CS2 skins vs S&P 500 vs Bitcoin: a 6-year backtest (2020 to 2026)
STI 30 returned +54.5% over 6 years. S&P 500 ran +154.4%, bitcoin compounded roughly 10×. The honest cross-asset comparison and where skins fit in a modern portfolio.
Update note (April 2026): The returns cited in this post reflect the original STI backtest (pre-fix April/24). Coverage corrections applied that day included ultra-rare blue chips (Dragon Lore, Howl, Fire Serpent) that were absent from the eligible universe. With the correction, STI 30 went to +78.9%, STI 100 to +88.9% (STI 500 stayed at +61.8%, STI Cases is now a separate fourth index at +763.9% since March 2021). The structural insight stands: STI 100 > STI 500 > STI 30 in performance order, negative Sharpe vs CDI, high transaction costs. Live values at /en/indices.
The methodology page covers what the STI index family tracks and how it's built. Now the question every investor asks before allocating a cent into a new asset:
Did it return more than just sitting on cash?
To answer honestly, I needed three different anchors:
- S&P 500, because skins are USD-denominated and international investors compare everything against it.
- Bitcoin, the most liquid alternative asset on the planet and the best proxy for "risk appetite" in recent years.
- CDI (Brazil's risk-free rate, set near the Selic policy rate — analogous to US Treasury T-Bills) included for emerging-market context. Brazil ran historically high real rates during the period; readers focused purely on US/EU benchmarks can skim past CDI rows in the table below.
The numbers, no theatrics
Period: April 1, 2020 to April 1, 2026 (exactly 6 years). All values are returns in US dollars before inflation (skins are priced in USD, CDI converted using Brazil's official exchange rate).
| Asset | Total return | CAGR | Volatility |
|---|---|---|---|
| Bitcoin | ~+990% | ~50%/yr | ~60% |
| S&P 500 | +154.4% | ~17%/yr | ~18% |
| CDI | +77.3% | ~10%/yr | ~0% |
| STI 100 | +72.7% | +9.5%/yr | 26.3% |
| STI 500 | +60.4% | +8.2%/yr | 24.7% |
| STI 30 | +54.5% | +7.5%/yr | 30.2% |
First surprise: cumulative CDI of 77.3% beat all three STIs in the period. Someone who simply bought Brazilian Treasury Selic bonds in April 2020 and never looked again outperformed the broad skin market. That's the real weight of the Brazilian high-rate window (Selic peaked at 13.75% in 2023).
Second: S&P 500 doubled CDI's return in dollars. That's the "exceptional US equity decade" — the American economy emerged from the pandemic with fiscal expansion + AI innovation + global dollarization, and returns reflect it.
Three stories in one chart
2020-2021 — The cheap money era
Fed at 0%, Selic at 2%, QE flowing. Risk assets flew together: S&P 500 up ~40%, bitcoin from $6k to $69k, and skins had their first historical peak. Skins moved hand-in-hand with S&P 500 in this period — important insight: they're not an "isolated" asset from the global market.
2021-2022 — When the music stopped
In April 2021, STI 30 began its biggest fall (-30.8% through December). Not a coincidence: that was exactly when the US Federal Reserve started signaling it would slow money printing and raise interest rates. Risky assets dropped across the board. CDI started looking competitive for the first time in years.
2022-2024 — Liquidity winter
Selic at the top (13.75%), Fed up to 5.5%. Skins and S&P 500 stagnated, BTC fell 70% then started recovering. CDI captured the period's best risk-adjusted return. This was the time when "sitting on cash" paid.
2024-2026 — The reopening
The Fed started cutting rates again. S&P 500 hit new highs. Bitcoin reached new all-time highs after the SEC approved BTC ETFs (regulated funds anyone can buy). STI 30 partially recovered, but did not match S&P 500 or BTC in speed — the skin market reacts more slowly to big economic shifts.
Adjusting for risk: skins look worse
Raw return is half the story. When you divide return by how bumpy the ride was (this is called the Sharpe ratio), the ranking gets even worse for risky assets in a period when interest rates were high:
| Asset | Sharpe vs CDI (6 years) |
|---|---|
| CDI | 0 (benchmark) |
| S&P 500 | ~+0.40 |
| Bitcoin | ~+0.65 |
| STI 100 | -0.20 |
| STI 30 | -0.24 |
| STI 500 | -0.27 |
All three STIs ended with a negative Sharpe — once you factor in the volatility, the broad skin market paid LESS than Brazil's risk-free rate did with zero ups and downs. S&P 500 and bitcoin both beat that bar, but both required stomach: S&P 500 had two 20%+ drops, bitcoin fell 70% in 2022.
Where skins fit
The most honest conclusion isn't "skins are bad." It's "skins have a specific profile":
- Positive nominal returns in dollars. Beat cumulative inflation in the period and preserved purchasing power.
- High correlation with US equities and crypto. Don't serve as a "hedge" against traditional markets — when S&P 500 falls, skins tend to fall together.
- Volatility similar to large-cap crypto, but with much lower liquidity. Selling an AWP Dragon Lore takes days or weeks; selling an S&P 500 ETF takes seconds.
- High transaction costs: Steam fee 15% (5% Steam + 10% Valve/CS2 publisher), 15% capital gains tax in BR, bid/ask spread frequently 2-5%. Third-party marketplaces (CSFloat ~2%, Skinport ~12%) reduce friction.
STI 30 (+54.5% in 6 years) fell short of CDI (+77.3%) and well below S&P 500 (+154.4%). It's not a replacement for Brazilian bonds or US stocks. It's an alternative asset that partially follows the global risky-stuff market — gained value in absolute terms, but didn't pay enough for the risk taken over the past 6 years.
This doesn't mean skins are a bad bet — even broad stock markets had stretches where they paid less than the risk justified. It means the next 6-year window could look very different, and having STI as a measurement tool lets you answer "is the skin market as a whole doing better or worse than last month?" with a number, not gut feel.
Next: the exceptions
In the next post, I move away from the average and focus on a group of 9 skins that returned +377% in the same period — counterintuitively almost 7× STI 30. The explanation isn't curation or luck: it's pure supply-demand economics. Contrabands, ended operations, maps removed from the competitive pool. Specific stories that explain outliers in the skin market.
Cite this post
Research, journalism, or blog use is welcome with attribution. Pick a format below.
Fernandes, J. (2026). CS2 skins vs S&P 500 vs Bitcoin: a 6-year backtest (2020-2026). Skin Trackers. https://skintrackers.com/en/blog/skins-vs-cdi-vs-bitcoin-2020-2026
Jorgin Fernandes, "CS2 skins vs S&P 500 vs Bitcoin: a 6-year backtest (2020-2026)," Skin Trackers, 2026, https://skintrackers.com/en/blog/skins-vs-cdi-vs-bitcoin-2020-2026.
@misc{skintrackers_skins-vs-cdi-vs-bitcoin-2020-2026,
author = {Jorgin Fernandes},
title = {CS2 skins vs S&P 500 vs Bitcoin: a 6-year backtest (2020-2026)},
year = {2026},
url = {https://skintrackers.com/en/blog/skins-vs-cdi-vs-bitcoin-2020-2026},
publisher = {Skin Trackers}
}https://skintrackers.com/en/blog/skins-vs-cdi-vs-bitcoin-2020-2026
For the full post with inline tooltips for every financial term and Brazilian-specific context, see the Portuguese original.