Three years ago, in May 2023, the S&P 500 was clawing back from a regional banking scare and trading near 4,190. Today it sits near 7,408 after a 2025 fueled by AI cap-ex and an Iran-war oil shock that lifted Energy to the top of the sector league. This article distills that journey into a single reference: index returns, the eleven S&P sectors, the Magnificent 7 dispersion that's quietly broken in 2026, and the small-cap rotation that finally arrived.
How to read this article. All return figures are price returns unless explicitly labeled total return (TR). 1-year window = May 19, 2025 → May 19, 2026. 3-year window = May 19, 2023 → May 19, 2026. Sector figures use SPDR sector ETFs (XLK, XLF, etc.) as proxies for S&P 500 GICS sector indices. Cumulative multi-year returns compound the published annual figures and are approximate.
| Index | Ticker | Level (May 15, 2026) | YTD 2026 | 1-Year | 3-Year |
|---|---|---|---|---|---|
| S&P 500 | SPX | 7,408.50 | +0.84% | +25% | +77% |
| Nasdaq 100 | NDX | ~26,150 | +3.5% | +38% | +135% |
| Dow Jones Industrial Avg. | DJIA | ~46,750 | +0.02% | +18% | +55% |
| Russell 2000 | RUT | ~2,470 | +11.1% | +22% | +38% |
| S&P 400 Mid-Cap | MID | ~3,510 | +4.2% | +19% | +44% |
Levels and percentages compiled from FRED, Yahoo Finance and S&P Dow Jones Indices. Multi-year returns approximate.
Small-caps spent four years lagging because they were the most rate-sensitive segment of the market and the most exposed to a domestic recession that never came. In 2026 the script flipped: trade policy whiplash and Iran-war tail risks hit multinational large-caps harder than domestic small-caps, and any whiff of a steeper yield curve still helps regional banks — a big Russell 2000 weight. Watch IWM relative to SPY: a sustained ratio break above 0.18 (currently 0.165) is the technical signal that rotation has legs.
Eleven S&P 500 sectors, ranked by trailing 12-month return. Tech still wins — but for the first time in a long time, Energy and Industrials are at the top with it. Defensives (Staples, Utilities, Real Estate) lag.
| # | Sector | ETF | 1Y Return | Headline Driver |
|---|---|---|---|---|
| 1 | Energy | XLE | +53% | Hormuz disruption, Brent above $105 for 8 months |
| 2 | Technology | XLK | +49% | AI cap-ex still accelerating; Nvidia, AVGO leadership |
| 3 | Communication Services | XLC | +34% | Alphabet +28% YTD alone, Meta ad revenue strength |
| 4 | Industrials | XLI | +30% | Reshoring, defense spend, infrastructure cap-ex |
| 5 | Financials | XLF | +27% | Steeper curve, no recession, capital-markets revival |
| 6 | Materials | XLB | +19% | Copper at multi-year highs, tungsten/rare-earth squeeze |
| 7 | Utilities | XLU | +16% | Data-center power demand bid; rate-cut hopes deflated |
| 8 | Healthcare | XLV | +11% | GLP-1 maturity, pharma policy overhang |
| 9 | Consumer Discretionary | XLY | +9% | Tesla drag, tariff pass-through worries |
| 10 | Consumer Staples | XLP | +6% | Defensive bid, but no yield tailwind |
| 11 | Real Estate | XLRE | +3% | 30Y yield at 5.13% crushed cap-rate compression |
Three years compresses three regimes into one chart: the 2023 AI awakening (Tech +56%), the 2024 melt-up (Communication +40%, Financials +33%, Discretionary +32%), and the 2025 oil-and-AI fusion (Energy +66%, Tech +58%). Compound them and you get a sector league with very wide tails.
| # | Sector | ETF | 2023 | 2024 | 2025 | 3-Year Cum. |
|---|---|---|---|---|---|---|
| 1 | Technology | XLK | +56% | +36% | +58% | +220% |
| 2 | Communication Services | XLC | +54% | +40% | +40% | +200% |
| 3 | Financials | XLF | +12% | +33% | +35% | +101% |
| 4 | Consumer Discretionary | XLY | +41% | +32% | +12% | +108% |
| 5 | Industrials | XLI | +17% | +19% | +28% | +78% |
| 6 | Energy | XLE | -2% | +6% | +66% | +72% |
| 7 | Materials | XLB | +12% | 0% | +24% | +39% |
| 8 | Utilities | XLU | -7% | +23% | +19% | +36% |
| 9 | Consumer Staples | XLP | 0% | +15% | +10% | +26% |
| 10 | Healthcare | XLV | +0.3% | +3% | +8% | +11% |
| 11 | Real Estate | XLRE | +12% | +5% | -4% | +13% |
What stands out: Tech and Communication produced almost 10× the cumulative return of Healthcare over three years. Two sectors built the bull market; the rest came along for the ride. Energy almost matched the S&P 500 over three years despite being down in 2023 and barely positive in 2024 — the entire move came in nine months of 2025.
