Judo Capital’s 9-insider buy cluster, read against CBA and NAB

JUDO CAPITAL HOLDINGS LIMITED sits in the awkward middle ground that usually produces the better insider reads. It is too small to be treated like a bank index proxy, too visible to be ignored, and just large enough that a director buy can still matter without pretending to be a grand strategic statement. The stock was around A$1.01 on 28 August, with market capitalisation around A$1.13 billion to A$1.14 billion, while the latest filing wave showed directors adding stock after the company had already put FY26 numbers on the table.

That matters because the Australian banking tape has not been handing out easy rewards. Commonwealth Bank posted cash NPAT of A$11.0 billion, up 7%, and return on equity of 14.0% for FY26, yet the big four have still been under pressure as rate cycles mature and margin questions hang over the group. Judo is not playing that game on the same scale. It is a specialist SME lender trying to win on service and speed, which is a very different business from sitting on a deposit franchise and a national branch footprint.

InsiderTrades data puts the relevant bucket, director-level buys at sweet-spot names, at a 52.8% 90-day win rate across 5,805 cases. That is useful context, not a promise. It points to a historical edge in the size band Judo occupies, where information is often less efficiently priced than in the mega-cap banks. It does not say this particular buy will work.

Hornery, Douglas, and the late-August cluster

The filing that catches the eye is the 28 August purchase by David Stephenson Hornery, an operating director who bought about EUR 30,440 of stock, euro-normalised at ingest. That is not a heroic sum in absolute terms, and it is not meant to be. The point is the pattern around it. Hornery was not alone. Jennifer Douglas filed a director buy on 27 August, and earlier on-market buys on 19 and 20 August came from Hornery, Christopher Bayliss, and others. The register has been active, and it has been active in one direction.

Our scoring likes this kind of setup for a reason. It rewards an operating director, a wide cluster, a small or mid-cap name, and a filing value that is tiny relative to market value. Judo checks those boxes. The company also sits in the sweet-spot size band, where insider information has historically been least priced-in. That does not make the trade good by itself. It does make the filing less easy to wave away as routine housekeeping.

The cluster detail matters more than the single ticket. Nine distinct insiders have traded the name in the same direction over the past quarter, with 12 recent declarations in the mix. That is a lot of activity for a lender that is still trying to prove it can scale without losing the discipline that made it interesting in the first place. A lone director buy can be noise. A cluster after results is a different read.

Judo’s FY26 numbers versus the market’s bank obsession

Judo’s FY26 report, released on 18 August, gives the filing some substance to lean on. Profit before tax came in at A$168.1 million, up 34%. Gross loans and advances reached A$14.7 billion, up 18%. Net interest margin was 3.13%, and cost of risk rose as the bank built provisions. Those are not the numbers of a business in distress. They are the numbers of a lender still growing into its niche while keeping an eye on credit quality.

That is where the comparison with Commonwealth Bank and NAB becomes useful. The majors have scale, funding advantages, and a broader earnings base. They also have the burden of being owned by everyone, which is why their shares can get treated like bond proxies one month and growth stocks the next. Judo does not have that problem. It has a narrower lane, and the market tends to judge it on whether SME lending growth can keep outrunning the usual banking trade-offs.

The sector backdrop is not especially forgiving. Australian trimmed-mean inflation held at 3.6% year on year to July, above forecasts, and major banks have shifted toward the possibility of another Reserve Bank of Australia hike, perhaps as soon as September or November. Household spending rose 1.1% in July, which keeps growth alive but also keeps inflation pressure in the frame. For a lender like Judo, that is a mixed bag. Higher rates can support margins for a while, but they also test borrowers and make credit discipline more visible.

Judo’s annual report already flagged slower economic momentum. So the filing does not arrive in a vacuum. It lands after a decent earnings print, in a sector where the big banks have still been punished despite solid results, and in a macro setting where the next policy move is still live. That is a better backdrop for insider buying than a clean expansion cycle with no credit stress. It is also a backdrop where management can be tempted to buy because the stock looks cheap, which is why the rest of the evidence matters.

CBA and NAB own the scale, Judo owns the niche

Commonwealth Bank and National Australia Bank remain the obvious comparables because they define the market Judo is trying to nibble at. CBA’s FY26 result was strong, and NAB has also posted solid growth. Bendigo and Adelaide Bank sits closer to Judo in spirit, a regional player with modest lending expansion and stable dividends, but it does not have the same SME focus. Judo’s pitch is narrower. It is a specialist business bank targeting small and medium enterprises, and that means it competes on relationship depth and turnaround speed rather than on sheer balance-sheet heft.

