Buy More Low (BML)

Objective — deploy leverage counter-cyclically: invest more when markets have fallen and fear is high, rather than at a fixed point in time.

Risk level — 3 — Higher risk. Risk justification — level 3 on the timing-judgement test: the strategy’s benefit depends on identifying when markets are “low”, which is a judgement the evidence does not support most people making reliably. Provisional — the mechanism is not yet specified (see Evidence status), so the tier is assigned on the concept’s shape, not on a defined rule. Benefit justification — unverified — needs Talbot. Jurisdiction — Canada for tax treatment.

Prerequisites

unverified — needs Talbot. Structurally it requires unused borrowing capacity held in reserve, the cash flow to service leverage taken on during a downturn, and the temperament to act when others are not.

Mechanism

unverified — needs Talbot. The mechanism has not been specified in any document available to this library.

What is established:

BML is named SDC intellectual property, listed among $MART DEBT Wealth offerings alongside “Credit-Accelerated Investing” products, and slated for advanced advisor training (“BML Advanced”).
SDC’s own course notes say: “Review all ideas in ‘Buy More Low’ Wealth Accelerator notes” — those notes are not in the vault, and are the missing source.
The concept it is built on is stated: “We simply attempt to be fearful when others are greedy and to be greedy only when others are fearful.” — Warren Buffett, 1986 Berkshire Hathaway annual report.
The data points named for the content: equity fund flows versus market returns, and margin debt balances versus market levels.

That is the honest extent of it. A specific deployment rule — what counts as “low”, how much to deploy, over what period — is not written down anywhere reachable, and this record will not invent one.

Benefits

unverified — needs Talbot. Directionally: deploying leverage after a decline improves the entry point relative to deploying at a fixed date, if the investor can actually do it.

Risks

Timing risk is the strategy. Every mechanism that decides “low” is a timing rule, and the record has no evidence that any specific rule works.
Leverage taken during a downturn is leverage taken when income and employment are also under pressure.
The Gemini strategic review in the vault flags this exact strategy as needing to be “framed around surviving these exact scenarios” and subject to mandatory stress-testing — an internal note, not published evidence, but it is the recorded internal position.

Failure modes

Deploying at a decline that continues, exhausting reserve capacity before the bottom.
Confusing a rule with a judgement — publishing a heuristic that reads as a signal.
Marketing a counter-cyclical strategy at exactly the moment it is emotionally hardest to execute.

Costs

unverified — needs Talbot.

Tax considerations

Interest on borrowing deployed into income-producing property remains deductible under ITA §20(1)(c); nothing about counter-cyclical timing changes that.

Who it may suit

unverified — needs Talbot.

Who should avoid it

Anyone treating this record as implementable. It is a named strategy with a missing specification, published so the gap is visible.

Implementation outline

Not available. Blocked on the “Buy More Low” Wealth Accelerator notes or on Talbot’s own specification.

Evidence status

unverified — named repeatedly across SDC documents as core IP with a stated conceptual basis, but no mechanism, rule, or evidence base is present in the vault. The referenced Wealth Accelerator notes are the blocking source.

This is one of two records in the library at unverified (the other is cash-damming). Both are published as explicit stubs rather than omitted, so the library shows what is missing instead of hiding it.

Counterarguments

Counter-cyclical deployment is market timing under a friendlier name, and the industry’s own consensus is against market timing. The strongest version of the objection, and it deserves a specific answer that this library does not yet have. Talbot’s recorded personal position — “While the financial industry advises against market timing, most in the industry do it in various forms. There appears to be empirical evidence that some momentum strategies reduce short-term risk and enhance long-term returns” — is a starting point for that answer, not the answer, and it is a personal-investing note rather than published SDC evidence.

Variants

“BML Advanced” — named as advanced advisor-training content; contents unverified — needs Talbot.
BML products within $MART DEBT Wealth; not specified.

The Market Drop Wins fold-in — decided 2026-09-22

Market Drop Wins’ T3-4 Buy More Low is this record. No new record is created. But unlike the other four overlaps, this one arrives carrying part of the specification this record was missing, and the partial matters.

