Mon-T Weekly Review — w/e 3 Jul 2026

Two stories, same desk, opposite outcomes. Treasury bonds, this week's Market of the Week and the desk's first-ever MOTW for the Full Desk tier's flagship fixed income product, were called BEARISH at 5/10 conviction and duly fell 1.51% as Kevin Warsh appeared at the ECB Forum in Sintra on July 1 telling the world 'prices are too high.' CNBC confirmed Treasury yields rose on the day, and the bond market obliged the desk's thesis about dangerous complacency in the MOVE index by repricing lower through Friday's close at 112.3125. A clean win on the desk's longest-running directional conviction of 2026.

Then there is silver. Called BEARISH at 7/10, the desk's highest conviction of the week and the seventh attempt to extend a streak that had delivered cumulative downside of roughly 24% since late May, the metal rebounded 6.14% from $59.18 to $62.81. CNBC reported gold was set for its first weekly rise in a month as investors scaled back Fed rate hike bets, and silver rode that same tailwind. The six-week bearish winning streak I have been celebrating in this column for over a month is finished, and it ended with the kind of move that makes you wonder whether the desk held on one week too many.

Four directional calls this week from thirteen markets, two correct. Fifty percent accuracy at an average confidence of 5.3. The nine NO CALL markets included the S&P rallying 2.1%, the Nasdaq surging 2.06%, gold bouncing 2.22%, and copper gaining 1.37%, all of which the desk watched from the sideline. If I sound like a broken record on the NO CALL theme, it is because the record keeps playing the same track.

Four directional calls this week, with two landing on the right side. The other nine markets got the NO CALL treatment. A 50% directional accuracy rate is the worst since the all-bearish coin-flip week of late March, and it sits uncomfortably at the threshold where paying attention to the desk offers no measurable advantage over guessing. The average confidence of 5.3, the lowest possible for directional calls, tells you the desk was barely whispering its views.

The confidence calibration is inverted in the most painful way possible. Silver at 7/10, the desk's strongest conviction, was the week's worst miss at +6.14% in the wrong direction. The other three calls sat at minimum confidence of 5/10, and two of those three landed correctly. When your highest conviction produces your worst result and your lowest produces your best, the system is telling you something uncomfortable about its own self-assessment. The ZB bearish call at 5/10 delivered the cleanest result at -1.51%, while crude oil's BEARISH at 4/10 scraped a technically correct -0.12%. Direction right, magnitude negligible.

The rolling twelve-week figure sits at 53.8% across 78 directional calls, with 98 no-call abstentions. That engagement split means the desk calls direction on fewer than half of all market-weeks, a rate that has been declining steadily since February's 70%+ pace. The number has slipped below the 55% line that had felt like a permanent floor for months. This week's 50% on four calls does nothing to help, and the wartime volatility weeks from April continue to exert gravitational pull. The desk's directional volume has fallen to a level where the rolling number barely moves from week to week, regardless of accuracy. At some point, you need to make enough calls for the accuracy to matter.

The Economic agent had a genuinely split week that exposes the desk's coordination problem at its starkest. On bonds, it was the star, driving the correct BEARISH call through its identification of Warsh's hawkish posture and the MOVE complacency signal. On silver, that same agent's framework about sustained real yields above 2.20% was the backbone of the bearish call that missed by 6.14%. The thesis was the same across both markets: hawkish Fed creates headwinds for duration and non-yielding assets. On ZB, the market agreed. On silver, the market decided the selloff had exhausted itself and that mean reversion from deeply oversold levels mattered more than the macro narrative.

The Fundamental agent, meanwhile, continues its year-long pattern of being overridden and then vindicated too late. On silver, it called BULLISH on the sixth-year structural deficit, was overridden by the bearish synthesis, and would have been closer to correct. On wheat, it called BULLISH on the 1972 production shortfall, was overridden to NO CALL by the miss reset, and the market moved only 0.3%. The agent that is right about medium-term fundamentals and wrong about timing remains the desk's most frustrating discipline.

The calendar thins dramatically into mid-July, with no FOMC until July 29-30 and the next major data catalyst being June NFP on July 5. The July 10 WASDE will determine whether wheat's drought thesis gets fresh validation or further dismissal. Crude oil at $68.78, nearly flat on the week, sits in a no-man's land between the IEA demand destruction narrative and J.P. Morgan's $60 Brent structural fair value floor. Silver's 6% rebound raises the question of whether the desk's bearish thesis has run its course or whether this was a dead cat bounce in a market that has lost half its value since January. Gold bounced 2.22% while the desk sat on its hands through a nine-miss mandatory reset. The S&P rallied 2.1%. The Nasdaq gained 2.06%. Both were NO CALL. The summer lull is here, and the desk needs to decide whether its chronic NO CALL posture is appropriate caution or learned helplessness. I will be watching.