> No matter how hard someone tries to convince me the Russians were shelling a nuke plant they controlled, I'm just not buying it. They were free to shut it down or blow it with explosives if they wanted
As I would expect. But Putin has decades of history of creating pretexts for his actions. That's always been his M.O. The Russian army is not Russia's most fearful weapon (obviously) — that honor belongs to their mouths. When Russian public figures go on TV, they are performing combat operations. But of course, they don't stop at just lying, they manufacture evidence (and somehow you’re absolutely certain that that’s not happening)... though they don't really put their hearts into it: https://twitter.com/DPiepgrass/status/1553257164386418688
> Putin is not going to back down, either. Losing the war would be the end of him, so he'll escalate until he wins or he gets what he wants
You were right about escalation, but that hardly means he'll get what he wants. He wants all of Ukraine, but probably can't get all of Donetsk.
I was thinking about it more tonight and realized that the negative consequences of mobilization may prevent Putin from announcing general mobilization, but partial mobilization is much less risky. So I put down a 40% chance on "A law or decree is passed by the government of Russia ordering the activation of reserve military forces and authorizing the use of these forces, as well as conscripts, in Ukraine or combat elsewhere outside of Russian territory" or that Russia declares war. And then just a few minutes later I hear the news:
> In a video address on Sept. 21, Russian President Vladimir Putin declared a partial mobilization. It will primarily target the Russian Armed Forces' reserve personnel. "Mobilization measures will begin today," Putin said.
Putin must think: Russia has 144 million people and I need 500,000 tops. I'm only mobilizing the poor saps who couldn't figure out how to get out of conscription earlier — politically powerless people — they are no threat to me. Plus they already have basic combat skills; I can use them right away.
The mobilization and sham referenda will probably let Putin hold (most of) his stolen land, but it makes the regime a bit less stable. Superficially it seems like LNR/DNR "joining" Russia should end the mobilization of residents there, but the Kremlin must have loopholes in mind.
I still give a 60% chance that the Kremlin will lose the Kherson region before February and 85% before 2024. Putin cannot reinforce it well enough (edit: make that 55% and 80% — maybe the mobilization will go well, but there are limits on how many reinforcements he can send). Russia will probably keep its land bridge to Crimea, though. Western weapon shipments to Ukraine are likely to speed up a bit; whatever shipments happen in the next month (assuming UKR soldiers are trained on them) will have the most impact.
Right now Ukraine would be wise to focus on "partisan" activity and IEDs (edit: and ops to educate Russian soldiers — I love the socratic method used with Russian POWs, asking questions that Russian State TV can’t answer; could it be adapted for the battlefield?). It'll be difficult to set up attacks in time for the sham referenda, but I wish them luck. If they have enough manpower and weapons, I figure they could try for a two- or three-pronged offensive around Melitopol to clip the land bridge before the end of the year. But that doesn't seem worthwhile unless they can secure the M18 bridge to Crimea, which prima facie requires far more manpower than was used in the Kharkiv offensive. Failing to secure the bridge would make gains there difficult to hold, so an offensive there is unlikely (25% edit: make that 12% before next May — it would be too risky).
Edit Sept 27: due to Russian mobilisation, my probability estimate that Ukraine will have more, or about the same, land area on May 13, 2023 than on May 13, 2022 is lowered to 85%.
Edit Oct 4: updated to 90% chance given Ukraine’s success in the Kharkiv and Kherson areas. Major counterattack near Melitopol/Berdyansk: 20%.
Edit Nov. 9: updated to 99% chance, as Russia is pulling out of Kherson. Major counterattack near Melitopol/Berdyansk: 12%. Nov 11: On second thought, 18%. Ukraine regained Kherson in line with expectations.
Nov 24: 30% chance Ukraine launches an offensive that goes at least halfway to cutting the land bridge to Crimea before May 2023. Putin’s mobiks are underequipped and underpaid; Ukraine’s capabilities are unclear.
