Schiff vs Saylor vs the Banks: Who Got 2026 Right?

Why Short-Term Price Windows Can't Score Long-Term Theses

A prediction scorecard is a retrospective accuracy audit that ranks public forecasters against realized market outcomes over a fixed window. In finance media, these scorecards compare figures like Peter Schiff, Michael Saylor, and major bank strategists against price targets they issued for a specific asset and date. The format assumes short-term price movement validates or invalidates long-term theses, an assumption that collapses under scrutiny.

Key Findings

• As of August 26-27, 2026, gold trades at $4,617-$4,647 and Bitcoin at roughly $79,000, gold up 16.9% year-to-date, Bitcoin down 23.4%, a divergence both assets have not shown simultaneously since 2011.
• Peter Schiff's 2026 gold-over-Bitcoin call is scoring correctly on a nine-month window, but the same forecaster has made 22 public "Bitcoin is dead" calls since 2011, according to the tracker bitcoindeaths.com, a 0-for-22 lifetime record on timing.
• Michael Saylor declared on August 3, 2026, "I have never sold mine. Not one Satoshi", the same day Strategy's SEC filing disclosed the company's third 2026 sale of 1,638 BTC at an average $63,957, below its roughly $75,400 cost basis.
• Standard Chartered's Geoff Kendrick has cut his year-end Bitcoin target twice since December 2025 ($300,000 to $150,000 to $100,000), then flagged on August 19-20, 2026 that $100,000 might now be "too low" after a Treasury bond-buyback announcement.
• The scorecard format itself is the story: it converts noisy, short-horizon price action into a simple verdict about who was "right," obscuring that Schiff, Saylor, and the banks are answering fundamentally different questions on fundamentally different timescales.

Thesis Declaration

The "who got 2026 right" scorecard is not a measurement of predictive skill, it is a format that converts short-term volatility into claims of long-term vindication or condemnation. This matters because retail capital allocation decisions increasingly follow personality track records rather than structural analysis, and the scorecard format is structurally incapable of distinguishing a correct thesis from a lucky window.

The Evidence: What the Numbers Actually Show

The Board tracks public forecasts and scores them on accuracy and calibration; this is the first monthly scorecard in that series. Start with the raw scoreboard, because the numbers are unambiguous even when their meaning is not. Bitcoin sits near $79,000 as of the August 26-27, 2026 dateline, down 23.4% year-to-date. Gold sits at $4,617-$4,647, up 16.9% over the same period. Ethereum trades around $2,450-$2,470. Silver sits near $69. This is, by the framing used across the tracked calls, the first time since 2011 that gold and Bitcoin have moved in opposite directions this sharply and simultaneously, a genuine anomaly, and the reason the scorecard genre has attention velocity right now.

But velocity is not validity. Consider each camp on its own terms.

Peter Schiff has been calling for Bitcoin's collapse in favor of gold since 2011. The bitcoindeaths.com tracker counts 22 separate public "Bitcoin is dead" declarations, the first when Bitcoin traded near $17. His most recent calls, made in early 2026 when Bitcoin was trading in the $60,000-$65,000 range, warned of an 85% crash. On August 20, 2026, he called Bitcoin's move above $72,000 a "fakeout, not a breakout," adding that gold is "the real winner." Ten days earlier, on August 10, as gold topped $4,400, he told followers to "sell MSTR and sell Bitcoin now!" Scored narrowly against the 2026 calendar year, Schiff is right: gold has outperformed Bitcoin by roughly 40 percentage points year-to-date. Scored against his 15-year public record, he is 0-for-22 on timing the actual death he keeps predicting.

Michael Saylor built Strategy's entire corporate identity around the phrase "never sell." His August 3, 2026 statement, "When I say 'Never Sell Your Bitcoin,' I speak as one saver to another. I have never sold mine. Not one Satoshi", was a personal pledge issued on the same calendar day that Strategy's own SEC filing disclosed its third Bitcoin sale of the year: 1,638 BTC transacted between July 27 and August 2 at an average price of $63,957, below the company's roughly $75,400 cost basis. Total 2026 sales across the three transactions reach 5,258 BTC, worth approximately $323 million, all executed below cost. The personal pledge is technically intact, Saylor did not sell his own coins. The corporate pledge, which is what actually moves markets and what retail holders associated with "never sell," has now been broken three times in eight months.

