Is Jiang Xueqin Credible? His Track Record Says Mostly Yes
Yes—provisionally. As of 2026-09-25, JiangPredictions.com tracks 400 predictions from Jiang Xueqin's lectures. Twenty-two have resolved: 16 confirmed, 6 wrong. That is a headline accuracy of 73% among resolved calls. The catch is that 361 predictions remain pending or unverifiable, so the 73% rests on a small denominator. Credibility, for now, is a pattern, not a verdict. The live numbers are maintained at accuracy.html.
The Iran war cluster is the strongest evidence
The bulk of Jiang's confirmed calls come from a single, interlocking story about a US-Iran war. In April 2024, he predicted the United States would go to war with Iran within two to four years (P011). He also said the US would bomb Iranian nuclear facilities (P012), that "shock and awe" would fail (P015), that nuclear weapons would not be used (P016), and that the conflict would last weeks or years (P018). Later calls added that Iran would close the Strait of Hormuz (P026), attack GCC countries including Dubai, Bahrain, and Qatar (P027), and that Mojtaba Khamenei would become Supreme Leader (P041). He also predicted China would not significantly participate (P071), that America would attack Venezuela (P082), that a decapitation strike would kill Ayatollah Khamenei (P088), and that the US/Israel would destroy the Iranian navy (P105). These were not vague gestures. They were dated, specific, and—when the events occurred—confirmed when the events themselves became news.
That cluster is why "predictive history credible" is not an empty phrase for his defenders. Jiang's method treats geopolitics as a structural system: energy, religion, demographics, and elite factionalism. The Iran calls hang together as a model, not a lucky guess. He also got Trump's 2024 win (P001) and JD Vance as a strong VP alternative (P002b). The Vance call is especially notable because the same lecture cycle included a wrong VP pick.
The misses are real, and they matter
Jiang predicted Trump would pick Nikki Haley as his vice president (P002a). That was wrong. He also predicted Putin would declare a nuclear umbrella over Iran, preventing US nuclear use (P017). That was wrong. The second miss is instructive: P016—that nuclear weapons would not be used—was confirmed, but P017's causal mechanism was not. A forecaster can land the outcome while missing the path. That distinction matters for anyone asking whether Jiang Xueqin is reliable. If you need the mechanism right, the record is more mixed. If you need the broad direction, the Iran cluster is stronger.
The wrong calls are not buried. They are indexed at predictions.html?status=wrong. That transparency is part of the credibility argument: the ledger includes falsifiable misses, not just retrospective wins.
The denominator problem
The biggest caveat is arithmetic. With only 22 resolved predictions, a single new confirmation or refutation moves the headline accuracy by several points. The 73% is not a stable property of Jiang's forecasting; it is a snapshot of a live, growing dataset. Most of the 400 tracked predictions are still pending or unverifiable—361 of them. Some have long horizons. P011 gave a two-to-four-year window from April 2024. P041 gave a five-year window. Others depend on events that may never be cleanly resolvable.
The system also draws on 180 source lectures, 179 of them transcribed. That is a large corpus, but volume of source material is not the same as volume of resolved outcomes. Anyone citing a hard accuracy number should say "as of 2026-09-25" and note the small resolved sample. A 73% hit rate on 22 calls is interesting. It is not statistically decisive.
How to judge him
Credibility is not binary. Jiang looks less like an oracle and more like a structural analyst with a distinctive model and a mixed tactical record. His Iran-war sequence is the strongest evidence that his framework captures something real. His VP and Putin-nuclear misses show that specific predictions can fail even when the broader story holds. For readers, the useful question is not "is Jiang Xueqin credible?" in the abstract, but "which of his claims are dated, falsifiable, and scored?" On that test, the tracker gives a clear answer: many are. The scoreboard is public, the misses are listed, and the pending pile is large enough to keep the question open.
Browse all tracked predictions · Accuracy analysis · All blog posts