S&P 5007,746.79-0.08%Nasdaq Composite26,515.93-0.34%Dow Jones Industrial53,955.79-0.04%Russell 20003,026.97+0.32%FTSE 10010,844.19-0.17%DAX26,391.42+0.26%Euro Stoxx 506,551.22+0.24%Nikkei 22566,970.22+2.08%Hang Seng25,652.82-0.06%Nifty 5024,471.70-0.46%Gold4,440.70+1.81%Silver65.02-0.13%Copper6.64+0.61%Crude oil WTI83.22+1.33%Crude oil Brent88.76+1.19%Natural gas2.76-1.36%VIX volatility15.34-0.78%UST 2Y4.25%UST 10Y4.72%UST 30Y5.25%

Calculator

What mixing two holdings actually did

A blend is not the average of its parts. Rebalancing mechanically sells whatever rose to buy whatever fell, and a mix usually varies less than the more volatile of the two — sometimes less than either. Whether it also returned more is decided by the history, so this runs it rather than asserting it.
10 instruments · five years of daily closes

Highest return

+150.3%

0% S&P 500 / 100% Gold

Least variable

28.1%

60/40 mix

Shallowest fall

-17.7%

20/80 mix, peak to trough

S&P 500 alone

+73.4%

Varying 36% a year

Gold alone

+150.3%

Varying 37% a year

Rebalanced

Monthly

In the worked example below

Two holdings, one mix

Choose the pair, the weight and the interval
$
%

The remaining 40% sits in Nasdaq Composite.

Rebalance back to those weights

Rebalancing sells whatever rose and buys whatever fell, to restore the weights. It costs nothing here; in reality it incurs dealing costs and may realise tax.

The blend returned

+76.0%

$100,000 became $175,988

Variability

39.8%

Annualised, from daily moves

Worst fall along the way

-28.7%

Peak to trough, within the window

Return per unit of variability

1.91

Crude, but the standard comparison
  • The blend
  • S&P 500 alone
  • Nasdaq Composite alone
S&P 500 alone+73.4%
Nasdaq Composite alone+78.9%
60/40 blend+76.0%
S&P 500 variability35.8%
Nasdaq Composite variability46.9%
Blend variability39.8%

Both holdings are measured over the trading days they share, so the comparison is like for like. A blend usually varies less than the more volatile of its two parts, and sometimes less than either — that is the one thing in investing that comes close to free, and it depends entirely on the two not falling together. Whether this particular mix beat holding one on its own is a fact about the window shown, not a property of the mix. Change the dates and the answer changes. Charges, dealing costs and tax are excluded.

Every mix of S&P 500 and Gold

Five years, rebalanced monthly
MixReturnVariabilityWorst fall
100 / 0+73.4%35.8%-24.8%
80 / 20+89.6%30.7%-21.9%
60 / 40+105.6%28.1%-19.0%
40 / 60+121.2%28.6%-18.0%
20 / 80+136.2%31.9%-17.7%
0 / 100+150.3%37.0%-23.3%

The mix with the highest return and the mix with the shallowest fall are rarely the same, and the one that looks best here is the one that suited the past five years. A different window gives a different answer, which is the honest limit of this exercise. How these markets moved relative to each other

Why a mix can vary less than its parts

If two holdings do not fall at the same time, the mix is steadier than either. When one drops and the other holds, the combined value moves less than the falling one did.

This is the one effect in investing that comes close to free. It depends entirely on the two behaving differently, and things that normally behave differently have a habit of falling together in a crisis.

What rebalancing does

Left alone, a mix drifts: whatever rises comes to dominate, and the blend slowly turns into the thing that went up. Rebalancing restores the weights by selling some of the winner.

That helps when holdings take turns and hurts when one trends strongly throughout. Set the interval to Never to see which happened here.

Reading the risk columns

Variability is the annualised standard deviation of daily moves. Worst fall is the deepest peak-to-trough decline inside the window, which is closer to what an investor actually experiences.

Return per unit of variability divides one by the other. It is crude — it treats upside and downside movement as equally bad — but it is the standard comparison.

What is excluded

Charges, dealing costs, spreads and tax are not modelled, and rebalancing incurs all of them in reality. Index figures are price levels rather than total return, so dividends are missing.

This is a record of what these prices did. It is not a recommendation, and it is not evidence about what any mix will do next.