What is now settled, from a named source:

Thresholds: −20% and −30%. Derived rather than chosen: −30% occurs once per 12.1 years across 16 bear markets 1929–2026 (once per 13.3 post-war), median depth −31.1%. “Once a decade” interpolates to about −28%, and the source deliberately rounds to −30% because “−28% is false precision on 16 observations.”
The crux, stated against the strategy’s own interest: given a −20% decline, roughly a 50% chance it ever reaches −30% — so half the time the second tranche never deploys.
Tranching does not maximize expected return. It maximizes the probability that you act at all. On expected capital deployed × expected discount, one-shot deployment at −20% wins (100% deployed, 20% average discount) and every tranched design scores lower; 50/50 at −20/−30 deploys 76% at a 23.5% average discount. Roughly four points of efficiency is the price of a plan that executes.
The reserve’s function is to enable the deployment, not to be deployed.
Seven design decisions settled: measured from the prior all-time closing high · the investor’s own domestic index, TSX for Canadians · local currency · accept a threshold never firing · deploy both if both fire quickly · re-arm on recovery to a new all-time high · nominal price index for the trigger, total return for outcome analysis.
The expense buffer is infrastructure, not drag — “an investor who must sell equities to eat during a decline cannot also be buying.”

What is still missing, and why evidence-status stays unverified: tranche sizing is not settled; the figures above are S&P 500 price-index numbers requiring recomputation on a Canadian total-return basis before anything ships; and the source records Buy More Low as a parked sub-project still needing TSX total-return data and LevPro scenarios. The Wealth Accelerator notes named in Open questions below remain the blocking source for BML as SDC’s own IP, as distinct from the generic decision.

The generic form of this decision is now its own record — post-decline-deployment (BM-2) — deliberately, so the general premise can be cited without adopting a particular tranching rule. This record stays the SDC-named tranched form.

Source: market-drop-wins-library-v2 T3-4; market-drop-wins-master-log-v2 §3.3, §3.4, §3.6, §5.2, §6.

post-decline-deployment · non-callable-first · standby-credit-capacity · interest-only-investment-loan · lifecycle-leverage-glide-path · conservative-leverage-ratio

Sources

Smart Debt Offerings — “BML (Buy More Low)”: the Buffett quote, the two named data points, the pointer to the missing Wealth Accelerator notes, and BML Advanced
SDC/Strategy/Identity/Internal View/Assets.md — BML named among SDC assets
Core/_WorkingOn/Later/Investment Strategy Upgrades.md — Talbot’s own recorded position on market timing (personal investing context, not published SDC evidence)
Core/_WorkingOn/Later/StrategicPlanUpgrades/Gemini/Strategic Plan-2.md — internal review flagging mandatory stress-testing for this strategy
market-drop-wins-library-v2 — T3-4, thresholds settled at −20% and −30%
market-drop-wins-master-log-v2 §3.3 (decline frequency and continuation), §3.4 (tranching table), §3.6 (the expense buffer), §5.2 (the seven settled design decisions), §6 (parked, and what it needs on resumption)

Open questions

Where are the “Buy More Low” Wealth Accelerator notes? BML as SDC IP is still blocked on them, even though the generic thresholds are now sourced.
Tranche sizing, which the Market Drop Wins work explicitly did not settle.
Recomputation of every figure on a Canadian total-return basis. The settled thresholds rest on S&P 500 price-index data; the source’s own standing rule forbids shipping those numbers as outcomes. Now unblocked (2026-09-24): a gated monthly S&P/TSX total-return series, 1956–2026 (StatCan to 2001, XIC.TO after, checked against XIC at 0.998 annual correlation), exists in ~/projects/cape-timing (load_canada()), alongside Canadian T-bills and CPI. The recomputation itself has not been run.
Valuation conditioning (MDW §3.1 bound). The AQR critique’s first bound, “re-test on the top valuation quintile”, was tested on 2026-09-24 in cape-timing. From top-5% CAPE starts, US stocks trailed cash over 10 years 55% of the time vs 14% overall, so the critique is materially weaker at extreme valuations, as the log expected. The post-decline side (tranche triggers) was not tested there.
What defines “low”, and on what evidence? — partly answered: −20% and −30% from the prior all-time closing high on the investor’s domestic index. What remains unanswered is whether that rule beats not having one.
Whether BML is a deployment rule, a product, or a course module. The documents use it as all three.