Standard Chartered's Geoff Kendrick offers the clearest case of forecast-as-lagging-indicator. His year-end Bitcoin target has moved from $300,000 in December 2025, to $150,000, to $100,000 by February 2026, two downward revisions in ten weeks, on top of a track record that includes three straight missed year-end Bitcoin calls historically. Then, on August 19-20, 2026, after the U.S. Treasury announced it would double long-dated bond buybacks, Kendrick reversed direction: "for the first time this year there is now a risk my end-year forecast [of $100,000] is too low." With Bitcoin at roughly $79,000 and four months left in the year, that target requires a 26% rally to hit even the lowered number.

Goldman Sachs cut its year-end gold target from $5,400 to $4,900 on June 20, 2026, citing a hawkish Federal Reserve, interest rate cuts pushed into 2027, and fading ETF demand, a call made while gold was down 8% for the month. Gold has since recovered to $4,617-$4,647, roughly $250-$280 below the revised target with four months remaining. The direction of Goldman's revised call may still prove correct; the timing, cutting into a local low right before a recovery, was poor.

Tom Lee requires more careful handling than the other four, because two of his claims are frequently conflated in casual coverage. At Consensus Miami, Lee projected Bitcoin at $150,000-$200,000 and Ethereum at $9,000-$12,000 by year-end 2026. Separately, at Proof of Talk in Paris in June 2026, he floated Ethereum at $250,000 as a long-term, no-deadline scenario tied to corporate validators and tokenization. These are not the same claim on different timelines, they are parallel claims with different resolution conditions, and treating the $250,000 figure as a "walked-back" version of the $9,000-$12,000 range misrepresents his position. What is verifiable: by August 2026, Lee was reframing the $9,000-$10,000 Ethereum level as arriving "easily" in 2027 or 2028 rather than by year-end 2026, while Ethereum trades at $2,450-$2,470, 75-80% below even the low end of his original year-end range.

Jim Rickards stands out for reaffirming rather than revising. On April 15, 2026, he restated: "$10,000 is still a realistic estimate," referring to gold before the end of 2026. Gold's current $4,620 level requires a 116% rally in four months to hit that target.

$323M, Total value of Strategy's below-cost-basis Bitcoin sales in 2026, disclosed the same year Saylor publicly reaffirmed "never sell."

The Scoreboard

Case Study: The Same-Day Contradiction

On August 3, 2026, Michael Saylor posted a message reaffirming Strategy's signature doctrine: "When I say 'Never Sell Your Bitcoin,' I speak as one saver to another. I have never sold mine. Not one Satoshi." The statement was framed as personal testimony, distinct from corporate treasury policy, and it was technically accurate, Saylor's personal holdings were untouched. But Strategy's own regulatory disclosure, filed the same day, told a different story: the company had sold 1,638 BTC between July 27 and August 2, 2026, at an average price of $63,957 per coin, below its roughly $75,400 average cost basis. This was the third such sale of the year, bringing 2026 total disposals to 5,258 BTC worth approximately $323 million, all sold at a loss relative to acquisition cost. The split between the individual pledge and the corporate action, disclosed on the identical calendar day, is the cleanest data point in the entire scorecard: it shows precisely where rhetoric and fiduciary reality diverge, and which one moves capital.

An Original Framework: The Horizon Laundering Matrix

Financial punditry scorecards fail for a specific, reusable reason: they measure forecasters against a single time axis while forecasters actually operate across two independent axes, claim horizon (how long until the thesis is supposed to resolve) and disclosure visibility (how publicly the position is tracked). Cross these two axes and you get four quadrants that explain every distortion in the current Schiff-Saylor-banks discourse.

Quadrant 1, Short Horizon, High Visibility. Schiff's "fakeout not breakout" call and Kendrick's year-end Bitcoin target both live here. These are the only claims a nine-month scorecard can legitimately grade, because the resolution date and the scoring date are close together.

Quadrant 2, Long Horizon, High Visibility. Schiff's 15-year "Bitcoin is dead" thesis and Rickards' standing $10,000 gold call live here. These claims are publicly falsifiable but on a timescale that a single-year scorecard cannot resolve, grading them against 2026 alone is a category error, akin to judging a chess opening by the board position after three moves.

Quadrant 3, Short Horizon, Low Visibility. This is where "the banks" actually live. Goldman's and Standard Chartered's public year-end targets are visible, but their internal prop-desk and principal-risk positioning, the capital actually deployed against the public call, is not disclosed. A bank's public target can be wrong while its actual book profits, or vice versa; the scorecard only ever sees the public-facing number.

