Actionable insights — $100,000 Annual Cash Flow (Portfolio Update)

1. Define fair value as a required-return threshold, and publish a buy-below price for every holding

1. Fix the return you require — 10% a year here — and hold it constant across every name and every period.
2. For each holding, solve for the price at which your earnings-growth model delivers exactly that return. That price is the fair value, and it is a decision threshold, not a valuation opinion.
3. Publish or record it before the market moves, so the buy decision is pre-made and the emotional work is done in advance.
4. Express the gap as a percentage so positions are directly comparable across currencies and price levels.
5. Recompute monthly. A fair value that only moves when you want to buy is not a threshold.

• A threshold definition that makes almost everything look cheap after a market fall — nearly the whole book screens undervalued here, which is what a required-return threshold does in a drawdown.
• Two tables disagreeing: Medpace reads -16.6% on one sheet and +5.0% on the other because the exit multiples differ. Pick the operative sheet and say which.

2. Build the portfolio's expected return bottom-up, with no re-rating assumption

1. For each holding, take current-year EPS and a forecast three years out. Convert the total change into an annual rate.
2. Add the dividend yield. That sum is the expected return with the multiple held constant — no assumption about sentiment at all.
3. Weight and sum across the book to get a single portfolio figure.
4. Publish the per-name components so any single forecast can be challenged without invalidating the whole.
5. Treat any multiple expansion as upside on top of this number, never as part of it.

• Consensus three-year EPS forecasts being optimistic in aggregate; the conservatism of the method does not extend to its inputs.
• A portfolio average carried by one or two extreme names — Kelly Partners at 30.5% flatters the total.

3. Track your own portfolio's cash-flow yield as a multi-year time series

1. Each period, compute the aggregate free cash flow of your holdings divided by their aggregate market value.
2. Record it. One reading is meaningless; a decade of readings tells you where your book sits relative to its own history.
3. Use it to answer the question you cannot answer stock by stock: is the portfolio as a whole cheap or expensive versus itself?
4. Pair it with the look-through cash-flow figure, so you can see whether a rising yield comes from more cash or a lower price.
5. Report both at the same cadence so the trend, rather than the level, is the signal.

• Composition changes flattering the series — adding cheaper names raises the yield without anything getting cheaper.
• A rising yield that reflects peak cash flow rather than a falling price.

4. Read breadth from the gap between index volatility and average single-stock volatility

1. Track two measures: expected volatility of the index (VIX) and the average expected volatility of its constituents (VIXEQ).
2. Compute the gap. A wide gap means individual stocks are moving a lot but in opposite directions, so the moves cancel at index level.
3. Interpret it as dispersion: the market is trading company-by-company rather than as a single theme.
4. Corroborate with a sector proxy — if the leading theme's index is falling while the broad index is flat, money is rotating rather than leaving.
5. Treat it as a description of the current regime, not a timing signal.

• Dispersion driven by an earnings season rather than by a regime change; the gap widens every reporting period.
• Confirmation bias — a manager underperforming on one style will find rotation signals attractive.

5. Support the "prices follow fundamentals eventually" claim with someone else's decade, not your own anecdote

1. Find a long, published record that reports both owner's earnings growth and share-price return for the same portfolio, year by year.
2. Split it into sub-periods where the two diverged, in both directions.
3. Show the full-period figures, where they converge. The convergence is the claim; the divergences are its cost.
4. State the duration of the divergences explicitly — this sets a realistic expectation for how long you may be wrong-looking.
5. Prefer third-party data to your own; it cannot be accused of being selected.

• A single manager's record standing in for a general law.
• The comfort of the conclusion obscuring the length of the discomfort — seven years is longer than most people's patience.

6. Give each holding an explicit BUY or HOLD, and let some of them read HOLD

1. Rate every position, every period, using the same threshold you would apply to a new purchase.
2. Be willing to publish HOLD on a name you like. A book where everything is a BUY is not being marked, it is being defended.
3. For each HOLD, state whether the constraint is valuation, position size or thesis — the three have different consequences.
4. Separate the rating from the decision to sell: expensive is a reason not to add, not automatically a reason to exit.
5. Revisit HOLDs specifically when the market falls; they are the positions where new money should not go.

• Unexplained ratings: the LVMH HOLD contradicts its own columns, which is exactly where a reader should ask why.
• A HOLD that has quietly been a HOLD for years — at some point that is a decision not to own it.

7. Classify the book by mechanism and check the weights against the story you tell about it

1. Assign every holding to one of a small number of mechanisms — founder-led, monopoly/oligopoly, share-count shrinker, and so on.
2. Compute the weights. This is the portfolio's actual structural bet, as opposed to its stated philosophy.
3. Compare with how you describe the portfolio. A mismatch is worth resolving in one direction or the other.
4. Ask what single condition would hurt the dominant category, and whether the smaller categories genuinely offset it.

• Key-person risk concentrated across many positions at once — it is the same risk repeated, not diversified.
• Categories that overlap, which inflates the apparent diversification.

8. Reconcile the published holdings list against the transactions you announced

1. After each period, list the transactions announced since the last report and tick each one off against the holdings table.
2. Investigate any gap immediately — a missing position is either a reporting lag, an unfilled order or an unreported change of mind.
3. Note the reason in the report so readers (and your future self) are not left to infer it.
4. Do the same for exits: a position that silently disappears is the most important thing a portfolio report can hide.

• Limit orders that never filled being quietly forgotten; a published limit below the market is a real risk of non-execution.
• Tables produced at different dates within the same post, which is also why two of them disagree on Medpace and LVMH.