chalabya

In discussing with forecasting partners, I learned of a recent Moldova rejection of Russian diplomats and Georgia rejection of restoring diplomatic relations with Russia. This leads me to think those may be slightly more likely invasion points, but I actually think my percentages may have been a bit high to begin with so I'm keeping them the same to become more accurate based on my information.

https://www.kyivpost.com/post/50856

Following the failed 30 hour Easter truce, I increased my predicted likelihood of a ceasefire lasting less than 30 days. However, I think I was too hesitant and increased it only incrementally. After discussion with others and looking at some of the evidence from the crowd forecast, I am more confident further raising the likelihood of this shortest ceasefire option.

Base rate:

https://www.history.com/articles/neutral-countries-world-war-ii

This may seem unrelated, but I feel it's important to think about how often countries invade neighbors in back-to-back conflicts. Usually a country needs time to recouperate after one invasion/conflict ends before starting another. And rapid invasions of several countries have usually meant that a global war is occuring. I think Russia is trying to avoid that, and therefore is unlikely to invade any of these countries in the next two years.

https://www.atlanticcouncil.org/blogs/new-atlanticist/the-real-reason-russia-invaded-ukraine-hint-its-not-nato-expansion/

This article hypothesizes that NATO expansion was not why Ukraine was invaded, but the relative weakness of NATO and Russia's distrust that the U.S. would continue to back up Europe (including non-NATO countries) to counter Russia.

Inside view:

https://www.express.co.uk/news/world/2043251/6-countries-feared-invaded-by-russia

Moldova and Georgia are hypothesized to be the next most likely countries to fear a potential invasion by Russia if it experiences anything akin to a victory in Ukraine. Likelihood still low, but possible.

After reviewing these sources, I am further contracting my likelihood for Armenia, but keeping a low but possible likelihood for Moldova and Georgia.

Base rate:

https://journals.sagepub.com/doi/10.1177/00220027221128300

This article is a detailed academic analysis of ceasefires. One key point is:

"The primary determinant of ceasefire sustainability is always the political will of the conflict parties. Ceasefires only occur when the conflict parties see strategic benefit in stopping fighting, and will always end if one or more of the conflict parties believe that a better deal can be achieved through renewed violence. As a result, ceasefires are more likely to break down if they are imposed, or occur before the conflict parties recognize the favorability of peace."

https://www.nytimes.com/2016/09/16/world/middleeast/another-cease-fire-in-syria-it-could-matter-even-if-it-fails.html

This article examines Syria but gives a bunch of relevant statistics about ceasefires.

"Take, for example, Yugoslavia, where there were 91 mediated truces or cease-fires from 1989 to 2000. Of those, 35% lasted less than a week and 13% lasted exactly a week."

https://www-tandfonline-com.proxy.library.georgetown.edu/doi/full/10.1080/13533312.2021.1894934#d1e285

"Moreover, violations can be expected to occur in the early phase of an arrangement, when trust between the parties is low..."

I'm still struggling to find the hard numbers of average ceasefire length in 21st century interstate conflicts, which I feel would give me a robust base rate. But this information is certainly helpful.

Inside view:

Evidently a one-day Easter truce was implemented this week, but I'm not sure whether that meets the ceasefire criteria of this forecasting question. However, even in the midst of that truce, both sides accused each other of violating it. So I'm standing by my relatively high likelihood of a ceasefire not lasting 30 days.

https://www.reuters.com/world/europe/ukraines-zelenskiy-says-russian-army-trying-create-impression-easter-ceasefire-2025-04-20/

https://www.bbc.com/news/articles/c1wdllj8lwxo

After reviewing these sources, I believe that a ceasefire is unlikely to last a full 30 days or longer. However, I think that international pressure could lead both sides to capitulate and agree to a lasting truce despite unfavorable terms.

I stand by my rationale. One insight my conversation partner offered was to look at the data of cases month-by-month to see if there are significant ebbs and flows throughout a given year. That could lead to increasing or decreasing what we would expect, particularly if we anticipate that in the second half of the given period there will be a habitual decline in the number of cases. However, I just assessed in a six-month-by-six-month fashion and didn't see immediate patterns of decline. I'll still need to do a more detailed analysis later, but I don't think Russia is generally in the habit of reigning in its disinformation campaigns. I stand by my forecast, and I'm still surprised that the crowd forecast thinks that 90+ is a low possibility when there were 41 cases in the first half of the year alone.

My discussion partner and I had the same order of likelihoods for invasion of each of these countries, and we had done very similar research into the bilateral relationships between each and Russia!

