The Missing 94,475 Contracts: What Last Week's CFTC Treasury Print Really Says

The ledger does not lie, only the narrative does. The CFTC's Commitments of Traders report for the week ended August 4 showed speculators cutting net short positions in CBOT U.S. Treasury futures by 41,225 contracts. The headline writes itself: bearishness receding. Then I did the arithmetic. It doesn't close. The disclosed line items show a two-year net short cut of 120,346 contracts and a five-year net short build of 179,319 contracts. The ultra-long net short fell by 5,723 contracts. Sum the disclosed changes: minus 120,346 plus 179,319 minus 5,723. That equals a net short increase of 53,250 contracts on the listed maturities. Not a decrease. The only way to reach the 41,225-contract decrease is to assume roughly 94,475 contracts of net short covering happened somewhere in products the statement did not name. The market's most important Treasury futures contract is the ten-year. It is missing from this print. That is not an editorial footnote. It is the story. The CFTC releases a Commitments of Traders snapshot every Friday, reflecting positions held through the previous Tuesday. I have gone through enough of these files to respect their structure. They separate commercial hedgers from non-commercial speculators. Asset managers, CTAs, hedge funds: the fast crowd. Their net positions are not a prediction. They are a footprint. Short contracts are the language of bearish expression. When speculators are net short, they are selling future exposure to a rate. When a net short shrinks, the seller steps back. When it grows, the seller presses. The release does not carry a year in the shared summary. That should make any reader cautious. A two-year short cover in the middle of a hiking cycle carries one meaning. The same cover inside a cutting cycle carries another. I will work with the data as given, week ended August 4, and let the next report supply the missing context. Now look at the maturity structure. The two-year Treasury futures contract is the most sensitive to Federal Reserve policy expectations. If a trader wants to bet that the tightening cycle ends, the two-year is the vehicle. The five-year lives at the intersection of growth and inflation. It is the duration that the market uses to price the middle of the next policy sequence. The ultra-long, the 30-year, is a referendum on fiscal sustainability, term premium, and long-run inflation expectations. When these three move in different directions, the market is not saying one thing. It is saying a sequence. That sequence is the real content of the report. The two-year short cover tells you the Fed stop is coming into view. The five-year short build tells you the landing will not be clean. The ultra-long tells you the long end is not where the fear is. Start with the two-year. Net short interest fell by 120,346 contracts. Big number. But remember the absolute level. The two-year net short position has been enormous for months. A 120,000-contract retreat from an extreme is a withdrawal, not a reversal. The position remains net short by a margin that would have seemed aggressive in any normal cycle. What does that mean? The most crowded bearish trade in the rates market is being quietly unwound. The people who were paid to be certain about higher-for-longer are now reducing exposure. It is not an embrace of lower rates. It is an admission that the price of being wrong is now higher than the price of being right. Now the five-year. The net short position expanded by 179,319 contracts. This is the largest single move in the report. It sits in direct opposition to the two-year. The market is not betting that rates will stay high forever. It is betting that the middle of the curve will stay sticky while the front end crumbles. The five-year is the bond market's version of a contested border. It is where the inflation path and the growth path argue. Adding short exposure there signals that the market doubts a smooth return to two percent. It says the first round of Fed easing will be bracketed by a long, flat, stubborn stretch of core inflation. That is a defensible position. It is also a very different position from the one that mainstream headlines will distill. Then the accounting hole. Do the bookkeeping again. The stated total net short change is a decrease of 41,225 contracts. The stated line items sum to an increase of 53,250 contracts. The residual is a decrease of 94,475 contracts. That residual has to live somewhere in the unmentioned products. The most likely destination is the ten-year note futures contract. The ten-year is the reference point for mortgages, corporate credit, government duration, and every macro portfolio on earth. A 94,475-contract short cover in the ten-year would be a serious piece of information. It would mean the bearish consensus is breaking beyond the front end. But the report as summarized does not name it. So we are left with two possible systems. In the first system, the ten-year was covered alongside the two-year. The broad speculative short base is shrinking. That is an early-cycle read. It would add confidence to the idea that the policy peak is in place, and the whole curve is repricing for the next cut. In the second system, the ten-year net short was actually increased. The disclosed residual comes from other maturities, and the curve trade is a spread, not a directional bet. Under that reading, the market is not becoming less bearish. It is expressing bearishness in a more surgical place. The two-year is no longer the weapon of choice. The five-year and perhaps the ten-year are. I have spent too many hours reconstructing failed systems to accept an unfilled summary. In the 2018 ICO cycle, I watched projects publish audit reports that omitted the one function that could drain the treasury. In 2022, I traced the Terra collapse