It is free, no registration, English, Ukrainian and Russian, and runs entirely inside the browser. It opens fully populated: 33 refineries with capacities, units and regions, the strike record reconciled to 20 September with 44 damaging events, and the constants, each carrying its source and the direction in which that source is most probably wrong.

There is a heat map of where the damage sits, a flow diagram of where gasoline comes from and who receives it, and a twelve-month space of scenarios against the thresholds below which particular consumers stop being served.

The Lab is also the engine room behind widget 1.3. (Petrol: administrative response ladder) of the panel Limits of Russian endurance in war at incidence.world.

Widget 1.3 was created to show what the regime was forced to do to respond to crisis, while the Lab was created to show what is forcing such response along with quantified consequences for up to 12 months at the same time predicting a moment when the next rung (and higher price to pay by Russia) becomes inevitable.

Both the Lab and the widget together provide another important measurement. When the Lab model predicts that the rung 6 is due while the Kremlin still holds at 5, it shows how much the Kremlin forces Russians to absorb (and thus pay for the war) through queues, prices and drawn-down reserves.

There is an important gap in how Ukrainian strikes on Russian oil refineries are monitored and analyzed.

For example, Reuters will report that Ryazan refinery lost AT-6 and AVT-4 equipment after Ukrainian strike on 6 September. Such reports are not connected to a model that tracks and predicts how much high-octane gasoline will be lost for Russia consumption in 1, 3 or 6 months. To calculate this a country-wide model is required which takes into account the equipment at each plant and tracks the race of damage versus repairment in such incomplete data environment with many uncertainties.

Another type of reporting is strike trackers which count strikes, sometimes equipment types and damage severity, but no thorough capacity aggregates or recovery projections, at best some aggregation is a rough automated approximation without important details taken into particularity. They glaring one is an ability of refineries with damaged high octane equipment to increase low-octane (AI-92) gasoline production.

Bloomberg and EA Analytics have upward bias: tanker traffic plus thermal imagery = a measure of activity, not of throughput, e.g. a “hot” equipment piece does not tell the mode of operation and amount it actually produces, tanker traffic can be used for storage or use previously stored barrels, not produced ones, and all such errors run upward.

The Lab is built to close exactly this step.

And it was also created for disagreement, experimentation, different scenarios probing. A plant can be set to a different state, and any of the eleven parameters can be moved within its stated range. An event card can be switched on, such as import financing, a strike moratorium or odd-even rationing. The model then derives what follows: station availability for AI-95+, the rationing ladder, coverage by region and by consumption tier, and the fiscal price of that bundle of assumptions.

New strikes go in as they happen, one at a time or in a batch. The dialog asks what was observed, such as a named crude train stopped, a plant which stopped taking crude, or a reformer identified in imagery. It then applies a published rule to turn the observation into a share of capacity, showing the rule as it goes, so the argument is with the rule and not with an invented number.

Two gates test the result. One checks against observed availability at the pump. The other replays the campaign from May and checks that the model reproduces today’s damage. A scenario which fails either gate is marked as failing, on screen, rather than quietly fudged until it passes.

Today’s state is not a stored snapshot but a derivation: the last hand-reconciled anchor plus every recorded strike since, evolved through the model’s own repair rules. Restarts are never reported reliably, so they are modelled, and the gap between what the model expects and what the next reconciliation finds is published as drift. That gap tests the repair assumptions, the one thing here nobody can observe and everybody guesses at.

Incidentally, it also tests the one charitable reading of the Bloomberg and EA estimates, namely that their higher runs are not method bias but plants coming back faster than the repair rules allow. If so, the next reconciliation finds more capacity running than expected, and the drift line shows the model running low, on screen.

The point is not to publish a number but to publish a machine which produces numbers from assumptions that can be inspected, rejected and replaced, and which reports when it has failed its own tests. Anyone tracking strikes is invited to try to break it, to send events, or to add them directly.

