Ukraine Military History Institute
Against a missile that can change trajectory, one interceptor is a guess
Rob Lee's most useful point in this part of the conversation is not about how many interceptors Ukraine has. It is about how many it can afford to fire at a single missile.
The Iskander's 9M723 can fly either a ballistic trajectory or a more quasi-ballistic one, and the defender does not know in advance which it will be. The MIM-104 Patriot has to guess. If you can only launch one interceptor, Lee says, that guess is often going to be a 50% chance of hitting or less.
His comparison is to American practice. If the United States were defending against this threat, it would launch multiple interceptors per missile. Ukraine does not have that option. The difference is not in the quality of the system. It is in the stock behind it.
The missile is hard even when the interceptors exist
Lee is explicit that the problem does not disappear when supply improves. The 9M723 is, in his words, an effective missile that is difficult to intercept even when Ukraine has enough Patriot.
He adds a second, less discussed threat: the 48N6U, the ground-attack version of the S-400 missile. Because it was built as an air defence missile, it manoeuvres in an unusual way, and that on its own confuses ballistic missile defence systems.
So the defender faces two distinct problems at once. One missile hides which path it will take. The other does not fly the way a ballistic threat is expected to fly.
Every alternative on the list is either weak, slow or unproven
Lee walks through the options, and none of them comes out well.
Ukraine has SAMP/T, but as far as he knows the system has not had much success against ballistic missiles, and he believes there are big issues with interceptor quantities as well.
Ukraine has struck the factories that produce Iskander components. Lee reports mixed views from the people he has spoken to. Some think it may be an option. Others think it is not effective at all, because Russia can simply produce elsewhere. He says plainly that he does not know, and that what it does show is that Ukraine is trying to do something.
Firepoint says it is coming with an option next year. Lee's response is "we'll see." There has been talk of licensed production of Patriot in Ukraine, but it now seems it will not be agreed until the winter, and even then it is a long-term measure.
His overall verdict is a single sentence: there is no good option right now. Ballistic missile defence is difficult, sophisticated and expensive to develop, and there is no way of doing it cheaply.
The backdrop he repeats is supply. Ukraine lacks Patriot, and the United States has worked through a large share of its own, around 65% by the figure he recalled reporting the previous week.
Ukraine knows where the launchers are, and that is not enough
Dmytro Putiata then says something that sounds like good news and is not.
Ukraine knows the places where Russia keeps its launchers. In most cases they are in forest areas, with underground shelters. He says Ukraine knows roughly where they all are, and even how many launchers sit in a given part of a given forest.
The problem is timing. To destroy a launcher you need to catch it while it is standing and trying to launch. Striking the forest on spec does not solve it. You could, he says, launch 200 drones against that forest and maybe be effective. Maybe. He does not know.
This is the concrete version of the idea Lee credits to Cook, going after the archer rather than the arrow. Knowing the archer's address does not help if he is only exposed for the moment he draws.
The admission about Ukrainian technology
Putiata is unusually candid about the limit on Ukraine's side, and ties it to a point he says he made earlier in the episode: Ukraine is not as technologically developed in some areas as it would like to show everyone.
His specific example is range. Without Starlink, Ukraine's technology lets it strike, when operated, out to about 100 kilometres. Lee had placed the launchers anywhere within about 500 kilometres of the border. "We have a lot of things to fix," he says, "and it's not gonna be that easy."
Coming from a Ukrainian analyst, that is not defeatism. It explains why the launcher problem is so much harder than the fight closer to the line.
Moscow is hit less often than Kyiv, and Lee explains why
Samuel Cook raises a question he heard that day: would you rather live in Moscow or in Kyiv? Lee's answer is immediate. Moscow is going to be hit less often.
Cook credits Ukraine's successes on the offensive side: the logistics campaign, the oil refineries, drones flying ever longer distances, and Flamingo missiles going through a number of iterations with rising production. Lee accepts that it is effective, and then describes its limits.
Ukraine's deep strike drones navigate by a combination of internal navigation and GPS or GNSS guidance, and that is not precise. He cites a video from the previous week in which a deep strike drone, possibly an FP-1, though he was not sure which type, missed a large factory by about 100 metres. These are not really precision-guided munitions, and Russia is jamming and defending. Most of the drones are defeated before the target. He puts the figure at probably 15% to 30%, in context the share that reaches the target.
That matters most against targets like refineries, where you need to hit a very specific place. With manual guidance to the end you can. With ordinary guidance, maybe you hit it, maybe you do not.
What to watch
- Firepoint's ballistic defence option. Lee noted the company says it will arrive next year. Whether a working system is shown in 2027 is a direct test of that claim.
- The licensed Patriot production agreement. Lee expected no agreement before the winter. An agreement signed earlier, or no agreement by spring, would each tell you something.
- The success rate of deep strike drones. Lee's estimate was 15% to 30% reaching the target. Any independent tally of launches against confirmed hits can be checked against that range.