Sondre Ulvund Solstad
banner
sondreus.bsky.social
Sondre Ulvund Solstad
@sondreus.bsky.social
@TheEconomist Senior Data Journalist &
@Harvard University @IQSS affiliate

Data/sims/models ➡️ articles++
not v active here, so best email: sondresolstad@economist.com
Lack of enough data precludes doing the same for Ukraine. But we do find that while Russian forward movement is still slow, it is speeding up. More at link: www.economist.com/interactive/...
July 9, 2025 at 5:19 PM
To estimate Russian losses, I lined up satellite data and shifts to areas of control in Ukraine with more than 200 estimates of total casualties there. This enabled me to construct a credible day-to-day range for the toll of the war since the invasion.
July 9, 2025 at 5:19 PM
To track the Israel-Iran war, @economist.com, I think uniquely, monitors 208 strategic sites in Iran for abnormal thermal activity. It can be hard to know how much maps of confirmed strikes may miss. We find broad overlap, but not perfect. www.economist.com/interactive/...
June 18, 2025 at 2:01 PM
Russia has attacked along virtually the entire front in the past week, and especially heavily in the past few days. But essentially no shift in the front lines. www.economist.com/interactive/...
June 10, 2025 at 4:37 PM
Our cover this week is about girls. The news is good: I find that the enormous global shortfall of baby girls has seen a stunning decline. www.economist.com/leaders/2025...
June 9, 2025 at 2:03 PM
What makes ceasefires, such as the one between Israel and Hamas, more likely to succeed? The experts I talked to and data from 2000+ recent halts in fighting suggest monitoring, politics and a strong start are key:
www.economist.com/graphic-deta...
February 3, 2025 at 10:13 AM
And, if one is willing to state moral priorities clearly, and focus only on charities with plenty of the right data available, one can try to find the best ones. It is precisely this task GiveWell, with a newly expanded team, has spent thousands of hours trying to do (see chart)
December 19, 2024 at 8:11 PM
Context re events in South Korea (Yoon Suk-Yeol at ~15% per
latest Morning Consult data) pro.morningconsult.com/trackers/glo...
December 3, 2024 at 4:35 PM
By matching open-source lists of dead with demographic data, as well as probable dead-to-wounded ratios, I was able to estimate lower bounds for the % of the pre-war male population killed or too injured to fight on.
November 27, 2024 at 11:55 AM
These compare to Russian dead of perhaps 200k so far. In addition come civilians killed, believed to number in the tens of thousands, with over 20k estimated dead in Mariupol alone.
November 27, 2024 at 11:55 AM
In 2022, as he was buying Twitter, posts about free speech surged. This was followed by a leap in 2023 and 2024 in talk of immigration, border control, the integrity of elections and “woke mind virus”.
November 21, 2024 at 2:03 PM
What does Elon Musk's online activity reveal about him? Going through nearly 40,000 posts over the past decade, and the policies implied or suggested in them, @ainsliejstone.bsky.social and I found a changed man
November 21, 2024 at 2:03 PM