That difference is why the valuation conversation is never clean. Judo trades at a lower absolute valuation multiple than the majors, but it is also carrying a different risk profile. The market is not paying for a universal bank franchise. It is paying, or refusing to pay, for evidence that the SME model can keep compounding without the usual banking slippage. The FY26 numbers helped. The insider cluster adds another layer. Neither one settles the argument.

Macquarie Group recently ceased as a substantial holder, which changes the institutional register in a way that matters more than a headline buy from a director and less than a full strategic reset. It removes one source of support, at least mechanically. Against that, the cluster of director purchases says the board and management are not waiting for the market to do the work for them. They are putting money in while the stock still sits near A$1.01.

That is where the comparison with the majors gets sharp. CBA and NAB do not need director buying to explain themselves. Judo does. It is a smaller name, a more concentrated model, and a stock where insider behaviour can still carry information content. If you want the cleaner read, you have to compare the filing against the business, not against a generic banking index.

What the insider pattern shows, and where it stops

The historical cohort data is the useful guardrail here. For director-level buys at sweet-spot names, InsiderTrades data shows a 90-day win rate of 52.8% and an average return of 3.68% across 5,805 cases. That is a modest edge, not a magic trick. It says these filings have historically leaned positive in this size band. It does not say Judo will follow the average, and it certainly does not say the stock will move in a straight line over the next quarter.

The strategy framework behind the bucket is also worth keeping in view, but only once. InsiderTrades data uses a 90-day holding window and a maximum position size of 0.08, with live out-of-sample placeholders of 0.81, 26.4, and 51.5 on a restricted EU venue universe. That is a screening frame, not a promise of future returns, and it survives only as long as the regime and the universe stay comparable. For this article, the point is narrower. Judo sits in a category where insider buying has historically had some traction, and this filing fits the pattern.

The fundamental screen is not the story, but it does help explain why the name is in the sweet spot. InsiderTrades data gives Judo a fundamental score of 76, with a value score of 85 and a quality score of 67. Those are transparent screen outputs, not an alpha claim. They line up with a company that is not obviously broken, is still growing, and is not priced like a bank with no execution risk. That is the kind of setup where director buying can matter more than it would in a fully mature franchise.

Credit, margins, and the part the filing cannot fix

The obvious risk is that SME lending is not a one-way trade. Judo’s FY26 report already showed cost of risk rising as the bank built provisions. That is prudent, but it also tells you the credit cycle is not a toy problem. If the RBA does tighten again, borrowers will feel it. If growth slows more sharply than expected, the market will stop caring about loan growth and start caring about arrears, provisioning, and funding discipline.

That is why the filing should be read as management confidence, not as a substitute for the next set of numbers. Hornery’s EUR 30,440 buy is meaningful because it came inside a cluster and after results, not because it can overpower a bad credit turn. The same goes for the earlier director buys. They tell you the board is willing to own the stock at this level. They do not say the next half-year will be easy.

The big-bank comparison keeps that honest. CBA and NAB can absorb more noise because their franchises are broader. Judo has less room for error, which is exactly why the insider cluster matters. When a smaller lender with a focused model sees multiple directors buy after a decent earnings print, the market should at least ask whether the stock has been marked down too far for the quality of the franchise. It should ask that question, then check the credit data.

What to watch while the register stays active

The next test is whether the buying remains clustered or fades into a one-off post-results burst. If more director-level filings follow, the market will have to decide whether this is a board that sees value at current levels or simply a group that likes the optics of buying after a strong year. The distinction matters. One is a view on valuation and execution. The other is theatre.

Watch the lending line first. Judo’s A$14.7 billion gross loans and advances base is the engine, and the 18% growth rate is what makes the story investable. Then watch margins and provisions. A 3.13% net interest margin is respectable, but it is not a shield if credit costs rise faster than expected. The bank has already shown it can grow. The question is whether it can keep doing that without paying too much for the privilege.

The comparison with Commonwealth Bank and NAB will keep framing the stock because the market always compares the challenger with the incumbents, even when the business models are different. Judo is not trying to be CBA. It is trying to prove that a specialist SME lender can earn its multiple. The late-August insider cluster says the board is willing to buy that case at around A$1.01. The next earnings update will tell you whether the market agrees.

This is not investment advice.