What did GSG do before 2006?

Before Aug 2006 the series is ^SPGSCI's. Those 21.5 years (Feb 1985–Jul 2006) show 4.5% a year with 18.3% annualized volatility and a worst drawdown of -51.6% (Sep 1990–Feb 1999). In major S&P 500 declines: 1987 crash, -1.2% versus -29.8% for the S&P 500 (^SPGSCI); 1998 LTCM sell-off, -11.8% versus -15.3% for the S&P 500 (^SPGSCI); 2000–02 dot-com bear, -6.9% versus -44.7% for the S&P 500 (^SPGSCI). The full extended history's worst drawdown is -88.7% (Jun 2008–Apr 2020).

How closely does ^SPGSCI track GSG?

Over 242 overlapping months (Aug 2006–Sep 2026), ^SPGSCI had a correlation of 0.978 with GSG, a tracking error of 4.8% a year and a tracking difference of +3.66% a year (grade: approximate).

How far back can you test GSG? To Feb 1985, with ^SPGSCI

Computed by the MarketHeist portfolio engine · Published 2026-10-10 · Updated 2026-10-10 · Data through Sep 2026

GSG (Commodities (S&P GSCI)) has its own monthly data from Aug 2006. Spliced with ^SPGSCI it reaches back to Feb 1985, which adds 21.5 years (52% of the extended history comes from proxies). Data through Sep 2026.

What you'd have missed

GSG's own data starts in Aug 2006; ^SPGSCI extends it to Feb 1985, 21.5 more years.
Its own history misses the 1987 crash, the 1998 LTCM sell-off and the 2000–02 dot-com bear.
Its worst drawdown, -88.7% (Jun 2008–Apr 2020), is already inside the ETF's own history; the extension does not change the worst case.
Over the 242 overlapping months ^SPGSCI returned 3.7% a year more than GSG, so the proxy years likely overstate what GSG would have earned by a similar margin.
DBC is extended with the same ^SPGSCI.

How the history is built

GSG (ETF) Aug 2006 → now

^SPGSCI (S&P GSCI commodity index, index) used Feb 1985 → Jul 2006; its own data runs Feb 1985 → now

Feb 1985Sep 2026

Solid bars are the months each fund contributes to the spliced series; the outline shows how far a proxy's own data runs. Each month uses the youngest fund that has data for it; returns are spliced, not prices (dividends reinvested via adjusted close).

How well each proxy tracked GSG

Proxy	Kind	Months contributed	Overlap with GSG	Correlation	Tracking error	Tracking difference	Beta	Vol ratio	Grade
^SPGSCIS&P GSCI commodity index	index	Feb 1985–Jul 2006258 months	Aug 2006–Sep 2026242 months	0.978	4.85%	+3.66%	0.96	0.98	Approximate

Statistics are over the months a proxy and GSG both have data. Tracking error is annualized; tracking difference is the annualized geometric gap (positive: the proxy earned more). Page grade: approximate, the worst grade among the proxies that contribute months.

GSG's own history versus the extended history

	The ETF's own history	Proxy years only	Full extended history
Window	Aug 2006 – Sep 202620.2 years	Feb 1985 – Jul 200621.5 years	Feb 1985 – Sep 202641.7 years
Growth of $10,000	$7,009	$25,712	$18,021
CAGR	-1.7%	4.5%	1.4%
Annualized volatility	23.2%	18.3%	20.8%
Sharpe ratio (risk-free 0%)	0.04	0.33	0.17
Max drawdown	-88.7%Jun 2008 → Apr 2020; not yet recovered	-51.6%Sep 1990 → Feb 1999; recovered Feb 2003	-88.7%Jun 2008 → Apr 2020; not yet recovered
Worst complete calendar year	2008 (-45.8%)	2001 (-31.5%)	2008 (-45.8%)
Best complete calendar year	2021 (38.8%)	1999 (46.2%)	1999 (46.2%)
Longest underwater (months)	219	148	219
Major declines covered	3	3	6

The columns cover different years, so a higher or lower CAGR is not a verdict on the fund; the point is which market regimes each window includes. Growth of $10,000 is the same one-asset, buy-and-hold calculation the live portfolio app makes.

Growth of $10,000

Log scale, so the early decades are not flattened. Grey is the proxy period; blue is GSG's own data. Dashed lines mark where each fund's returns begin.

Drawdowns

Decline from the prior peak at month-end, the "underwater" curve.

