Jiang Xueqin YouTube: Where to Watch His Lectures Online

Jiang Xueqin's lectures are free to watch, and they live in three places: his Predictive History channel on YouTube, where each lecture is posted in full; his Substack, where the same material appears in written form; and the by-video index on this site, which maps every tracked prediction back to the specific lecture it came from. If you only want one link, start with the index — it is the fastest route from a claim to the video that made it.

The primary source: YouTube

Everything begins on the Predictive History channel. Lectures typically run between thirty minutes and an hour, and they are delivered unscripted, which is part of both their appeal and their difficulty. Jiang tends to build an argument through historical analogy — imperial overstretch, debt cycles, the recurring mechanics of hegemonic war — and then convert that analogy into dated, falsifiable calls about the near future. The channel is organized chronologically, not thematically, so a lecture from April 2024 that predicted a US war with Iran (P011, P012) sits hundreds of videos away from the 2026 lectures where those calls were resolved.

That structure is why the jiang xueqin youtube archive is easiest to use with a map. Search the channel for a topic and you will get fragments; search our index and you get the full chain.

Substack and written companions

Jiang also publishes in text. His Substack newsletter carries essays and written versions of arguments that appear in the lectures, and it is the better format if you want to quote him precisely rather than paraphrase from audio. The written posts are not a complete mirror of the video catalog — some lectures never appear there — so treat the Substack as a supplement rather than a replacement.

A practical workflow: watch the lecture on YouTube for the delivery and the question-and-answer passages, then check the Substack and our index for the exact wording of any prediction you intend to hold him to.

The by-video index

Our by-video index lists each source lecture alongside the predictions extracted from it, with dates and current status. As of 2026-10-17 the system holds 183 source lectures, 182 of them transcribed, feeding 450 tracked predictions. If you want to hear Jiang argue that the US would bomb Iranian nuclear facilities, or that Iran would close the Strait of Hormuz (P026), the index takes you to the video rather than asking you to scrub through a catalog.

What the archive currently shows

The numbers, as of 2026-10-17: 450 predictions tracked, 22 resolved, 403 still pending, 17 confirmed and 5 wrong — a headline accuracy of 77% among resolved calls.

That figure deserves an explicit caveat. Twenty-two resolved predictions is a small denominator, and several come from a single cluster: the US-Iran war. When one large event confirms many related calls at once — the strikes (P012), the failure of the shock-and-awe approach (P015), the escalation against civilian infrastructure (P020), the decapitation strike that killed Ayatollah Khamenei (P088) — the accuracy rate moves in a lump rather than a trickle. A single event breaking the other way would shift it just as sharply.

The confirmed list is not only about Iran. P001 was confirmed when the Associated Press called the 2024 presidential race for Trump, and P002b correctly identified JD Vance as a strong vice-presidential option two months before the convention. Others were confirmed after US strikes on Venezuela were reported (P082) and after Washington opened trade fronts against Canada, Mexico and the European Union (P325). The misses are documented at the same granularity: P002a, the Nikki Haley vice-presidential call, is recorded as wrong, not quietly dropped.

The bulk of the catalog — 403 predictions — remains unresolved. Most carry timelines that stretch into 2027 and beyond, which means the accuracy rate will keep moving for years regardless of how the next few months go.

How to watch efficiently

• Start with the index, not the channel. Find the prediction you care about, then open the linked lecture.
• Watch at 1.25x or 1.5x. The arguments are dense but rarely rapid.
• Note the date on the video. A call made in April 2024 and a call made in March 2026 are different claims, even when the wording matches.
• Check whether a prediction has already resolved before treating it as a forecast. Roughly one in twenty tracked calls has been settled.

All of it — jiang xueqin lectures online, in video, in text, and in index form — is free. The only cost is attention.

Browse all tracked predictions · Accuracy analysis · All blog posts