Quantum BiasNvidia Stock Split: Do Stocks Really Go Up After a Split?
In theory, a split changes nothing. Here's what prices did anyway.
In this article
In the announcement week, yes. After the split, not reliably, and that includes both Nvidia splits. Across six big-tech splits announced from 2020 to 2024, the stocks beat the S&P 500 by a median of about 10 percentage points in the announcement week, and by a median of about 28 points by the time the split took effect. In the year after the split itself, four of six trailed the index. Three of the six announcements came with quarterly earnings, so the "split effect" is hard to separate from the news.
| Question | Data | Answer |
|---|---|---|
| How did big-tech stocks behave around their splits? | Six splits: Apple 2020, Tesla 2020, Nvidia 2021 and 2024, Alphabet 2022, Amazon 2022 | Strong around the announcement; mixed after the split took effect |
What a split is, and isn't
In a 4-for-1 split, each share becomes four shares worth a quarter of the price. The company's value, earnings and ownership don't change. In a perfectly rational market, a split would be a non-event.
But splits get a ton of attention, and research has long found they're not quite neutral:
- In 1969, Eugene Fama, Lawrence Fisher, Michael Jensen and Richard Roll used stock splits in one of the first modern "event studies", The Adjustment of Stock Prices to New Information, published in the International Economic Review. They found that stocks tended to rise strongly before splits, and argued that splits signal management's confidence, often followed by dividend increases.
- In 1996, David Ikenberry, Graeme Rankine and Earl Stice studied 1,275 two-for-one splits in What Do Stock Splits Really Signal?, published in the Journal of Financial and Quantitative Analysis. They found excess returns of 7.93% in the first year after the split and 12.15% over three years.
We looked at the six most-watched splits of recent years.
The events
| Company | Ratio | Announced | Split-adjusted trading began | Announced with earnings? |
|---|---|---|---|---|
| Apple | 4-for-1 | July 30, 2020 | August 31, 2020 | Yes |
| Tesla | 5-for-1 | August 11, 2020 | August 31, 2020 | No |
| Nvidia | 4-for-1 | May 21, 2021 | July 20, 2021 | No |
| Alphabet | 20-for-1 | February 1, 2022 | July 18, 2022 | Yes |
| Amazon | 20-for-1 | March 9, 2022 | June 6, 2022 | No |
| Nvidia | 10-for-1 | May 22, 2024 | June 10, 2024 | Yes |
What we found
Returns relative to the S&P 500 (SPY), in percentage points, using weekly closes:
| Company | Announcement week | Announcement to split | 13 weeks after split | 52 weeks after split |
|---|---|---|---|---|
| Apple 2020 | +13.0 | +24.1 | −7.2 | −5.9 |
| Tesla 2020 | +13.9 | +40.3 | +38.4 | +62.2 |
| Nvidia 2021 | +5.7 | +31.4 | +12.9 | −2.5 |
| Alphabet 2022 | +6.6 | −29.6 | −14.3 | −4.0 |
| Amazon 2022 | +2.7 | −15.2 | +16.8 | +0.4 |
| Nvidia 2024 | +15.1 | +40.1 | −13.6 | −3.7 |
| Median | +9.8 | +27.8 | +2.8 | −3.1 |
Exact values from table 3 in this article. Bar lengths share a zero baseline; no new calculation or simulated price path. Read the table and its stated sample and limits in this article.
The announcement week was strong every time. All six stocks beat the S&P 500 in the week of the announcement, by 2.7 to 15.1 points.
From announcement to split, results went two ways. The 2020, 2021 and 2024 splits came during strong rallies in their stocks. The two 2022 splits came during the 2022 bear market, and both stocks fell hard before the split took effect.
After the split, there was no consistent edge. Over the next 52 weeks, only Tesla and, barely, Amazon beat the index. The median was −3.1 points. Tesla's +62.2 points pulls the average up and shows how much one case can dominate a small sample.
Nvidia's two stock splits, side by side
Nvidia split its stock twice in three years, which makes for a useful pair. Its 4-for-1 split, announced in May 2021 without earnings, beat the S&P 500 by 5.7 points in the announcement week and by 31.4 points by the time it took effect. Its 10-for-1 split, announced with a record quarter in May 2024, beat the index by 15.1 points that week and by 40.1 points up to the split. Both then trailed the S&P 500 over the next 52 weeks, by 2.5 and 3.7 points. The excitement came before the split, not after.
Exact values from table 3 in this article. Bar lengths share a zero baseline; no new calculation or simulated price path. Read the table and its stated sample and limits in this article.