The bill for 2026 is ₽16.1 trn above revenue, 124% of the entire National Defence line. Over January to July, 58% of what had to be financed was closed outside the market, 51 points of it from Treasury balances = more than half of seven months’ financing came out of money already sitting on the Treasury account. When the “market” reopened on 2 September, after 63 days without a single completed placement, it gave 99% of the quarter’s borrowing in one session. In 2022, 2024 and 2025 this emergency mode came in November and December. This year it came in September.

In the second quarter the Bank of Russia was the only central bank selling gold in any real volume, 21.77 t against 4.23 t for the next seller, and it did so in a quarter when the official sector as a whole bought 289 t, more than in any second quarter on record. In 2025 it sold 6 t: a small number, but the first time in twenty years the stock went down over a full year. In the first seven months of 2026 it sold 49.7 t.

Every other conversion mechanism on the panel has a ceiling the Kremlin sets for itself; the ceiling on gold is set in Beijing, New Delhi and Ankara.

incidence.world

What is interesting: the decree is not about facility only, it is about any property of an owner and ANY assets inside Russia. Rosimushchestvo, the federal state property management agency, will be able to manage and execute every owner’s rights and power with only one exception: disposal, but without explicit end date. While it’s masqueraded as not a nationalization de jure (title stays), de facto control will be fully seized, and the cash flow with it, and a change of ownership will most probably follow down the road, each time “due to special situation” and each time “on special terms”.

Article 4 settles the brutal new truth for owners: 1) this “temporary management” costs are met out of income of the facility/property: Russian state declared it cannot defend infrastructure important objects inside country and punished owners for its own doing by taking their income (not profits, but any incoming cash flow) to pay for defense. 2) if the facility/property to remain under an owner’s management as before, the owner still will spend for such protection, but still will bear the responsibility “for not defending it” and this is in the situation when Russian army cannot defend the strategic port and oil terminal in Novorossiysk with all the anti-aircraft defences, navy anti-air guns and experienced crews in the main navy base.

The Incidence panel named this in advance in the M.3 widget with explicit statement what widget would count as significant movement: both the permission extending beyond energy and strategic facilities, and the requirement of protection at the owner’s expense as the only way to proceed with “usual” operations. And this decree No. 604 does both. The panel’s data run to 21 August. The decree is dated the 24th, and it is quite earlier than author was expecting, such movement signals acceleration of the conversion and war burden shifting further along the ladder. And this time with unhindered direct or even blatant signal that emphasizes the urgency of the need and inability “to fill the gap” with other means.

What is especially important is the reason #3 (the “failure to repair”) for the “temporary management” since nobody legislates against behaviour that is not happening (or that is not definitely expected to widen), because that is the Kremlin’s own confirmation that owners of oil refinery plants that are repeatedly struck have stopped rebuilding, while owners of e-commerce platforms, like Wildberries, for example, are redirecting stock rather than rebuilding what was hit. This is sound business judgement and pure accounting arithmetic rather than a desire to show disloyalty, the most logical business behaviour when the same facility was struck and most probably will be struck again and again.

On the panel’s Oil Refining Ladder of the M.3 widget that behaviour is the last 7th rung: repairs ceasing step is used as an expensive signal that shows admission that what will be even repaired cannot be protected. This ladder in the panel release from 21 of August, stands at the 5th rung, and for now it stays there, because a rung will be triggered with observable events and proofs. But still, the decree tells us which way the observable is running. The 7th rung will be triggered when at least 50% of the damaged plants will be stopped being repaired de facto. This new decree is an attempt to make the logical owner’s answer illegal and/or seize the property and potentially give it away (via nationalization later or via de facto control on cash flow) to loyal figures who will be paid by a virtually free for them property to use some part of this newly gained cash flow to defend the property.

A regime whose claim on power rests on being able to enforce (that comes only along with ability to protect) decided it can tell its largest owners in writing that it can no longer perform its ultimate duties as a state (not to mention as a military enraged state ruling its society via military hysteria). The decree enforcement is the admission and a cry of inability.