A note made by my partner that has influenced my thinking is: Rather than doing research into base rate/inside view and then picking percentages based on vibes from the research, I could be more methodical about starting from 0% (because invasions are not a common global occurence) and then increasing incrementally based on how likely that scenario seems based on the research. I think I was setting the likelihood too high because all of these percentages were "low enough" that the differences didn't seem significant, but they really could mean a lot.

My partner and I had pretty similar research and reasoning for this forecast this week. I could play around with the percentages slightly, but I don't think that would be based on any specific data so I won't. I'm confirming the same forecast because I think that a ceasefire is most likely to either be quickly breached -or- become a lasting end to the war. I think the next most likely is that a 30 day ceasefire is just barely held onto for the duration without technically being breached, but then fighting resumes right afterward.

My initial forecast was way too high. I think I had reasonable rationale for why overall probability was low, but did percentages based on vibes rather than really thinking through how unlikely these scenarios are.

Base rate (bilateral relationships between Russia & each country):

Moldova

Moldova does not share a border with Russia. Russia is increasingly incensed at Moldova cozying up to the EU and weaning itself off of Russian gas since the war in Ukraine. But it still likely is playing the long game and not wanting to invade soon, especially as its resources have been spent in Ukraine.

https://carnegieendowment.org/research/2024/10/moldova-russia-strategy?lang=en¢er=russia-eurasia

"Moscow’s approach to Moldova is to play for time and keep the country in geopolitical limbo by stoking internal divisions, stalling reforms, and fueling disenchantment with the pro-European course."

"Instead of trying to placate Russia and get some breathing space to build ties with the EU, as it did in 2020–2021, since 2022 Chișinău has moved decisively to cut ties with Moscow. It has weaned itself off its dependency on Russian gas and expanded its EU accession track by adding a security and defense partnership with the bloc. Although neutrality is part of its constitution, Moldova has also strengthened cooperation with NATO and elaborated a new National Security Strategy that defines Russia as the key threat."

Armenia

https://carnegieendowment.org/research/2024/07/armenia-navigates-a-path-away-from-russia?lang=en¢er=russia-eurasia

https://oc-media.org/after-years-of-deterioration-armenias-relations-with-russia-appear-to-be-warming/

Armenia seemed to be moving away from Russia briefly, but then the relationship has recently (this year) been improving and economic ties overall have strengthened since the war in Ukraine.

Georgia

https://www.wilsoncenter.org/blog-post/georgia-russia-quasi-partnership-based-elite-interests

"The two countries have no diplomatic relations and parts of Georgia’s internationally recognized territory have been occupied by Russia since the 2008 war. Most tellingly, 69 percent of Georgians view Russia as an enemy and blame Moscow for derailing this South Caucasus country’s path to joining the European Union."

https://carnegieendowment.org/russia-eurasia/politika/2023/07/playing-with-fire-georgias-cautious-rapprochement-with-russia?lang=en

Yet, "the notoriously troubled relationship between Georgia and Russia has, to the surprise of many, entered a new period of increased stability." The current Georgian government still won't formally open diplomatic ties with Russia and has not changed language that identifies it as the occupying force in territories taken during the 2008 war, but behind the scenes the Georgian goverment increasingly follows Russia's lead on certain key policies and has increased the countries' economic interdependence.

Kazakhstan

https://www.chathamhouse.org/2024/02/russias-influence-kazakhstan-increasing-despite-war-ukraine

https://neweasterneurope.eu/2024/09/06/mapping-scenarios-for-future-russia-kazakhstan-relations/

This relationship is better than ever. The two countries cooperate on many diplomatic and economic fronts.

Inside view (current news):

https://babel.ua/en/news/117010-ihe-leader-of-gagauzia-was-released-under-house-arrest-she-is-suspected-of-facilitating-the-financing-of-a-political-party-by-russia

Even in Moldova, the riskiest of these relationships, Russia's tactics seem to be more subtle than gearing up for a full-scale conflict. (I could be wrong and this is a precursor to Russia invading Moldova to "liberate" it from "oppressive government" that criminalized this pro-Russian political party.)

Forecast:

After reviewing additional base rate information about each of these bilateral relationships, I now understand why the crowd forecast favored an invasion of Moldova over an invasion of Georgia. The crowd, indeed, knew more about these relationships than I did, and I am now convinced that that relationship is the most strained and likely to lead to a conflict/invasion in the next two years. I still believe, overall, that any of these is a low possibility due to the state of Russia's military following the long conflict in Ukraine.