My take is that Putin’s pursuing a resounding victory in Ukraine. He’d have preferred it to be easy. But a hard fight can be spun as just showing that Ukraine was a credible security threat that needed to be taken down. Plus, he now needs a justification for the blood that’s been spilled, along with the sanctions. Zelenskyy is an international symbol now, and leaving him in power would humiliate Putin.
At this point, Russia has had 1.2 troops killed for every 1 Ukrainian; there are about 2 wounded for every 1 killed. At that rate, to kill or wound 25% of Ukraine’s military (215,000 strong) would require about 20,000 Russian deaths and 40,000 Russian wounded—about 5% of their total military manpower, at about half of the casualties America suffered in Vietnam. Note that Russia’s already a sixth of the way there, six days into the conflict.
Beyond that point, I expect that dwindling Ukrainian manpower would lead to accelerating Russian advantage. Very, very roughly, I expect that Russia needs another month to reduce Ukraine’s military to 160,000. The fighters will take the brunt of casualties, and the tooth-to-tail ratio means that this may represent a disproportionately large fraction of their real fighting power. So I will go out on a limb, predicting very far from my areas of knowledge, and estimate Russia is 75% likely to capture Kyiv by April 20th, 2022. I will also put them at 25% likely to stop the invasion without capturing Kyiv and installing a puppet government.
edit: my second prediction means “Russia is 25% likely to stop the invasion, without having captured Kyiv and without having installed a puppet government in Ukraine.”
edit 2: I was foolish to believe any casualty numbers coming out of this war. However, I think the numbers I quoted are Ukraine’s propaganda, so I’ll take them as an exaggerated upper bound of how well they could possibly be doing. Given that, I think their prospects are even more grim than I originally thought. I’m going to just leave up the prediction as it stands, though.
I appreciate you registering your falsifiable predictions on a public forum.
[Putin would] have preferred it to be easy. But a hard fight can be spun as just showing that Ukraine was a credible security threat that needed to be taken down.
I believe precisely the opposite. I will take the reverse side of your prediction. I am confident Putin cannot sell a hard fight in Ukraine to the Russian people using the logic “Ukraine’s fierce resistance to the Russian invasion proves Ukraine was a credible security threat”. (More generally, I predict Putin cannot sell a hard fight in Ukraine to the Russian people at all. (Coercion doesn’t count.))
Consider this comment to be a public, registered counter-prediction.
edit: my second prediction means “Russia is 25% likely to stop the invasion, without having captured Kyiv and without having installed a puppet government in Ukraine.”
Why not just go back and actually edit the comment?
Epistemic legibility. LW isn’t a scientific publication, but it is standard practice not to edit published scientific work directly. Instead, new versions and addendums are released. This way, there’s a record of the changes it’s gone through, and if a version gets cited, the person citing it doesn’t look like they misquoted the document if a new version comes out.
My take is that Putin’s pursuing a resounding victory in Ukraine. He’d have preferred it to be easy. But a hard fight can be spun as just showing that Ukraine was a credible security threat that needed to be taken down. Plus, he now needs a justification for the blood that’s been spilled, along with the sanctions. Zelenskyy is an international symbol now, and leaving him in power would humiliate Putin.
At this point, Russia has had 1.2 troops killed for every 1 Ukrainian; there are about 2 wounded for every 1 killed. At that rate, to kill or wound 25% of Ukraine’s military (215,000 strong) would require about 20,000 Russian deaths and 40,000 Russian wounded—about 5% of their total military manpower, at about half of the casualties America suffered in Vietnam. Note that Russia’s already a sixth of the way there, six days into the conflict.
Beyond that point, I expect that dwindling Ukrainian manpower would lead to accelerating Russian advantage. Very, very roughly, I expect that Russia needs another month to reduce Ukraine’s military to 160,000. The fighters will take the brunt of casualties, and the tooth-to-tail ratio means that this may represent a disproportionately large fraction of their real fighting power. So I will go out on a limb, predicting very far from my areas of knowledge, and estimate Russia is 75% likely to capture Kyiv by April 20th, 2022. I will also put them at 25% likely to stop the invasion without capturing Kyiv and installing a puppet government.
edit: my second prediction means “Russia is 25% likely to stop the invasion, without having captured Kyiv and without having installed a puppet government in Ukraine.”
edit 2: I was foolish to believe any casualty numbers coming out of this war. However, I think the numbers I quoted are Ukraine’s propaganda, so I’ll take them as an exaggerated upper bound of how well they could possibly be doing. Given that, I think their prospects are even more grim than I originally thought. I’m going to just leave up the prediction as it stands, though.
I appreciate you registering your falsifiable predictions on a public forum.
I believe precisely the opposite. I will take the reverse side of your prediction. I am confident Putin cannot sell a hard fight in Ukraine to the Russian people using the logic “Ukraine’s fierce resistance to the Russian invasion proves Ukraine was a credible security threat”. (More generally, I predict Putin cannot sell a hard fight in Ukraine to the Russian people at all. (Coercion doesn’t count.))
Consider this comment to be a public, registered counter-prediction.
Why not just go back and actually edit the comment?
Epistemic legibility. LW isn’t a scientific publication, but it is standard practice not to edit published scientific work directly. Instead, new versions and addendums are released. This way, there’s a record of the changes it’s gone through, and if a version gets cited, the person citing it doesn’t look like they misquoted the document if a new version comes out.