December 17: 20% chance Ukraine launches such an offensive. Ukraine’s military is looking a tad weak for dealing with ~300,000 mobiks. Dec. 21: Make that 30% again, as it’s hard to verify Ukrainian strength, and Ukraine’s best chance to attack is winter. Confusingly a Ukrainian official says he thinks Russia will go for Kyiv again; maybe a PR move to keep international focus on Ukraine? In any case, from where I sit, Ukraine and Russia are equally likely to launch a large assault in the next few months, though a Ukrainian assault is likely to take more territory.
Jan. 2: Lowering to 20% chance based on a general sense that several factors work against Ukraine, including having too few troops relative to Russia. At the same time, Russia can’t advance very far due to Ukraine’s western weapons and Russia’s low military competence at all levels.
Jan. 11: 15% chance for large UKR offensive; 45% for large Russian offensive beyond just Bahkmut. Russia knows how to look menacing, I’ll give ’em that. Edit: the most likely scenario is a Russian offensive of “medium” effectiveness, in which they capture Bakhmut and a few other small cities and towns in Donbas and in other areas by early May. Ukraine may also launch an offensive, but it is unlikely to advance halfway to the Azov Sea by May.
Jan. 25: Chance of big, successful Ukraine offensive by May is now up to 30% due to pledges of tanks & other helpful hardware, though delivery of at least the U.S. tanks is reportedly delayed until end-of-year. Jan 28: Big Russian offensive down to 15% given these new weapons, and because (i) RU just seems a bit too weak for a truly big offensive, (ii) the expected second wave of mobilization hasn’t materialized, and (iii) “Conventional Russian forces are likely replacing exhausted Wagner Group forces” (ISW). While it’s likely that Bakhmut will fall in the next 15 days (65%), Bakhmut was always Wagner’s goal; Russia wants it too, but won’t necessarily prioritize it. I think the most likely scenario is more of the same for a month or three, as each side quietly tries to strengthen forces, followed by a medium-size offensive by one side or both, producing one or two salients less than 50 km deep. If the Kerch Bridge is hit, an immediate offensive by Ukraine becomes pretty likely.
Feb. 3: After nearly a year of war I’ve still found no good sources to reliably estimate key statistics such as personnel losses on each side, or the extent of corruption on each side. Meanwhile, most commentators I’m seeing have only average-to-decent epistemics (i.e., typically fairly poor). Perun is good, but is unsure himself which sources to trust. Predictions are actually getting harder over time, as it becomes harder and harder to tell who has the advantage. Probably one side is losing, but which side is very hard to say. The best statistics that are available look fairly good for Ukraine, e.g.
Ukraine is clearly “better at this warfare thing” by Oryx data, but on the other hand, nearly its entire air force is gone and Russia is still 3x the size, has more than 3x the weapons, and can mobilize more men if Putin says the word. Who will gain territory, then, depends on all kinds of uncertain details from the exact level of Western support, to how willing Ukrainian leaders are to improving military command structures, to Russian competence, to Putin’s propensity to toss more Russians in the meat grinder. Overall I’m more pessimistic after reporting from e.g. Willy OAM & HistoryLegends; the Putinists are likely to grab more land than the Ukrainians, and what new tanks Ukraine eventually gets may simply replace losses. On the plus side, staying in a defensive posture seems better for minimizing attrition, and I doubt Putin will get back as much territory this year as he had in July 2022, let alone March. Edit: after watching Perun’s latest, I think Ukraine’s Western allies will give Ukraine just barely enough hardware to let it hold ground or maintain the upper hand — so they can achieve successful offensives, but only small ones. Putin’s risk-aversion and the yes-man culture he created are a bad combo w.r.t. his goals, but for now, Russia is on the attack and will likely gain small amounts of territory.
Feb. 10: Ukraine has an equipment/ammo shortage, so chance of medium-size Ukrainian offensive by May is at 25%. Russian offensive is already started. Chance of medium success: 40%. Chance of at least modest success beyond Bakhmut (>10km advance): 75%. Chance of successful medium Ukrainian offensive by end of year: 50%. Bakhmut is holding out better than I expected, but I still expect Ukraine to withdraw this month (65%), though Russia may shift focus away from Bakhmut if they have a breakthrough elsewhere. As always, Russia cannot take a large city such as Kharkiv or Zaporizhzhia (98%) but perhaps they can force Ukraine to reallocate troops by attacking a city indiscriminately.
