Mon-T Weekly Review — w/e 22 May 2026

There is a version of this week where I open by celebrating the desk's best crude oil call in months. There is another version where I linger on platinum's 6.7% collapse against a bullish call, or gold dropping 4.7% while the desk sat on its hands with a NO CALL. I am going to try to do all three, because that is what happened, and the honest scorecard contains all of it.

Six of eight directional calls correct, a 75% hit rate that matches the prior week's ungraded estimate and represents the desk's most consistent recent stretch. Eight directional calls at an average confidence of 6 shows measured commitment rather than the whispered 5.25 of the May 1 hibernation period, and seven NO CALL markets demonstrate the agents still know when to fold their arms. Crude oil, the desk's longest-running headache since the Iran war began in February, was called BEARISH at 5/10 and dropped 6.84%. After months of bearish calls getting torched by Hormuz headlines, the ceasefire timeline finally bent in the desk's favour. That single call was worth the price of admission.

But the misses cut deep. Platinum was called BULLISH at 6/10 ahead of the WPIC Platinum Quarterly report on May 18, and the metal promptly fell 6.7% from $2,068 to $1,930. The Russell 2000, after weeks of bullish glory that I praised in every review, was called BEARISH at 6/10 and rallied 2.54% to prove the desk's bearish pivot was premature. And gold, where the desk issued NO CALL despite a mild bullish lean in the synthesis, crashed 4.68%. I said last week that the CPI-driven gold thesis needed resolution. It resolved, and the desk was not in the room.

Eight directional calls this week, with six landing on the right side. The other seven markets got the NO CALL treatment. A 75% directional hit rate is solid, and the confidence calibration was clean: the two highest-conviction calls at 7/10 were Wheat BULLISH (+1.77%, correct) and Soybeans BULLISH (+0.36%, correct). The two misses came from 6/10 calls, Platinum BULLISH and Russell 2000 BEARISH, which is exactly how you want the distribution to work. Your high-conviction calls should outperform your moderate ones, and this week they did.

The NO CALL markets tell a more complicated story. Gold dropped 4.68% on a NO CALL, which the desk will feel given its own synthesis showed a mild bullish lean. Sterling rallied 0.91% and the yen weakened 0.70%, both misses by the NO CALL scoring framework. Copper bounced 1.93% after a brutal breakdown the desk correctly identified but chose not to trade. When four of seven NO CALL markets produce moves above the noise threshold, the framework is protecting the accuracy percentage at the cost of opportunity. That trade-off is defensible. It is not always comfortable.

The rolling twelve-week figure sits at 57.6% across 99 directional calls, with 96 no-call abstentions. That engagement split tells you the desk is calling direction on roughly half of all market-weeks, a rate that has fallen steadily since the 70%+ pace of February when the precious metals thesis was printing money. This week's 75% accuracy on eight calls nudges the number in the right direction, and the December-January catastrophes have now fully aged out of the window. If the desk can maintain 70%+ accuracy while increasing its directional volume to ten or twelve calls per week, we should finally see the rolling figure crack through 60% on a sustained basis. The 57-58% range has been home for too long.

The Economic agent had the strongest week across the directional calls, correctly supporting the bearish crude oil thesis through demand destruction and OPEC downgrade frameworks, driving the bearish bond call that landed cleanly, and providing the macro backdrop for the silver bearish stance. Its identification of April CPI at 3.4% as a regime-shifting catalyst that removed Fed easing hopes proved to be the most useful analytical lens. The Fundamental agent had a mixed week: its agricultural work drove correct BULLISH calls on wheat and soybeans through the WASDE production shortfall and renewable diesel demand theses respectively, but its BULLISH conviction on platinum's structural deficit was overwhelmed by the very macro forces the Economic agent warned about.

The weakest collective performance came from the equity disciplines on the Russell 2000 bearish call, where four of six agents showed bearish leans following the May 15 breakdown but small caps promptly rallied 2.54% to prove the thesis premature. The desk's observation that RSI at 33.73 was 'oversold without bullish divergence' turned out to be exactly the kind of oversold condition that triggers bounces rather than continuations. After weeks of correctly riding the RTY rally, the desk flipped bearish at the worst possible moment. The Sentiment agent, which remained neutral on RTY this week, was the only voice not actively contributing to the wrong call.

June is catalyst season. The June 5 ECB meeting will determine whether the euro's twelve-week NO CALL hibernation finally ends with a directional view. The June 16-17 FOMC, Kevin Warsh's first as Chair, is the macro event of the quarter, with markets pricing zero cuts and the new dot plot potentially reshaping the rate trajectory that has dominated every asset class since March. The June 18 BoE follows immediately after. For commodities, the June 10 WASDE update on wheat will test whether the worst US production shortfall since 1972 gets worse or stabilises. Crude oil's ceasefire timeline per the EIA projects Strait normalisation in late May to early June, which means the geopolitical premium that has defined 2026 markets could finally collapse in the coming fortnight. The desk will have its Sunday views. I suspect the NO CALL count decreases as the catalyst density increases.