Quadrant 4, Long Horizon, Low Visibility. Saylor's corporate treasury strategy sits here. The "never sell" pledge is a Quadrant-2-style long-horizon public claim, but the actual capital allocation decisions, 5,258 BTC sold in 2026 alone, are disclosed only through SEC filings that most scorecard consumers never read. The personal pledge (Quadrant 2) gets the headline; the corporate reality (Quadrant 4) gets a footnote.

The Horizon Laundering Matrix explains why every 2026 scorecard reads as tribal validation rather than analysis: commentators are grading Quadrant 2 and Quadrant 4 claims using Quadrant 1 methodology, and the mismatch is not a bug in any individual writer's judgment, it is structural to the format itself.

Predictions and Outlook

PREDICTION [1/4]: Bitcoin will close December 31, 2026 below Standard Chartered's revised $100,000 target, meaning Kendrick's twice-cut forecast will mark a fourth consecutive missed year-end call (65% confidence, timeframe: December 31, 2026).

PREDICTION [2/4]: Gold will close December 31, 2026 below $6,000, meaning Jim Rickards' reaffirmed $10,000 target will go unmet for the year he specified (65% confidence, timeframe: December 31, 2026).

PREDICTION [3/4]: Ethereum will close December 31, 2026 below $5,000, meaning Tom Lee's Consensus Miami year-end range of $9,000-$12,000 will be missed and the resolution date will be publicly reframed to 2027 or later before the year closes (68% confidence, timeframe: December 31, 2026).

PREDICTION [4/4]: Strategy will disclose at least one additional below-cost-basis Bitcoin sale in an SEC filing before year-end 2026, while Michael Saylor's personal "never sell" statements continue uninterrupted (65% confidence, timeframe: December 31, 2026).

What to Watch

• Whether Standard Chartered issues a fourth target revision before December 2026, a reversal back downward would confirm the trend-following pattern rather than genuine risk repricing.
• Whether financial media outlets report Strategy's next 8-K filing with the same prominence given to Saylor's public pledges, asymmetric coverage is itself a measurable signal of narrative capture.
• Whether Schiff's 0-for-22 lifetime record on bitcoindeaths.com gets cited alongside his 2026 "win" in year-end retrospectives, or whether the base rate is quietly dropped from the narrative.
• Whether gold and Bitcoin's simultaneous divergent performance, unprecedented since 2011 by the tracked framing, persists through Q4 2026 or mean-reverts, which would retroactively make the entire 2026 scorecard look like noise rather than signal.

Historical Analog: Revisions That Track the Tape

Standard Chartered's consecutive downward revisions on Bitcoin, followed by an abrupt upward-risk flag after a single Treasury announcement, resemble a pattern often associated with sell-side equity analysts during the 1999-2002 dot-com collapse, who issued serially shrinking price targets on falling tech stocks, targets that moved with price rather than ahead of it. Kendrick's pattern, $300,000 to $150,000 to $100,000, then a flag that $100,000 might be "too low" the moment Treasury policy shifted, reads as the same lagging-indicator behavior dressed as forward-looking analysis. Separately, Schiff's 22-call, 15-year "Bitcoin is dead" campaign echoes the broader pattern of permabear commentators who maintain a standing thesis for decades and count any temporary price alignment as vindication.

Counter-Thesis: Maybe the Scorecard Still Does Something Useful

The strongest objection to this critique is that scorecards, however crude, are the only accountability mechanism retail audiences have against forecasters who otherwise face zero cost for being wrong. Schiff's 22 failed Bitcoin obituaries would carry no reputational weight at all if no one, anywhere, kept a running count, bitcoindeaths.com exists precisely because informal accountability failed. Similarly, without journalists cross-referencing Saylor's August 3 pledge against the same-day SEC filing, the corporate sale would have passed with far less scrutiny; the contradiction only became visible because someone applied scorecard logic to it. A scorecard that grades a nine-month window is imperfect, but a world with no scorecard at all lets every forecaster claim a permanent, unfalsifiable long horizon, Schiff's "eventually I'll be right" defense is itself a Quadrant-2 evasion that scorecards, whatever their flaws, at least force into the open. The honest position is not that scorecards are worthless, but that they are only valid when scored against the specific horizon and disclosure quadrant a claim actually occupies, which is exactly the discipline the current "who got 2026 right" genre refuses to apply.

Stakeholder Implications

For regulators and policymakers: The SEC should require corporate treasury disclosures, like Strategy's Bitcoin sales, to be cross-referenced against contemporaneous public statements by controlling executives within the same filing cycle, not left for journalists to discover independently. The gap between Saylor's August 3 pledge and the same-day 8-K is a disclosure-timing issue that a modest rule change (mandatory same-week correlation flagging for executive public statements versus treasury transactions) would close without restricting speech.

