first signal fires. That is the entire system. No earnings analysis. No macro view. No emotional decisions. The ratio itself tells you the regime.

Equity rallies on growth, earnings and rate cuts. Gold rallies on inflation, rupee weakness and fear. These triggers rarely fire at the same time, which is why their leadership regimes last for years, not days. The 20 day breakout is just a clean filter to identify when the regime has actually changed.

The results over 18 years

Over 18.34 years, from January 2008 to May 2026: CAGR 18.01 percent, and a final corpus of ₹20.84 lakh on ₹1 lakh of starting capital. Sharpe 1.05. Win rate 58 percent. Profit factor 3.72.

Max drawdown 24.95 percent. Compare that to Niftybees alone at minus 59.71 percent during 2008. The system pulls you out of equity before the worst of any crash.

It beats every passive alternative: Niftybees buy and hold, Goldbees buy and hold, and the standard 50/50 portfolio.

In every major regime, the system did the right thing. It sat in gold through 2008. It rode the equity bull from 2014. It rotated through COVID. It caught the recent gold rally from 2024.

Simple enough to ignore

Most people will never run this because it is too simple. They will keep searching for the next algorithm, the next pattern, the next guru. The edge in the market is not always hidden in complexity.

The edge in the market is not always hidden in complexity.

Sometimes it is sitting in plain sight, and the only thing protecting it is the fact that most people will not believe something this simple actually works.

I have shared this system in detail on Vijay Thakkar's podcast.

I first wrote this as a thread on X. Follow me there for new ones.

  • asset rotation
  • Niftybees
  • Goldbees
  • backtest
  • momentum

Kirubakaran Rajendran

Full-time algo trader, 18 years in the markets. Founder, Vecktor. I teach trading the way I work: write the rule, test it on data, then automate it.

More about me

Learn it step by step

Turn these ideas into rules you can test.