The forecast: despite business community asking for it officially for long time, there will be no general funding mechanism for civilian counter-drone/counter-missile protection till the end of this year, including no dedicated fund or subsidy line or budget programme beyond the profit-tax deduction that is already in place (and stated in the panel). The owner will pay the defense bill one way or another. One federal instrument that pays for it refutes this forecast, and I do not need to be the one who finds it.

The panel: incidence.world — v1.9, data as of 21 August 2026, in English, Ukrainian and Russian. CC BY 4.0 · DOI 10.5281/zenodo.22068421

Even the floating-rate 29028 issue failed despite a floating-coupon bond carries no interest-rate risk for the buyer. No bids were submitted (at the price the Ministry was willing to tolerate) = the offered price was not tolerated by the market. While formally, though, it is a failure of both side, actually, only one of them matters for this case. This fact alone does not mean that Russia cannot borrow any longer. Of course, it still can but at a higher interest (to reflect a rising risk), but it was decided not to offer such higher interest.

This fact also does not mean that the channel has been exhausted and the proof is the fact that on 23 July, the Ministry registered RUB 1.5 trillion (USD 18.3 billion) of floating-rate OFZ bonds. And just 5 days later it secured their inclusion in the higher listing tier with nothing happened after that: it was not activated = forced absorption by banks has been prepared for the future action. Logically, since everyone interested saw it, a mechanism that has been prepared but not used has already been spent as a political resource that is a different signal and it’s not the same as keeping it in reserve.

Of course, one of the main reasons is the State Duma elections that will take place on 20 September, so it’s also not so much about market, this pause should end after them (plus several weeks for Duma to make certain decisions) if it was only electoral. It is about market although in the aspect of the price of the bonds and signals that can be sent and decided not to be sent before elections.

Forecast: the third-quarter plan will not be fulfilled to 70% or better, and auctions of fixed-coupon issues will not have resumed before 31 October 2026 – the date that is both: after the elections and after publication of the borrowing plan for the 4^th quarter.

Another interesting fact that is strongly adjacent to the one above: H1’26 federal debt servicing is 10.3% of budget revenues: at the first glance, it simply looks better than planned, but it is better than planned because the market is closed: placements that do not happen do not create coupon obligations. It signals the opposite reading of both signals: this indicator improves as the backbone situation underneath it deteriorates. Of course, if the value is measured against GDP rather than revenue (as it is usually done), that understates the burden 3-4 times, but the payment can be made only from revenue not from GDP. And there are multiple cases when with higher debt-to-GDP the burden-to-revenue is much lower than Russia has.

Another important aspect: while the federal budget is one accounting perimeter; the obligation lies across the entire system. Across the federal centre, regions, funds, systemically important state-owned companies, the defence industry and the social sector, the total requirement for 2026 amounts to RUB 16.1 trillion (USD 196 billion): 40% of federal budget revenues and about 96% of everything budgeted for defence and domestic security combined. A region that cannot pay a contractor; a defence plant whose bank refuses to roll over working capital; an unpaid salary: these are three different paths (and evolving simultaneously) to the same failure, and all of them are financed from the same rubles and for all of them obligation falls on the state (to finance them no matter what). Such requirement has been met so far. What the panel measures is by what means and at whose expense.

The particularities of the situation discussed above are some of the reasons why the panel at https://incidence.world was created. And this Incidence Notes channel will contain: 1) panel updates (at approximately monthly cadence), 2) verification of the author’s forecasts (upon checkpoint dates), 3) important updates to the ladders progress (for 4 important ladders of the panel), and notes on rapidly evolving mechanisms (one post about a rapidly evolving mechanism accelerated yesterday by Putin’s decree will be published in a day or two).

The next update of the Incidence panel is planned to be done and released after the 20 September elections. A separate post with a verification of the Forecast above will be done after 31 October, and regardless of the outcome. Findings that contradict forecasts and arguments will be published first, this rule applies both to the panel and to this resource.

The panel: incidence.world. Version 1.9, data as of 21 August 2026, in English, Ukrainian and Russian.
CC BY 4.0 · DOI 10.5281/zenodo.22068421