Jiang Xueqin YouTube: Where to Watch His Lectures Online
Jiang Xueqin's lectures are available online in three main places: his YouTube channel, Predictive History; his Substack of the same name; and the by-video index on this tracker, which maps each source lecture to the predictions extracted from it. If you want the primary video feed, start with YouTube. If you want written posts and archives, use Substack. If you want to see which lecture produced which call, use the by-video index.
The Predictive History YouTube channel
The main hub is the Predictive History channel on YouTube. That is where Jiang posts his long-form lectures on geopolitics, history, and forecasting. The channel is the original source for most of the material tracked here. As of 2026-09-26, the tracker has logged 181 source lectures, 180 of which are transcribed. Those lectures have produced 414 tracked predictions.
The videos are not short clips. They are lecture-style monologues, often running well past an hour, and they are not organized as a news broadcast. Jiang will spend the first part of a lecture on historical framing, then move into specific calls. That format is why the tracker exists: the predictions are in there, but they are not timestamped or summarized by the channel itself. If you are looking for a specific call — say, the US-Iran war sequence — you can search the channel or use the by-video index to find the relevant lecture.
Substack and written archives
Jiang also publishes on Substack under Predictive History. The Substack is useful if you prefer text, want to skim arguments, or want to share a specific piece without sending someone a two-hour video. It is not a one-to-one transcript of every lecture, but it covers many of the same themes and gives you a written record of his forecasting framework.
For researchers, the combination matters. YouTube gives you the full lecture, including the offhand predictions and the caveats that do not make it into a written post. Substack gives you the cleaner argument. The tracker uses both, but the prediction IDs are keyed to the lectures, not the Substack posts.
The by-video index on this tracker
The by-video index is the most practical way to watch Jiang's lectures if you are following a specific prediction. It lists the source lectures in the system and connects them to the predictions extracted from each one. So if you want to see the lecture behind a confirmed call — for example, P001, the call that Trump would win the 2024 presidential election, or P011, the call that the United States would go to war with Iran — the index tells you which video to watch.
That is different from a simple YouTube search. On YouTube, you get the upload date and the title. On the by-video index, you get the prediction IDs, the category, the timeline, and the resolution status. It is a research tool layered on top of the video archive.
What the tracker has found in those lectures
As of 2026-09-26, the tracker has recorded 414 predictions from 181 source lectures. Of those, 22 have resolved: 17 confirmed and 5 wrong. The headline accuracy is 77%. That number comes with a large caveat: 22 resolved calls is a small denominator, and 368 predictions are still pending or unverifiable. The 77% will move as those calls resolve. It is not a stable long-run batting average.
The confirmed calls include some of Jiang's most specific geopolitical forecasts. P001 (Trump wins 2024) resolved after the election. P011 (US goes to war with Iran) and P018 (the war lasts many weeks, possibly many years) were confirmed as the conflict unfolded. P026 (Iran closes the Strait of Hormuz) was confirmed when Iran closed the strait. P088 (decapitation strike kills Ayatollah Khamenei) was confirmed when news sources reported the strike. P315 (Putin visits China shortly after Trump leaves China) was confirmed in May 2026. The wrong calls include P002a, the prediction that Trump would pick Nikki Haley as his vice president; he picked JD Vance instead, which Jiang had also tracked as a strong alternative (P002b).
That mix is why the lectures are worth watching directly rather than only reading the scoreboard. The tracker records the hits and the misses. The lectures show the reasoning. If you want to evaluate the reasoning, you need the video. If you want to check the record, you need the index.
How to start
- For the full lecture feed: go to the Predictive History channel on YouTube and sort by upload date.
- For written posts: go to the Predictive History Substack.
- For prediction-by-prediction sourcing: use the by-video index.
- For the live scoreboard: the tracker's main page lists every prediction ID, status, and timeline.
The lectures are free to watch. The tracker is independent and mirrors news sources; it does not produce the predictions. When a prediction is marked confirmed, that means a news source reported the event, not that the tracker verified it through Jiang's channel. The by-video index is the bridge: it takes you from the scoreboard back to the lecture where the call was made.
Browse all tracked predictions · Accuracy analysis · All blog posts