Feb. 17: Wagner is being starved of resources, so Bakhmut likely won’t fall in February (70%). Feb. 19: Russian forces seem degraded and I’m tempted to downgrade the chance of a >10km Russian advance, but on second thought, >10km isn’t very much so 75% still seems about right (IOW the original probability should’ve been higher). There’s a lot of expectations for a Ukrainian spring offensive, so although I’m missing the signs that such a thing is practical, I’m raising the chance of “medium success” (>25km advance by mid-May) to 40% and the chance of a >10km advance to 70%.
Feb. 26: Russia seems to be running out of steam; dropping chance of >10km Russian advance outside Bakhmut by May to 65%. Raising chance of >25km Ukrainian advance to 50% by May and 65% by year-end, because I think the UM guy really knows something; however, Russian defenses should be much tougher in south Ukraine than they were around Balaklia.
March 4: On second thought, 50%/65% is too high; let’s say 35%/60%. Russians are likely to take the rest of Bakhmut in the next two days (60%).
March 6: Either Ukrainians are really stubborn or Wagner’s too exhausted. Maybe both. Not even sure if Bakhmut will fall by May (55%). Both sides claim they’re trying to attrit the enemy in Bakhmut and both sides overestimate the extent to which this is true. I’m 98% sure Ukraine is taking fewer casualties than Russia, but only 65% sure they will win back a substantial amount of territory this year. Russia still won’t capture any major city like Zaporizhzhia or Kharkiv this year (99%).
March 29: I gave a lot of uncertain probabilities and it occurs to me I should say what I’m more certain about. As mentioned, Russia won’t capture any major city this year (99%). Belarus will not invade Ukraine (97%) and Russian troops won’t launch an offensive from Belarus (90%). In the last 9 months of 2023, Russia will fire fewer missiles at Ukraine from Russia/Crimea than they did in the first month of their invasion, i.e. 1200 or less (80%). Russia won’t nuke Ukraine or the West (98%). The Russian economy will shrink more in 2023 than 2022 if military expenditures are ignored (75%) but I give a 50% chance that Rosstat figures agree.
Ukraine will launch a offensive this year (96%) in the south (92%) and it will advance further than the Russian offensive near Bakhmut did, i.e. more than ~10 km (80%), while taking fewer casualties (80%) although this last point will be hard/impossible to assess. It looks like a southern offensive is starting already but it could be a feint or recon. Public information suggests they don’t have enough equipment yet (or even enough trained soldiers) but Ukraine’s allies could have delivered more equipment secretly and/or Ukraine may take the risk of using up old equipment before new equipment arrives. Contrary to Russian claims, Ukraine will not capture anything in Russia beyond 5 km from the Russian border (95%); stunts by Russian militants don’t count. Ukraine also won’t capture more than 0.2% of Crimea this year (80%) and certainly less than 2% (95%). Even money says they cut the land bridge to Crimea (50%), but conditional on doing so, they have a good chance of also retaking the Zaporizhzhia nuclear plant within a year (80%). Russia won’t hold any Ukrainian territory more than 30 km beyond current front lines by the end of 2023 (75%). Russia will not capture the entire Donetsk oblast (90%) and they probably won’t even recapture Luhank (80%). Conditional on the southern Ukrainian offensive failing to reach 40% of the way to the Azov sea, they probably won’t cut the land bridge by the end of 2024 either (65%), mainly because the West is likely not to send enough weapons. On the other hand, conditional on cutting the land bridge, Ukraine is likely to strike one or both Kerch bridges again (70%) but is unlikely to capture 10% of Crimea within a year afterward (75%).
For fun here are my probabilities on Medvedev’s predictions for 2023: (1) 8% (2) 1% (3) 0.1% (4) 0.0001% (5) 0.0001% (6) 0.00005% (7) 5% (8) 0% (9) 0.5% (10) 0.01%. In other words, Medvedev is full of sh*t and the chance that even a single one of his predictions comes true is about 10%.