The Magnificent 7 returned +76% in 2023 and +156% cumulative through end-2024 while the rest of the S&P 500 produced low-double-digit returns. That trade is no longer one trade. Through mid-May 2026 the basket is up only ~9% as a group, but the spread between the best and worst name is now 44 percentage points.
| Stock | Ticker | YTD 2026 | Story |
|---|---|---|---|
| Alphabet | GOOGL | +28% | Gemini monetization + search ad strength; cloud cap-ex paying off |
| Meta Platforms | META | -6% | 1-month -12% on cap-ex worry; ad-revenue cycle wobble |
| Nvidia | NVDA | +12% | Still the AI tax collector, but multiple compressing |
| Amazon | AMZN | +6% | AWS margin pressure from custom silicon competitors |
| Apple | AAPL | +2% | Services flat, iPhone cycle muted, China headwinds |
| Tesla | TSLA | -9% | Delivery growth stalled, robotaxi narrative slipping |
| Microsoft | MSFT | -16% | Azure AI cap-ex digestion + Q1 guidance miss |
The tables above use S&P 500 sector ETFs as proxies. For our own readers who execute via Tickmill MT5 CFDs, here is the same picture but at the single-stock level, sliced from the live Tickmill US CFD universe (484 instruments, 422 US with 1-year data). The single-stock distribution is far wider than the sector averages suggest — the Technology median is +20.8% over 1 year, but the mean is +79.8% because a handful of memory/storage/semi names compounded into the triple digits.
| # | Ticker | Name | Sector | 1Y % | Spread (bps) |
|---|---|---|---|---|---|
| 1 | LITE | Lumentum Holdings | Technology | +1,031% | 16.2 |
| 2 | WDC | Western Digital | Technology | +809% | 11.6 |
| 3 | MU | Micron Technology | Technology | +630% | 6.5 |
| 4 | VSAT | Viasat | Technology | +539% | 34.0 |
| 5 | IRWD | Ironwood Pharmaceuticals | Healthcare | +528% | 54.5 |
| 6 | TTMI | TTM Technologies | Technology | +452% | 17.8 |
| 7 | VIAV | Viavi Solutions | Technology | +425% | 8.2 |
| 8 | INTC | Intel | Technology | +408% | 2.7 |
| 9 | ARWR | Arrowhead Pharmaceuticals | Healthcare | +364% | 24.6 |
| 10 | PLUG | Plug Power | Industrials | +334% | 59.7 |
| 11 | FORM | FormFactor | Technology | +267% | 20.3 |
| 12 | AMD | Advanced Micro Devices | Technology | +256% | 3.3 |
| 13 | UCTT | Ultra Clean Holdings | Technology | +250% | 16.8 |
| 14 | AMKR | Amkor Technology | Technology | +233% | 15.2 |
| 15 | SMTC | Semtech | Technology | +230% | 21.7 |
12 of the top 15 are Technology — and within Tech, 9 are memory / storage / optical / fab-tool semis. This is the single cleanest signature of the AI cap-ex super-cycle: not the headline GPU names, but the picks-and-shovels supply chain that feeds them.
| # | Ticker | Name | Sector | 1Y % | Spread (bps) |
|---|---|---|---|---|---|
| 1 | RPD | Rapid7 | Technology | -71.3% | 29.4 |
| 2 | LBRDK | Liberty Broadband | Communication Svcs | -67.4% | 6.1 |
| 3 | FMC | FMC Corporation | Basic Materials | -67.3% | 23.4 |
| 4 | QDEL | QuidelOrtho | Healthcare | -66.8% | 17.9 |
| 5 | ASAN | Asana | Technology | -63.9% | 30.5 |
| 6 | HAIN | The Hain Celestial Group | Consumer Defensive | -62.9% | 137.9 |
| 7 | WING | Wingstop | Consumer Cyclical | -60.0% | 55.2 |
| 8 | CWH | Camping World Holdings | Consumer Cyclical | -59.1% | 29.1 |
| 9 | MSTR | MicroStrategy | Technology | -58.4% | 9.6 |
| 10 | JACK | Jack in the Box | Consumer Cyclical | -57.6% | 44.8 |
| 11 | PRGS | Progress Software | Technology | -54.4% | 20.9 |
| 12 | SRPT | Sarepta Therapeutics | Healthcare | -53.9% | 23.9 |
| 13 | CHWY | Chewy | Consumer Cyclical | -52.1% | 10.0 |
| 14 | GTLB | GitLab | Technology | -51.6% | 12.0 |
| 15 | PLAY | Dave & Buster's | Communication Svcs | -51.0% | 28.3 |
5 of the 15 worst are Consumer Cyclical / Defensive — the consumer is the marginal weakness in this market. Tech makes the list too (Asana, GitLab, Rapid7, Progress, MicroStrategy), confirming that owning XLK is not the same as owning every tech name — within the sector, dispersion is enormous.
