1,026 Nobel Prizes, Four Assumptions, One Dataset (Live Nobel Foundation Data)
This post is a live demonstration of QuantumLayers, our own analytics platform. We loaded a public dataset into it, let it chart and test the data, and embedded the results below. The charts are drawn live in your browser from the dataset itself. The underlying data is public and comes from the Nobel Foundation’s open API; QuantumLayers is the tool doing the analysis.
Every October the Nobel announcements arrive, and with them a familiar picture: a single brilliant mind, a phone call at dawn, a lifetime of solitary work rewarded. It is a good story. We wanted to know how much of it survives contact with the full record, so we loaded every Nobel Prize ever awarded into a public dataset on QuantumLayers: 1,026 awards made to 1,018 people and organizations across 122 award years, from 1901 through the 2025 prizes.
Rather than walk through the data column by column, we picked four things people commonly assume about the prize and checked each one against the numbers. Every chart is fetched live from QuantumLayers’ public chart endpoint when the page loads, with no API key and no login, and each one links to its own standalone ql-chart page.
Assumption 1: “It happens like clockwork, every year”
The prizes are meant to be annual, and for most of their history they have been. But the record has two unmistakable holes. Split the 125 years into equal periods and count the awards in each, and the bars sag twice: 18 awards in the stretch covering the First World War and just 15 in the one covering the Second, when several years passed with no prizes at all. Everywhere else the trend runs steadily upward, from about 40 awards per period in the early 1900s to 90 in the most recent one.
Figure 1: Nobel awards per period, 1901 to 2025, fetched live from QuantumLayers. Source: Nobel Foundation.
▶ Open the live ql-chart on QuantumLayers
The verdict: mostly true, with an asterisk. The rhythm held except when the world was at war. The more interesting question is why the recent bars are more than twice as tall as the early ones, when there are still only six categories. That is the next assumption.
Assumption 2: “It’s a prize for the lone genius”
Each award in the dataset records how the prize was divided: a whole prize, a half, a third, or a quarter. That single column tells a story about how discovery itself has changed. Group the awards by split and look at when they were given, and the pattern is stark. The median whole, undivided prize was awarded in 1958. The median half prize, in 1979. Thirds and quarters, which only become common as teams grow, have medians of 1998 and 2002.
Figure 2: Distribution of award years for each way a prize can be split, fetched live. Source: Nobel Foundation.
▶ Open the live ql-chart on QuantumLayers
The verdict: true in 1901, false today. QuantumLayers’ analysis of variance ranks this as one of the strongest effects in the dataset: how a prize is split explains about 18% of the variation in when it was awarded (η² ≈ 0.18, F ≈ 74, p < 0.0001), and the gap between whole prizes and thirds is large (Cohen’s d ≈ 1.18). The platform also detected a structural break when laureates are ordered by birth date: for people born after the turn of the twentieth century, the average split jumps by about 60%. The lone genius did not disappear, but the laboratory, the collaboration and the three-person citation replaced it as the norm. That is also why the bars in Figure 1 keep rising: more prizes are being shared among more people.
Assumption 3: “Every Nobel is the same kind of honor”
If sharing were simply a sign of the times, every category would drift the same way. They do not. Average the split across each field and two different worlds appear. Literature sits almost exactly at 1.0: 114 of its 122 awards went to one writer, alone. Physics and medicine sit at the other extreme, near 2.4, and in physics only 47 of 230 awards, about one in five, went to a single person.
Figure 3: Average split by field, where 1 means one laureate took the whole prize and 4 means a quarter share. Fetched live. Source: Nobel Foundation.
▶ Open the live ql-chart on QuantumLayers
The verdict: false. QuantumLayers found a significant association between field and split (Cramér’s V ≈ 0.33, p < 0.001). A novel still has one author; a particle detector has thousands, and the prize rules cap the winners at three. The sciences are where that cap bites, and where the typical laureate now stands on a crowded stage. Peace, at about 1.6, sits in between, partly because it is the only category that can go to organizations: all 31 organizational awards in the dataset are Peace Prizes.
Assumption 4: “Women are catching up”
Of the 1,026 awards, 68 went to women, about 6.6%. Where those awards landed matters as much as how many there are. More than half are in just two fields: 20 in Peace and 18 in Literature. In literature, women hold roughly 15% of all awards. In physics they hold 5 of 230, a little over 2%, and in economics 3 of 99.
Figure 4: Nobel awards to women, by field, fetched live. Source: Nobel Foundation.
▶ Open the live ql-chart on QuantumLayers
The verdict: slowly, and unevenly. Two separate results from QuantumLayers back this up. First, gender and field are significantly associated (Cramér’s V ≈ 0.34, χ² ≈ 240, p < 0.0001): the mix of fields women win in is measurably different from the mix men win in. Second, gender has a significant effect on award year (p < 0.001, Cohen’s d ≈ 0.51). The average award to a woman came in 1991, against 1974 for men, which means most of those 68 awards are recent. The direction is real. The pace, especially in the sciences, is not yet anything like parity.
One finding we set aside, and why
Not every significant result is a discovery. QuantumLayers also reported that field has a large effect on award year, with the biggest gap between literature and economics (Cohen’s d ≈ 1.23). That is statistically sound and historically trivial: the economics prize was only established in 1969, so its awards are younger by construction. We mention it because it shows what automated analysis is good for, and where a human still has to read the result. The platform finds the pattern and scores it; knowing that a pattern is an artifact of the rulebook takes context.
There is also a question this dataset cannot answer directly. It records each laureate’s birth date and the award year, but not their age when they won, so “are laureates getting older?” needs a derived column first. That is a good next step for anyone who wants to pick up the thread.
Pull the thread yourself
The dataset is public, so every cut of it is one link away. A few starting points beyond the charts above:
All awards by field Men, women and organizations How physics prizes are split
How this was built
The Nobel Foundation publishes its full laureate list as a CSV through a public API. We pointed QuantumLayers at that URL, and it imported all 1,026 rows, typed each of the twenty columns, and ran its automated statistical tests: analysis of variance for numeric columns across categories, chi-square tests with Cramér’s V for pairs of categories, and structural break detection along the date columns. Findings are ranked by importance and corrected for multiple comparisons before they are reported. Each chart above is drawn with Chart.js from a single unauthenticated POST to QuantumLayers’ ql_get_chart_data endpoint, which returns a ready-to-draw configuration. The numbers quoted in the text reflect the dataset at the time of writing.
Data: Nobel Foundation, Nobel Prize laureates API (api.nobelprize.org), including the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel. Each row is one award, so laureates honored twice appear twice.
This is one in a series of live demonstrations of QuantumLayers: each post loads an open dataset into the platform and shows what its charts and automated statistics reveal. To load a public dataset of your own and get the same charts and findings, start building at quantumlayers.com.