For investors and capital allocators: Do not allocate based on any single-year scorecard result. Score forecasters against the specific horizon and quadrant their claim occupies, a 15-year permabear thesis cannot be validated or invalidated by nine months of price action, and a corporate "never sell" pledge should be weighted against actual SEC-disclosed transactions, not personal social media statements. Track the base rate (0-for-22, repeated target revisions) as the primary signal, not the most recent headline.

For financial media and industry operators: Publish base-rate context in the same paragraph as any "win," not in a separate follow-up piece. If a scorecard credits Schiff's 2026 gold call, the 0-for-22 lifetime record belongs in the same sentence, not a footnote. If Saylor's pledge is quoted, the same-day filing belongs in the same paragraph. The credibility of the scorecard genre depends entirely on whether outlets apply this discipline symmetrically across bullish and bearish figures.

Frequently Asked Questions

Q: Who predicted Bitcoin's 2026 drop? A: Peter Schiff's early-2026 calls, made when Bitcoin traded between $60,000 and $65,000, warned of an 85% crash and correctly anticipated relative underperformance against gold; Bitcoin has since fallen 23.4% year-to-date while gold rose 16.9% as of the August 26-27, 2026 dateline. Standard Chartered's Geoff Kendrick also cut his Bitcoin target twice over the same period, though his revisions tracked the decline rather than anticipating it.

Q: Is Peter Schiff right about gold in 2026? A: On the narrow 2026 calendar-year comparison, yes, gold has outperformed Bitcoin by roughly 40 percentage points year-to-date, and Schiff's August 10 and August 20 calls favoring gold over Bitcoin and MicroStrategy have held up so far. His lifetime record is separate: bitcoindeaths.com tracks 22 public "Bitcoin is dead" calls since 2011, none of which correctly timed a permanent collapse.

Q: Did Michael Saylor sell Bitcoin in 2026? A: Saylor personally has not sold his own Bitcoin holdings, consistent with his public pledges. Strategy, the company he leads, disclosed three separate 2026 sales totaling 5,258 BTC (approximately $323 million), all executed below the company's roughly $75,400 cost basis, with the most recent disclosed the same day Saylor stated "I have never sold mine. Not one Satoshi."

Q: What did Tom Lee predict for Ethereum in 2026? A: At Consensus Miami, Lee projected Ethereum reaching $9,000-$12,000 by year-end 2026, alongside a Bitcoin target of $150,000-$200,000. Separately, at Proof of Talk in Paris in June 2026, he floated a long-term, no-deadline scenario of $250,000 tied to corporate validator adoption and tokenization, a different claim, not a revision of the year-end figure. By August 2026, Lee was reframing the $9,000-$10,000 level as arriving in 2027 or 2028, with Ethereum trading around $2,450-$2,470.

Q: Why do bank price targets keep getting revised? A: Goldman Sachs cut its year-end gold target from $5,400 to $4,900 on June 20, 2026, during a monthly decline, only to see gold recover close to the original level within weeks. Standard Chartered cut its Bitcoin target twice before flagging in August that the lowered figure might now be too conservative. Both patterns suggest targets are being revised in reaction to recent price movement rather than ahead of it, a lagging-indicator problem, not a forecasting one.

New calls tracked this month

Two fresh, dated, falsifiable calls enter the tracker as pending predictions:

1. Standard Chartered (Kendrick), August 19-20, 2026: first upward-risk flag of the year on the $100,000 year-end Bitcoin call, triggered by the Treasury buyback announcement. Resolves December 31, 2026.
2. Peter Schiff, August 20, 2026: Bitcoin's move above $72,000 is a "fakeout, not a breakout," made explicitly against gold outperformance. Falsifiable within weeks.

Synthesis

Nobody "won" 2026, a nine-month window resolved a subset of claims that were never designed to be judged on that timescale, while the claims that were built for short horizons (Kendrick's target, Schiff's fakeout call) remain open until December 31. Schiff's gold call and Saylor's same-day contradiction are both real, verifiable, and simultaneously true, which is precisely the point: the scorecard format cannot hold two accurate but incompatible facts about the same person without collapsing into tribal shorthand. The only durable takeaway is procedural, not directional, grade every forecast against its own stated horizon and disclosure quadrant, or the scoreboard measures nothing but who posted loudest this week.