All positions below are paper trades only, run to build a live track record before any real capital is committed.
High-yield OAS sits at 2.74%, down 42 basis points over the past three months — a regime that rewards carry and punishes hedges. That compression is a double-edged signal: spreads this tight historically precede either a long grind higher or a sharp reversal, and the distribution is not symmetric. For BDC investors, the practical implication is that marks hold until they don't, and filing lags become landmines.
The BDC sector is broadly bid alongside credit, but the aggregate masks a widening dispersion in coverage ratios and PIK composition. Our cash-coverage alerts fired on ten names this week — that's not noise. TPVG's cash coverage is essentially zero (0.001x on a 34.3% PIK share); ICMB sits at 0.243x with 13.7% PIK. Neither is a new story, but the cluster of alerts against a benign spread environment is a regime-divergence worth tracking. When OAS eventually reprices, portfolios with this much accrual income will be first to see realized NII erosion.
SCM has not filed its latest 10-Q; the most recent data in our system is stale as of 2025-03-31. That filing gap is the trade. Across 29 borrowers SCM shares with peers that have filed, marks have deteriorated materially: HV Watterson Term Loan moved from 0.99 to 0.58, AD Net from 1.00 to 0.77, Red S All Natural Centeotl from 0.99 to 0.47. When those re-marks propagate into SCM's NAV at filing, our model implies approximately -85bps of NAV downside. The bull case: SCM's managers made idiosyncratic credit decisions that outperformed peers on these exact names — possible, but the base rate is against it. This is the NIM Signal 1 lead-lag strategy, explicitly unvalidated at scale; size is small and deliberate.
PSBD trades at 0.786x NAV as of March 31, 2026 — a 21.4-point discount to book. The pnav_book_divergence alert is flagging something more specific: book dropped 302bps quarter-over-quarter while the price-to-NAV ratio improved by 5.9 points. That combination means the market is getting more comfortable with a portfolio that is objectively deteriorating. Either the market knows something about stabilization that the marks don't yet reflect, or this is a valuation trap. We don't have enough enriched borrower data on PSBD to resolve that question this week, so no trade — but we are watching the next filing closely for whether the mark deterioration is concentrated or broad-based.
Long OBDC was opened from the human watchlist this week. The structural thesis is straightforward: OBDC is one of the few large-cap BDCs with genuine portfolio scale ($18B+), consistent first-lien concentration, and a Blue Owl fee structure that doesn't penalize shareholders on the income line. The alert that gives us pause: the debt-to-equity conversion on Gloves DBA Protective Industrial Products ($7.5M face value) is a completed restructuring signature — debt wiped, equity received. That's a realized loss event. It is small relative to the portfolio, but it is the kind of signal our scorecard tracks. The debt-to-equity conversion signal currently shows a 0% hit rate across 4 events at 60 days, meaning the market has not yet punished holders. We are long, but we are watching the borrower stress feed on OBDC's portfolio more closely than usual.
OCSL at 0.778x NAV with a 163bps quarter-over-quarter book decline and a price-to-NAV that improved 6.4 points looks like the market is front-running a stabilization narrative. We are not buying that narrative yet. Cash coverage is not in our alert set for OCSL this week, but the trajectory of book value — down meaningfully for at least two consecutive quarters — combined with a portfolio that has seen ongoing credit migration makes this a name where the discount is earning itself rather than representing opportunity. The thesis would reverse if OCSL's next filing shows stabilized marks and net origination outpacing repayments; we will revisit.
BBDC was opened long from the watchlist, but the cash-coverage alert complicates the picture: 0.966x total coverage looks adequate until you strip PIK — at 15.2% of the book, cash coverage drops to 0.819x. That is not a crisis, but it is a meaningful spread between what BBDC reports and what it actually collects in cash. The risk is dividend sustainability if base rates fall faster than expected and PIK positions fail to convert. What would prove us wrong: PIK share rising above 20% in the next filing, or any indication that PIK positions are concentrated in the same borrowers flagging stress at peers.
The regime_long strategy was passed both days this week — OAS at 2.74% with a three-month decline of 42bps is exactly the environment where backtest v4 shows the long-quality leg generates noise, not alpha. The v3 borrower-stress event channel also returned nothing tradeable; ten days of silence on enriched tickers is a signal in itself, and manufacturing a trade there would be the kind of drift that destroys a track record. FSK was shorted from the watchlist without a published primary thesis in the data — that position will be reviewed next week before we formalize it in this letter.
The filing calendar for Q2 2026 BDC reports begins in August; that is when the stale-mark thesis on SCM either validates or dissolves, and when the PIK-heavy names on our watch list have to show their hand.
Screening research, not investment advice. All data from public SEC filings.