The Math Behind the Final Stretch - Part 1

Starting conditions in a market priced beyond history

A few weeks ago, I published a framework outlining how I see this market cycle ending. Click Here if you missed it

Not tomorrow.

Not next week.

But soon enough that pretending otherwise is starting to feel analytically dishonest.

This post is the follow-through I promised.

I’ve sat with it longer than I expected.

No charts for shock value.

No timing theatrics.

No hero narratives.

No calling sixty of the last two crashes like Michael Burry.

Just the math.

The boring, uncomfortable kind that usually ends cycles.

Think of this as guardrails.

A map of what happens if nothing changes.

If earnings normalize.

If credit stops pretending nothing can go wrong.

If valuations behave the way they always have when optimism peaks.

Stories start cycles.

People and leverage stretch them.

Math eventually closes them.

I. Starting Conditions (January 2026 Baseline)

Let’s bring this into reality before anyone accuses me of manipulating data.

Market Level & Valuation ( as of close January 23rd)

• S&P 500: 6,946

• Trailing EPS: 221.4

• Trailing P/E: 31.05

• Shiller CAPE: 40.63

• Forward P/E (12m): 22.16

• Forward EPS 310.2

• Implied forward earnings growth: 40%

• Price-to-Book: 5.61

• Price-to-Sales: 3.45

• Dividend Yield: 1.14%

Every one of these sits above the 95th percentile of history, back to 1905.

That’s where we are.

Not just elevated.

Not simply rich.

Statistically extreme by any measure.

If this were housing, people would already be whispering the iconic B word.

“Bubble”

People don’t like that word.

Markets don’t really care.

That brings us to earnings and margins.

Earnings & Margins

• Q4 2025 YoY earnings growth: +8.3%

• CY 2026 EPS estimate: $311.30

• Net profit margin (S&P 500): 12.8%

• After-tax corporate profit share: 10.9% (2nd-highest on record)

• Labor share of GDP: 53.8% (lowest since 1947)

Margins and overall earnings sit at historic extremes while labor share collapses.

That combination defines a classic late-cycle configuration.

History has never been kind to this setup.

Credit & Liquidity

• High-yield spread: 2.69%

• IG OAS: 0.73%

• AA OAS: 0.44%

• SOFR: 3.64%

• Total repo exposure: $12.6T (record)

• Fed repo injections (7 months): $420B+

• Money market assets: $7.8T (record)

• M2 YoY growth: +4.3%

• M2 level: $26.7T (record)

Credit is priced for no defaults, no margin pressure, and no policy mistakes.

Absolute perfection.

And for anyone who hasn’t lived through this before:

When the credit market is priced for perfection.

Run.

Labor & Consumers (Stress Beneath the Surface)

• Private payrolls ex-healthcare: -300k in 8 months

• Temporary help employment: lowest since 2020

• Layoffs (2025): 1.2M (+58% YoY)

• % consumers expecting to miss a payment: 15.3% (near 2020 highs)

• Employment-population ratio: ~60% (downtrend)

Labor deterioration is late-cycle and broadening, not isolated.

When healthcare is carrying labor growth, something underneath is already breaking.

Executives worry about balance sheets.

Households worry about rent, heat, and groceries.

Those two conversations rarely stay separate for long.

At this point, the recession call becomes a matter of timing.

While labor stress builds underneath, market risk concentrates at the top.

Concentration

•Top S&P 500 weights ≥3%: 29% of index

•VIX futures positioning: net short (historically extreme)

•NVIDIA alone: ~7% of S&P 500 weighting

Wealth concentration now rivals levels last seen in the late 1920s.

Different decade.

Same mechanics.

II. Earnings Scenarios

Historical earnings drawdowns outside systemic banking collapses cluster tightly. Three scenarios are modeled.

Scenario A: Soft Earnings Recession

• Earnings decline: -10%

• EPS: $201

Typical of 1990, 2018, 2011.

Scenario B: Base Case (Cycle Reset)

• Earnings decline: -20%

• EPS: $179

Occurs with:

• 2000–2002 (ex-bubble sectors)

• 1973–74 (real terms)

• 1987–1991 margin resets

Scenario C: Credit-Stress Tail

• Earnings decline: -30%

• EPS: $156

This outcome requires stress, but not a complete catastrophe.

This occurs when:

• Credit spreads normalize above 6%

• Capex pulls back

• Labor softens meaningfully

• Policy reacts late but aggressively

Importantly, this scenario does not require:

A systemic banking collapse

or

A 1929-style financial implosion

It simply reflects what happens when leverage unwinds faster than earnings expectations adjust.

Why This Matters

Notice what’s missing here:

heroics, guesswork, or tail-event obsession.

These scenarios don’t rely on panic or magic.

They rely on math, margins, and credit doing what they’ve always done once cycles turn from assets to liabilities.

The debate should not be whether earnings fall in general.

It should focus on how far normalization is allowed to run before policy steps in and how much damage is done before that moment arrives.

---

III. Valuation Regimes (What Multiples Do Late-Cycle)

Valuations reset in stages as leverage slowly fades, earnings confidence begins to weaken, and policy credibility starts to erode.

Multiples usually contract well before earnings hit their low point and before credit losses show up in the data.

The catalyst tends to end up being mechanical rather than emotional.

Leverage pulls back.

Duration gets repriced.

Risk quietly becomes expensive again.

They rarely wander back to normal at the end of a cycle.

Overwhelmingly history shows they exaggerate to the downside just as they exaggerate today to the upside.

The current market sits in a very late-stage optimism regime.

As of January 2026, the numbers look like this:

• Trailing P/E: 31.05

• Shiller CAPE: 40.63

• Forward P/E: 22.16

• Price to sales: 3.45

• Price to book: 5.61

• Market cap to GDP: 222.94%

• Market cap to M2: 306%

• Dividend yield: 1.14%

Across every major valuation framework, the market stands well beyond even historical bubble thresholds.

When cycles unwind from these levels, valuation compression tends to follow a familiar sequence.

First comes the optimism peak.

Concentration builds, leverage expands, and narratives do most of the explanatory work.

• That’s where we are today.

Then credit stress emerges, often between 14 and 18 times earnings, when spreads widen and refinancing risk becomes visible.

• That’s what’s in store for the second half of this year.

Finally, balance sheet repair takes over in the 12 to 15 times range, marked by a series of forced deleveraging and full earnings resets.

• We will hit this in the back half of 2027 or the first half of 2028.

Reversion almost never pauses neatly at long-term averages.

It tends to overshoot.

Earnings estimates trail reality. Credit losses surface late.

Early policy easing often adds volatility rather than restoring confidence.

Start with today’s elevated multiples and record index concentration. Add weakening labor internals and credit markets priced for near perfection.

Put those together, and the probability-weighted valuation floor in this cycle clusters around 13 to 15 times earnings, reflecting historical overshoot following simultaneous valuation, margin, and credit extremes.

Periods of stress can push multiples temporarily lower than average.

That range aligns with historical outcomes following valuation peaks of this scale.

At this point in the cycle it’s important to point out that when earnings, margins, and credit all sit beyond the 95th percentile simultaneously, markets do not find a stable equilibrium.

Prices must fall, earnings must fall, or risk must reprice. In every prior historical instance, all three adjusted.

The only open question is not whether normalization occurs, but exact date, how much is absorbed before policy responds and how much damage is done before that response arrives.

Markets can delay arithmetic. They cannot avoid it. Translating these constraints into price outcomes is not a matter of prediction.

It is a matter of math.

That comes next.