| Sector | n | Mean 1Y % | Median 1Y % |
|---|---|---|---|
| Technology | 99 | +79.8% | +20.8% |
| Basic Materials | 16 | +54.0% | +46.2% |
| Energy | 15 | +52.8% | +46.9% |
| Healthcare | 67 | +35.3% | +16.8% |
| Industrials | 46 | +26.9% | +14.4% |
| Financial Services | 42 | +11.9% | +11.2% |
| Utilities | 6 | +9.9% | +20.6% |
| Communication Services | 26 | +9.2% | -5.6% |
| Real Estate | 19 | +3.5% | -3.0% |
| Consumer Cyclical | 62 | -0.4% | -2.8% |
| Consumer Defensive | 24 | -5.1% | +4.1% |
The mean-median spread is the dispersion story. Tech is +79.8% mean / +20.8% median — a typical tech name returned ~20% but the basket return is dragged up to ~80% by a handful of triple-digit winners. Communication Services is the opposite: mean +9.2% but median -5.6% (Alphabet +133% pulls the mean up alone). The XLK ETF is between these two — closer to the mean than the median because it is market-cap weighted.
Practical Tickmill spread observation. Mega-cap US CFD spreads are under 3 bps across the board (NVDA 0.9 bps, AMZN 0.77 bps, AAPL 1.01 bps, MSFT 1.91 bps, GOOGL 2.04 bps, META 2.32 bps). Even on the active small/mid names, spreads typically sit at 10-30 bps. That means a 60-90 day mega-cap hold pays roughly 1.5-2.0 bps round-trip — small enough that single-name alpha edges above 1% per quarter are still worth executing.
Hyperscaler capital spending tripled between 2023 and 2025. Nvidia's data-center revenue went from $15B annual to $200B+ run-rate. That single line item is why XLK is up 220% in three years. The 2026 question is who pays — hyperscalers booking these costs as multi-year amortization vs. genuine ROI on AI workloads. Microsoft's -16% YTD is the first public read on that question.
The Strait of Hormuz event-window beginning Q2 2025 took Brent from ~$72 to a sustained $100+ regime. Energy (XLE) was down in 2023 and flat in 2024, then ripped +66% in 2025. Defense names (LMT, RTX, NOC) inside XLI added another leg. As of May 2026 Hormuz is still closed in spot terms even after the Beijing off-ramp; the Energy bid has moderated but not unwound.
The 30-year Treasury yield just printed 5.13% — the highest since 2007. April CPI re-accelerated to 3.8% YoY. The market has fully ruled out a Fed cut in 2026 and started pricing in a December hike. Long-duration sectors (Real Estate, Utilities, low-FCF tech) feel this immediately. Financials with floating-rate books benefit. Gold and equities have decoupled on this — gold sold off $250 the week of May 16.
Russell 2000 spent four years underperforming and finally reversed. Two structural drivers: (a) tariffs hit multinationals harder than domestic small-caps; (b) regional banks (~16% of IWM) benefit from a steeper curve. The IWM/SPY ratio is breaking out of a five-year base. If it holds, this is the most important rotation of the cycle.
Multi-year returns are seductive but they mask three things every trader should price in:
We don't sell signals. We build the dashboards we use ourselves to read sector rotation, sentiment, and risk in real time. Free, no signup, updated continuously.
One-line takeaway. The three-year US equity bull is real and AI-led, but the leadership is narrowing fast. In 2026 the index is flat, Mag-7 dispersion is the widest in five years, and small-caps are finally taking a turn. If you own indices, you're already long the rotation. If you trade single names, the next twelve months will reward selectivity in a way the last three years did not.
Data sources: S&P Dow Jones Indices, FRED (St. Louis Fed), Yahoo Finance, State Street SPDR sector data, Novel Investor sector return tables, and fxcryptobots internal market notes. Returns are price returns unless labeled total return. Multi-year cumulative figures compound published annual returns and are rounded to the nearest percent. Figures as of market close May 15, 2026 unless noted.