michal_dubrawski

I'm still at 5%. I don't see new signals useful for updating in my news search.

However, the time horizon is long for this forecast, so I may be overconfident here as a result of relying too much on the current mental models of the world and my ignorance of the future state of the world, which may be very different from the current one in some important ways.

Scenario analysis exercise may be helpful with questions like this one, and we have some experts within the forecasting team (I do have useful experience with it as well). It would be quite a bit of work, but I believe it would be very beneficial for long term questions. With specific scenarios in mind, we could know what to look for - we could then use indicators/signposting/flagwatching to observe if we are getting closer to each of these scenarios.

This is so unlikely to happen in the remaining time window (one month left) with the Trump administration strongly pushing in the opposite direction.  I rarely feel that confident about a forecast below <1% as I feel here.

As I wrote here, I think in my previous forecast I may have read too much about Russia's longer-term intent from their current actions against Poland and other Eastern flank countries. It may be that they need any edge over Ukraine they can get right now, so even if they have plans against Baltics or Poland they do not care that much right now. And if European NATO countries feel threatened and arm themselves, at least in short-term that works for Russia, as they may see the need to keep more of these weapons to themselves instead of giving more of them to Ukraine. Also, let's say that there is such a plan, I don't think many people would know about it within Russia, so people planning operations against Ukraine support on the NATO territories may not have the knowledge of such plans. Also plans of aggression against each Baltic country may exist in Russia and be systematically updated just as an exercise and as a part of some standard military obligations. This does not mean that there will be an intent, opportunity, capabilities, and decision to use some of them. I highly doubt that there is a set date for a Russian invasion of any NATO country. It is possible that no one in Russia knows that yet, if they were to attack any NATO country within 5 to 10 years. They may think they know, but the situation may change, or the decision-makers may change. I guess a lot depends on what reality the future will bring us. So the interpretation of current actions as indicators for their plans and decisions in a few years time may be of limited predictive value.

I still think it is very unlikely in the given time-frame, but the level of hostility is constantly rising between Russia and NATO, and I would expect that given the opportunity, there would be enthusiasm inside Russia to do more against NATO.

@ctsats has posted some interesting articles in his forecasts here.  

I am increasing my forecast a bit. The reasons for this are twofold.

First is that I revisited my previous forecast, and after reflecting on it, I got the impression that I may have oversimplified the complexity of the situation with my models, or overrationalized things. The warning window before the construction start may be smaller than I anticipated, and my initial correction for the uncertainty of it looks to be insufficient. I looked at some other Chinese infrastructure projects, and how long did it take from the announcement of the agreement and construction start, and it looks like a reasonable time to assume would be 3 months, but it could be even shorter at times.

The second reason is based on the two articles below, which may be read as weak signals (cause for a slight update IMO) that China may change their approach of not engaging their military in local conflicts other than peacekeeping missions or anti-piracy missions, as their Belt and Road projects experience increased setbacks from terrorism (in Mali, Niger, and Burkina Faso), and as the second article points to the benefits of Chinese PLA  engagements in so called "small wars" for gaining testing ground for their technology and combat experience for the personnel.

The Diplomat: "China’s Sahel Gamble Falters as Insurgencies Rage"

Center for International Maritime Security, guest author Michael Hanson: "China’s Coming Small Wars"

I am tagging @Akkete ,  @predictionman , @LogicCurve , @grainmummy , @sebawi , @Jeroen , @Rene  and @YJAung as we either had a discussion about this question previously or they upvoted or referenced my previous forecast in their comments.

I am slightly reducing probabilities for Latvia and Lithuania in this timeframe (17 months or 1.4 year remaining). A lot can change in what is almost 1 year and a half, but I don't think Russia will be able to act that fast based on how their forces are still engaged in Ukraine. I think Ukraine will remain their main focus for quite some time.

I believe that maintaining the impression that there is real and current threat of invasion to the EU and NATO countries in Baltics and NATO Eastern Flank is currently beneficial for Russia. Look at the recent incidents like with those armed individuals in unmarked uniforms near Estonia border and some media reaction with warnings that the little green men have already come to NATO's doorsteps.