transaction by transaction, and the moment the mint/burn numbers stopped summing was the moment the system was already dead. The same principle applies here. A data release that does not reconcile is not a rounding issue. It is a selection issue. Someone chose what to show. The missing ten-year is the tell. What is the market doing? Let me build the position. A trader wanting to express a steepener buys short-dated bonds or reduces short-dated shorts, and sells longer-dated bonds or adds longer-dated shorts. The two-year short cover and the five-year short build fit that structure perfectly. It is a 2s5s steepener. That is the strongest signal in the entire report. The market is not making a single bet on the level of rates. It is making a bet on the shape of rates. It is pricing the end of the hiking cycle and an uncomfortable middle. There are two flavors of steepener. A bull steepener happens when short-term rates fall faster than long-term rates. A bear steepener happens when long-term rates rise faster than short-term rates. The CFTC data, as far as it goes, points toward the bull version. The ultra-long short cover of 5,723 contracts is small but telling. It suggests the long end is not under attack. If the trade were a bear steepener driven by fiscal supply fears, the 30-year short base would be expanding, not contracting. It is contracting. That bakes in a very specific narrative: short rates will come down, long rates will stay anchored, and the middle will absorb the pain. That is a plausible mix before a policy pivot. It is not a mix that screams recession. A recession signal would show shorts collapsing across every maturity as speculative capital runs for the exit. That is not what this print shows. Now translate to the inflation argument. A two-year short cover is the market's way of saying the next three CPI prints will not force another hike. A five-year short build is the market's way of saying the second half of disinflation will not be as easy. An ultra-long short cover says long-run inflation expectations are not breaking. Put it together. The market is pricing a softish landing. Not soft. Not hard. Sticky in the middle. That is exactly the positioning I would expect to see one or two macro data releases before the Fed openly admits the cycle has ended. Cross-market implications follow from the curve trade. Two-year rates heading lower gives a discount-rate tailwind to long-duration growth equities. Five-year rates staying high means the economic cycle is not strong enough to lift cyclical value stocks. That combination favors growth over value. Gold benefits from lower real short-end rates, but the five-year short build puts a ceiling on any upside. The dollar loses some carry support at the short end, but the five-year floor prevents a violent decline. None of this is a hero forecast. It is a repricing of timing. The market is saying: the next move is a pause, not a crash, and the next recession is a problem for another quarter. Now the contrarian angle. The bulls are not entirely wrong. They are just early and they may be reading the wrong half of the table. The short covering in the two-year is a legitimate signal. When the most crowded bearish trade starts to unwind, it is often the first visible crack in the old consensus. That crack can widen. Long-duration assets, including growth equities and gold, should benefit as the front end reprices. The bulls have correctly identified that the policy peak is closer than the market was willing to admit a month ago. That part is real. What they get wrong is the five-year. The largest single move in the report is an expansion of bearishness at the exact point where the inflation war will be won or lost. You cannot read the two-year cover and ignore the five-year build. The market is telling you that the last mile of inflation is still contested. The policy rate may have peaked, but the neutral rate may settle at a level that keeps the middle of the curve heavy for longer than the equity market would like. That is a caution flag, not an all-clear. I have learned not to mistake a spread trade for a directional prayer. A curve steepener is often the market's way of admitting uncertainty while still booking a premium for taking a side. It is a defense mechanism. You can be short the front end and long the middle at the same time because the policy path is not clear. It does not mean conviction. It means the trader wants to get paid for carrying the knot. Panic is just poor data processing in real-time. But complacency is worse. Complacency reads a two-year short cover and ignores the five-year short build. Complacency sees the headline 'net shorts cut' and does not ask why the line items do not sum. Complacency treats a missing ten-year as a data hiccup instead of a deliberate omission. The bulls have a real edge: they are paying attention to the right asset duration. But they are ignoring the middle of the curve at their own risk. Next Friday's CFTC release will settle the debate. If the two-year short continues to shrink and the ten-year shows a large net short cover, the old bearish consensus is breaking. The pivot trade will have legs. If the ten-year net short expands, then last week's print was not a pivot. It was a pair trade. It was a signal that the market expects the front end to ease and the middle to stay tight. The ledger does not lie, only the narrative does. The narrative says dovish. The ledger says divided. The two-year says I was too short. The five-year says I am not convinced. The missing ten-year says the story is not final. I will keep my equation simple. Emotion is a variable I exclude. The next data print is a better indicator than any opinion. Watch the ten-year. Watch the residual. Structure outlives sentiment, and the ledger is the code of the market. The synthesis will come from the next report, not from the headlines.