How GSG did in each major US stock-market decline since 1985

Decline	S&P 500	GSG over the same months	Source of those months
1987 crashSep 1987–Nov 1987	-29.8%	-1.2%	^SPGSCI
1998 LTCM sell-offJul 1998–Aug 1998	-15.3%	-11.8%	^SPGSCI
2000–02 dot-com bearSep 2000–Sep 2002	-44.7%	-6.9%	^SPGSCI
2007–09 financial crisisNov 2007–Feb 2009	-50.8%	-51.8%	ETF
2020 COVID crashJan 2020–Mar 2020	-19.4%	-42.6%	ETF
2022 inflation bearJan 2022–Sep 2022	-23.9%	20.3%	ETF

The decline windows are computed, not chosen: every peak-to-trough fall of at least 15% (month-end) in the S&P 500 series since 1985. The fund's figure compounds its monthly returns from the month after the S&P 500's peak through the trough.

GSG's own five deepest drawdowns (extended history)

Peak	Trough	Depth	Recovery	Source of the decline months
Jun 2008	Apr 2020	-88.7%	not yet recovered	ETF
Sep 1990	Feb 1999	-51.6%	Feb 2003	^SPGSCI
Nov 1985	Jul 1986	-25.4%	Dec 1989	^SPGSCI
Jul 2006	Jan 2007	-23.1%	Dec 2007	ETF
Feb 2003	Apr 2003	-22.6%	Feb 2004	^SPGSCI

This covers declines the equity windows miss, such as the bond bear markets of 1994 and 2022.

Caveats

^SPGSCI is an index level, not a fund: it has no fees and is not directly investable. The measured gap to GSG while both existed is +3.66% a year.
Monthly month-end data: intra-month drawdowns were deeper than shown.
Dividends are reinvested; no taxes or trading costs are modeled.
Our data source's series for ^SPGSCI begins Feb 1985; the fund itself may be older.

Test it yourself

Open GSG as a one-asset portfolio in the live app, where every number recomputes from fresh data.

Open GSG (100%) in the portfolio builder →

Used in portfolios: Global Asset Allocation

Used in guides: Global Asset Allocation

All fund histories →

Frequently asked questions

How far back does GSG data go?: GSG (Commodities (S&P GSCI)) has its own monthly data from Aug 2006. Spliced with ^SPGSCI it reaches back to Feb 1985, which adds 21.5 years (52% of the extended history comes from proxies). Data through Sep 2026.
What did GSG do before 2006?: Before Aug 2006 the series is ^SPGSCI's. Those 21.5 years (Feb 1985–Jul 2006) show 4.5% a year with 18.3% annualized volatility and a worst drawdown of -51.6% (Sep 1990–Feb 1999). In major S&P 500 declines: 1987 crash, -1.2% versus -29.8% for the S&P 500 (^SPGSCI); 1998 LTCM sell-off, -11.8% versus -15.3% for the S&P 500 (^SPGSCI); 2000–02 dot-com bear, -6.9% versus -44.7% for the S&P 500 (^SPGSCI). The full extended history's worst drawdown is -88.7% (Jun 2008–Apr 2020).
How closely does ^SPGSCI track GSG?: Over 242 overlapping months (Aug 2006–Sep 2026), ^SPGSCI had a correlation of 0.978 with GSG, a tracking error of 4.8% a year and a tracking difference of +3.66% a year (grade: approximate).

Methodology

Source: Yahoo Finance via MarketHeist, monthly adjusted close (dividends reinvested). Months run to the last complete calendar month.
Splice rule: Each month uses the youngest fund that has data for it; returns are spliced, not prices (dividends reinvested via adjusted close).
Numbers: Computed by the MarketHeist portfolio engine as a one-asset, 100% buy-and-hold portfolio, the same calculation the live app makes. Sharpe uses a risk-free rate of 0%. Drawdowns are month-end.
Correlation: Pearson correlation of monthly returns over the months both funds have.
Tracking error: Standard deviation of the monthly return differences (proxy minus GSG) times the square root of 12.
Tracking difference: Annualized geometric return of the proxy minus that of GSG over the overlap. Positive means the proxy earned more.
Beta and vol ratio: Beta is the covariance of the proxy with GSG divided by the variance of GSG; vol ratio is the proxy's standard deviation divided by GSG's.
Grades: Close: correlation of at least 0.95 and tracking error of at most 3%. Approximate: tracking error of at most 8% and either a cash-like fund (volatility of at most 3%) or correlation of at least 0.80 with a vol ratio of at most 1.25. Category stand-in: anything weaker.
Generated: 2026-10-10; data through Sep 2026. Reproduce hash of the full extended one-asset portfolio: eb0fcc91cfe6b0d4

Historical simulation, not investment advice. Past returns, including spliced proxy returns, do not predict future results.