Is Jiang Xueqin Credible? His Track Record Says Mostly Yes

The number-driven answer is yes — with an honest caveat about sample size. As of 2026-08-01, this independent tracker has extracted 372 distinct predictions from Jiang Xueqin's lectures and resolved 22 of them against real-world news. Of those 22, 16 were confirmed correct and 6 proved wrong, for a headline accuracy of 73%. That figure is strong, but it is built on a thin base: 333 predictions are still pending or set aside as unverifiable. The credibility argument is not a single percentage. It is that every call is timestamped, sourced from his recorded lectures, and open to inspection — including the ones he got wrong.

The call that put him on the map

Jiang's most visible confirmed forecast was his April 2024 prediction (P001) that Donald Trump would win the 2024 presidential election — committed to video seven months before the vote, when most mainstream models were hedging. He paired it with a sharp secondary read: a May 2024 call (P002b) that JD Vance was a strong vice-presidential pick for Trump, well before Vance sat on any conventional shortlist. Both landed.

But the deepest cluster of confirmed calls is the one he staked his name on most loudly: a US-Iran war. Starting in April 2024 (P011), he argued the United States would go to war with Iran within two to four years — a forecast many dismissed as alarmist. When the conflict erupted in early 2026, a striking run of his specific sub-calls followed it: that the US would bomb Iranian nuclear facilities (P012), that "shock and awe" would fail to end the war quickly (P015), that the fighting would drag on for weeks or years rather than days (P018), that Iran would close the Strait of Hormuz (P026) and strike Gulf states (P027), and that the Iranian navy would be destroyed (P105). His prediction that a decapitation strike would kill Ayatollah Khamenei (P088) was confirmed when major news organizations reported the strike around March 1, 2026.

He also got the harder, negative calls right. He argued nuclear weapons would not be used in the Iran war (P016), and that China would not meaningfully intervene on Iran's behalf (P071). Both held. Outside the Middle East, his December 2025 call that the US would move against Venezuela (P082) and his May 2026 forecast that AI companies would ship companion products to drive engagement (P129) both resolved his way.

Where he was wrong

Credibility is built on showing the misses, not hiding them. Two of Jiang's calls are now scored wrong. In May 2024 he predicted Trump would pick Nikki Haley as his running mate (P002a); Trump chose Vance instead — the very pick Jiang had simultaneously gotten right under P002b, making the miss notably self-canceling. More substantively, he argued that Vladimir Putin would declare a nuclear umbrella over Iran before or at the outbreak of war (P017), which he believed would deter US nuclear use. That did not happen. The full list of misses is on the wrong predictions page, and the running score is on the accuracy dashboard.

The honest caveat: a small denominator

Here is what keeps that percentage from being the final word. Of 372 tracked predictions, 333 are still waiting on events that have not happened, and only 22 have fully resolved. A track record scored across 22 calls is suggestive, not conclusive — a couple of late reversals would move the number meaningfully. The Iran cluster in particular is a set of tightly related calls that resolved in the same direction at the same moment. Statistically, it should be read as one large bet on a war, not as a dozen independent forecasts. Judging Jiang Xueqin reliable means weighting that clustering honestly rather than counting headlines.

The receipts are the argument

The case for his PredictiveHistory channel being credible does not rest on charisma or hindsight framing. It rests on the fact that these were committed predictions, dated and extracted from 176 source lectures (174 transcribed), before the events occurred — not recollections reshaped after the fact. You can check the prediction IDs, the dates, and the confirming reporting yourself rather than take a highlight reel on faith. Six predictions already sit in the wrong column, and the system is built to add to that column rather than bury it.

The fuller judgment is still accruing. With 333 predictions outstanding, the track record is genuinely live, and the next stretch of geopolitics will tell us whether the current rate is a floor or a ceiling. For now the evidence says Jiang Xueqin is more right than wrong — and serious about letting you verify it.

Browse all tracked predictions · Accuracy analysis · All blog posts