Who is TheValueist? The record behind "results never lie"
By Maya Koeva · August 31, 2026 · 5 min read

The bio on @TheValueist carries three words that read like a dare: "Results never lie." We agree. Quantral has tracked the account since May 5, 2026, and in under four months it has produced 4,949 stock posts in our coverage, more than any other account we track. We grade those calls against what prices actually did. This page covers who the account is, what it posts, and what the graded results say.
Who they are
TheValueist is pseudonymous, and this profile stays within what the account states publicly. The X profile dates to September 2024 and counts 37.2K followers across 18.4K posts. The bio describes a discretionary long/short approach focused on tech and energy, and the feed matches it: dense, framework-heavy threads on AI infrastructure and semiconductors on one side, and energy, from grid power to the refining complex, on the other.
Beyond the feed, the account runs Atlas Peak Research, a research site with 94 reports pitched at an institutional reader: IPO frameworks for SpaceX, Quantinuum, and Cerebras, post-earnings debriefs on names like GlobalFoundries and Vishay, and cross-asset notes on everything from TPU roadmaps to ERCOT power filings. It coins its own frameworks too; the flagship is the CRAVE Thesis (Cost Reduction Value Expansion), an argument that collapsing AI inference costs expand the market rather than shrink it. A recurring format is the audit post: the account takes a circulating claim, an Oracle backlog number, a SemiAnalysis chip estimate, and cross-checks it against filings and disclosures in public. An account that grades other people's claims should have no objection to being graded itself.
What they post, in our data
Since May 5 the account has averaged more than 1,200 tracked stock posts a month. About two thirds carry a clear direction, and those lean six to one bullish. The most-posted names, with each one's bullish-to-bearish split among the directional posts:
| Ticker | Posts | Bullish : bearish |
|---|---|---|
| NVDA (Nvidia) | 473 | 267 : 31 |
| MU (Micron) | 409 | 223 : 25 |
| SNDK (SanDisk) | 343 | 180 : 21 |
| LITE (Lumentum) | 337 | 179 : 13 |
| VLO (Valero) | 118 | 47 : 9 |
| PSX (Phillips 66) | 113 | 43 : 9 |
| DINO (HF Sinclair) | 111 | 43 : 8 |
| MPC (Marathon Petroleum) | 110 | 42 : 8 |
| GOOG (Alphabet) | 91 | 41 : 16 |
| INTC (Intel) | 72 | 35 : 15 |
Two things stand out. First, the top of the table is one trade expressed four ways: Nvidia plus the memory and optics names that ride the same AI buildout. Second, the middle of the table is a basket nobody else in our coverage touches. The four refiners sit within eight posts of each other with near-identical splits, one energy thesis posted as a unit, and within the accounts we track, TheValueist effectively is the refiner coverage: of the 115 tracked HF Sinclair mentions in our whole dataset, 111 come from this one account. That is the "TMT+Energy" bio made literal, and worth knowing when a refiner's mention count looks like a crowd: it is one voice.
The two-sided tells are Intel and Alphabet, the only top names where the bearish side runs above one in three. Even a six-to-one bull posts with more doubt where the story is contested.
The graded record
Volume is not a track record. On Quantral's six-month accuracy board, @TheValueist ranked sixth of every account we track: 71 graded calls, 54.9% right, with a best graded call of MRVL at +52% in seven days. If 71 sounds small next to 4,949 posts, that is the methodology working as intended: repeated posts on the same name inside a grading window collapse into a single call, so posting a thesis fifty times earns exactly one grade. A firehose gets no volume discount, and no volume bonus either.
The rest of the board posts a fraction of the volume: @jukan05 graded 58.6% on 70 calls, @aleabitoreddit 59.5% on 274, and the board's number one, @citrini, 64.2% on 81. TheValueist's 54.9% sits a few points behind, on a similar sample. Posting ten times more than everyone else bought the account a wider audience, not a better hit rate. And the honest context is the same one every profile in this series ends on: the best graded account we track is right about 64% of the time, so the gap from TheValueist to the top of the board is about one call in ten. Nobody in the set escapes the error bar, which is exactly why the number is worth measuring.
The open bets
A hit rate is history. The account's current convictions are live, and right now the biggest ones are being tested. The AI-memory and optics complex it has backed all summer sold off hard in the second half of August: SanDisk (180:21 in the account's directional posts) closed August 28 about 17% below its August 17 high, a drawdown the wider trusted room is also riding, Micron (223:25) sat about 8% below, and Lumentum (179:13) gave back its recovery in Friday's optics selloff. On the other side, the account's heaviest name paid: its Nvidia lean sat on the right side of the August 26 print, though our own three-week tape grade on that one is still open. The refiner basket has no grade yet from us at all. Some of these bets will resolve for the account and some against it, and the point of grading is that we will not have to argue about which.
Follow the calls, graded
"Results never lie" is a good motto. It is also a testable one. Their voice page has the profile and the graded highlights, and the live record, updated in the Quantral app as each call resolves, sits next to every other account we track, including the rest of the accuracy board. Check the record before you check the follower count.
Graded-call figures are Quantral's own methodology (subject calls graded over fixed windows; see how a track record is graded) as published on the board linked above. Post counts and splits cover the account's tracked stock posts from May 5 to August 31, 2026; price references use closing prices through August 28, 2026. Claims attributed to the account are its own, linked to their sources. Quantral surfaces signals and context from public sources to support your own research. Nothing here is financial advice or a recommendation to buy, sell, or follow anyone. Past performance is not indicative of future results.