Why the announcement weeks were strong
Three explanations, and more than one can be true:
- The news around the split. Apple, Alphabet and Nvidia (2024) announced their splits with quarterly results. You can't separate strong earnings and a split in the same release with weekly data. Apple's announcement week included its reaction to a strong quarter. Nvidia's 2024 week included its reaction to a record quarter.
- Signaling. As Fama and his coauthors argued, management usually splits after a strong run and when it expects that run to continue. The split tells investors something about management's confidence.
- Attention and accessibility. Lower share prices make stocks easier to buy and more visible to individual investors, as our article on attention bias explains. Fractional shares have made price per share matter less in practice, but not in people's heads.
Why the long-run result differs from the research
Ikenberry and his coauthors found positive excess returns over one to three years after splits. Our six cases don't show that, but six cases can't overturn a result built on 1,275. And in mega-caps followed by thousands of analysts, any predictable drift is more likely to get traded away.
What this means for traders
- Read the release, not the ratio. When a split comes with earnings, the earnings drive most of the reaction.
- Don't buy the split date expecting a pop. The effective date is known weeks ahead of time. Nothing new happens that day. In our six cases, the next 13 weeks ranged from −14 to +38 points relative to the index.
- If you trade NQ or ES, split announcements by the biggest index members can add to a strong earnings reaction, which moves the index. Our concentration article shows how much weight these names carry. The split itself doesn't change index weights: the index adjusts for it.
How to check it yourself
- Find the announcement date and the first split-adjusted trading day in the company's press release or SEC filing.
- Measure the stock's return minus the index's over the announcement week, the period to the split, and 13 and 52 weeks after.
- Note whether earnings came out the same day.
Limits
- Six events. These are case studies, not a statistical test.
- Weekly data. Announcement reactions are measured over the full week, which includes other news.
- Mixed-up announcements. Three of six came with earnings.
- Survivorship. These companies are today's giants. Splits by companies that later struggled aren't in the sample.
How we did it
- Weekly adjusted prices from Alpha Vantage (
TIME_SERIES_WEEKLY_ADJUSTED), through October 9, 2026. Returns are measured relative to SPY's total return. - Announcement dates come from the company releases, filings and press reports listed below. The first split-adjusted trading weeks were confirmed in the price data.
Sources
- Fama, E. F., Fisher, L., Jensen, M. C., & Roll, R. (1969). The adjustment of stock prices to new information. International Economic Review, 10(1), 1–21. IDEAS/RePEc
- Ikenberry, D. L., Rankine, G., & Stice, E. K. (1996). What do stock splits really signal? Journal of Financial and Quantitative Analysis, 31(3), 357–375. IDEAS/RePEc
- Apple. (2020, July 30). Apple reports third quarter results [press release]. apple.com
- Tesla, Inc. (2020, August 11). Form 8-K (five-for-one stock split). U.S. Securities and Exchange Commission. sec.gov
- NVIDIA Corporation. (2021, May 21). Form 8-K (four-for-one stock split). U.S. Securities and Exchange Commission. sec.gov
- Alphabet Inc. (2022, February 1). Alphabet announces fourth quarter and fiscal year 2021 results [press release, including the 20-for-1 split]. abc.xyz
- CNBC. (2022, March 9). Amazon announces 20-for-1 stock split, $10 billion buyback. cnbc.com. Accessed October 12, 2026.
- Capoot, A. (2024, May 22). Nvidia announces 10-for-1 stock split after AI-driven boom in share price. CNBC; republished by NBC Miami. nbcmiami.com. Accessed October 12, 2026.
- Alpha Vantage. TIME_SERIES_WEEKLY_ADJUSTED (AAPL, TSLA, NVDA, GOOGL, AMZN, SPY). Data retrieved October 2026.
FAQ
Does a stock split increase a company's value? No. It splits each share into more shares at a lower price. The company's value doesn't change.
Do stocks go up after a split is announced? In our six big-tech cases, all six beat the S&P 500 in the announcement week, but three announcements came with earnings.
Do stocks go up after the split takes effect? In our six cases, four of six trailed the S&P 500 over the next 52 weeks. Older research on 1,275 splits found positive excess returns on average.
Why do companies split their stock? To lower the share price, often after a long rise, and, according to research, to signal confidence.
Did Nvidia stock go up after its splits? Before them, yes: both splits beat the S&P 500 by more than 30 points from announcement to split. In the 52 weeks after each split, Nvidia trailed the index slightly.
Research and education only. Not financial advice. Past results do not predict future returns.
How this article was made: drafted with AI assistance from public data and the sources listed above; facts and figures checked by Gabriel Décarie and Karel Panneton.
The week just closed, summarized, and the economic calendar for the week ahead.