If Russia was planning to invade NATO Eastern Flank anytime soon they would not be doing what they are doing now - I mean for example showing Poland the need to invest in drone detection and antidrone protection on their Eastern border.  Ok, if one plans the attack and wants the element of surprise to be on their side, they may want to make numerous incursions into target airspace and territorial waters creating the impression that this is something which happens often and which is "normal". Also something like conducting regular military exercises at the target country's border as a cover up for the military buildup for future invasion, as in case of USSR's 1968  invasion of Czechoslovakia or Russia's 2022 invasion of Ukraine. However, I think for now this is all about the posture, image, about signaling, putting pressure, making NATO countries' policymakers uncomfortable, trying to influence their societies, showing the world "we can do that", also I guess it is good for internal audience in Russia. The result is that European NATO countries are making preparations to defend themselves from future Russian invasion, and if Russia does not plan to invade any NATO country anytime soon, this is good for them as it reduces how much money that is assigned for military purposes and how many weapons and ammunition can be sent to Ukraine. My point is that if policymakers in EU and European NATO were sure that we are safe from Russia because we are in NATO, then we would be able to sent more of the equipment to Ukraine. Even the US during the Biden administration was calculating how many weapons they could give to Ukraine and how many they needed to keep in reserve in case of other conflicts.

We have 14 months left in the forecasted time window. I like @lzhang suggestion here about looking at how long it took between negotiation and opening of the base. For our purpose, a base under construction still counts, so while this is only a single case and other negotiations of similar matters may be different, we have initial news media reporting about this from 9 May 2015. We have the next news media report from 21 January 2016 that the negotiations were concluded. That is 257 days or 8 months and 1 week. And the construction is reported to have started in March 2016, but the first official confrmation I found is from April and is mentioned here: "On April 8, 2016 China’s Ministry of Defense reported that construction of China’s first oversea military base had begun." So that gives us 335 days or 11 months between first reporting of negotiations and the reporting of the construction. Ok, negotiations can take a shorter time or media can learn about them in later time than in case of Djibouti (I would expect US Intelligence providing them with such a report, like in case of Equatorial Guinea), but still, this is useful information, as it allows us to reduce the time window of when the information should appear for this likely to take place in remaining time. If we reduce number of remaining days till 1 January 2027 (427) by 335 days it gives us only 92 days or 3 months, even if multiply this by 2 assuming that things could happen faster this is only 6 months and not 14 months. 

Time invariant calculations of Laplace's rule of continuity and Jeffreys prior assuming the remaining 184 days 

Case A — since first success - construction start of base in Djibouti, assuming March 2016,  (variable window,  S=1,T=3,531

Laplace: p≈4.95%

 ​Jeffreys: p≈2.51%

 ​

Case B — since first reporting of another attempt (fixed window,  S=0,T=1,426)

Laplace:  p≈11.43%

 ​Jeffreys:  p≈5.89%

Also, this article makes interesting point:
"China has shifted its strategy in the Indian Ocean region, favouring loyal intermediaries over direct military bases. (...) China’s approach in the region has evolved. Its first overseas military base, established in Djibouti in 2017 on the Red Sea’s edge, marked a turning point but also triggered strong diplomatic pushback from regional and global powers. Since then, overt bases have proven costly in terms of perception and strategic trust. As a result, Beijing appears to favour arrangements that deliver similar benefits— access to ports, surveillance routes, logistic support— without formal PLA installations or permanent overseas personnel openly acknowledged in foreign territories.

By empowering Pakistan as a maritime proxy, China leverages its existing defense ties, ship deliveries (e.g. Hangor-class submarines), and naval cooperation to project influence. Meanwhile, under agreements with Somalia and Turkey, external actors can carry out security operations, training, and oversight in strategic areas— key sea lanes, coastal areas, and zones vulnerable to piracy or external pressures— all while maintaining plausible deniability for Beijing. This creates a web of influence that can advance China’s security and economic goals without provoking formal alliances or accusations of encirclement."