Travis Valuation · Workfile No. 4

Dick’s Sporting Goods

Everything behind Issue #4 — including the thing the email did not have room for, which changes the picture.
SubjectNYSE: DKS
Date of valueSeptember 8, 2026
Price$132.37
Issue publishedSeptember 6, 2026
Prepared byCurtis Travis, AACI (Ret.)

Summary of findings

A category-leading retailer, cut nearly in half, with four directors buying the crash with their own money. The published issue told that story and it was true. This file adds what the email could not fit, and one item materially changes how the insider signal should be read.

The directors bought $3.8 million in late August. The same insider group had sold about $47 million earlier in the same year, at roughly $200.

Neither fact cancels the other. Open-market director purchases with personal money remain the rarest bullish signal there is. But a reader given only the buying has been shown one side of a ledger, and this file exists so that does not happen twice.

The one-line conclusion

A healthy core business dragged down by one badly-timed acquisition, priced roughly at its conservative floor rather than below it. Not a Graham bargain. A contrarian turnaround bet with real leverage attached — and a valuation whose upside case rests on earning power that may have been a pandemic artifact.

1 · What the company is

Dick’s Sporting Goods is the largest sporting-goods retailer in the United States, founded in 1948 and public since 2002. Most readers have stood in one. In 2026 it acquired Foot Locker, adding thousands of mall-based footwear stores and a great deal of debt.

Its most recent quarter was, in the core, excellent: revenue up 53% to $5.59 billion and flagship comparable sales up 4.9%. What broke the stock was guidance — management moved the new Foot Locker unit from a projected profit to an operating loss, and the market repriced the whole company on that one line.

2 · The earnings history, and a warning inside it

$19$0 $2.19$3.40$18.27$10.22the pandemic year 2011 2025
Earnings per share, 2011–2025. Nine years of a slow grind from $2.19 to $3.40, then a pandemic explosion to $18.27 in 2021 as the country bought bicycles, home gyms and golf clubs. Everything since has been a descent from that spike, not a decline from a trend.

Why this matters more than it looks

Several valuation methods below read that 2021–2024 stretch as growth and project from it. It was not growth. It was a demand pull-forward that is not coming back on the same terms.

This is the single biggest reason to treat the high-end valuations in section 3 with caution, and it is why the conservative Graham Number — which uses only current earnings and book value — is the more defensible anchor for this company.

3 · Every tool, on one company

Net-Net / NCAV−$55.79Owner-earnings ceiling$22.23Graham Number$121.75Price, Sep 8$132.37Analyst consensus$156.44Earning Power Value$235.53Graham 1974$298.81Graham 1962$324.61
What each method says the shares are worth. Note that the price sits above the conservative Graham Number, and that the three highest readings all depend on growth rates drawn from the pandemic years.
MethodReadingWhat it says
Graham Number$121.75Price is above it — no margin of safety
Graham 1962 formula$324.61Growth-based; see the caution above
Graham 1974 formula$298.81Growth-based; see the caution above
Earning Power Value (B)$235.53Uses a 7-year average that spans the spike
Earning Power Value (A)No discount rate available
Net-Net / NCAV−$55.79Negative — liabilities exceed current assets
Lynch PEG1.20Reasonable against growth
Acquirer’s Multiple (EV/EBIT)13.38×Middling — not deep value
Buffett owner earnings$2.22/shPositive, but implies a $22 ceiling
Return on capitalROIC 7.1% · ROCE 9.5%CROIC 3.4% — modest
Piotroski F-Score4 of 9Weak financial momentum
Altman Z-Score2.51Grey zone — not distress, not safe
Beneish M-Score−0.90Above the −1.78 threshold — flags

The Beneish flag, and why it is probably not what it looks like

The M-Score screens for accounting profiles that resemble earnings manipulators. Anything above −1.78 raises a hand. Dick’s reads −0.90. That is a flag, and it did not appear in the published issue because the issue had no room for it.

It is very likely a false positive. The model keys on sudden jumps in sales growth, asset quality, depreciation and accruals — and a company that has just absorbed Foot Locker will show enormous moves in every one of those inputs for entirely innocent reasons. A large acquisition breaks the Beneish assumptions.

Recorded anyway, because a workfile that only prints the flattering readings is worthless. What would distinguish innocent from otherwise: the same score staying elevated once the acquisition has annualised, alongside receivables or inventory growing faster than sales.

Reading the disagreement

The methods split by what they trust. Current-fact methods — Graham Number, net-net, owner earnings — put fair value at or below today’s price. Growth-projecting methods put it two to three times higher, because they are extrapolating from pandemic-era earnings. The app’s blended average of $245.18, a 45.6% discount, is the arithmetic mean of those two camps, and it inherits the assumption of the more optimistic one.

Cheap only if the 2021 earning power was real. Fairly priced if it was a spike.

4 · The Graham screen

Verdict: Borderline. It fails on three tests, one of which is a data limitation rather than a genuine failure.

TestResultDetail
Current ratio ≥ 2Fail1.48
P/B ≤ 1.5Fail2.05
Twenty years of dividendsUnverifiedOnly 15 years of data available — a provider limit, not a miss
Adequate size, earnings stability, debt test, P/E, growthPassP/E 13.96, ROE 14.7%

Three warning signs fire, fewer than Lennar’s five: earnings declined over three years, a dividend cut sits in the history, and total debt now exceeds book equity — debt-to-equity of 1.39, which is the Foot Locker deal showing up on the balance sheet.

5 · The insiders — the full ledger

Issue #4 was built on the late-August cluster of director purchases, and that cluster is real, unusual and worth respecting. It is also only half the year.

$240$200$160$120−$41.6MMar 31Stack−$0.9MApr 17Lodge-Jarrett−$4.6MMay 28Hobart+$3.7MAug 264 directors+$0.1MSep 1ColomboCircle area ≈ dollar value · red sold, green bought
Insider transactions in 2026, plotted at the price they were done. Circle area is roughly proportional to dollar value. Red is selling, green is buying.
DateWhoActionPriceValue
Mar 31Edward W. Stack, Executive ChairmanSold 210,478~$197~$41.6M
Apr 17Julie Lodge-Jarrett, EVPSold 4,140~$224~$0.9M
May 28Lauren Hobart, President & CEOSold 20,083~$228~$4.6M
Aug 26–27Barrenechea, Colombo, Eddy, Mathrani — four directorsBought 28,650~$129~$3.7M
Sep 1William Colombo, directorBought 913$133.19~$0.1M
About $47 million out at roughly $200. About $3.8 million back in at roughly $129.

How to read that fairly, because the raw ratio overstates it

These are not the same kind of transaction. Edward Stack is the founder and still holds roughly 6.5 million shares — the March sale was about 3% of his position, the sort of diversification a founder does on a schedule. Lauren Hobart sold roughly 6% of hers. Neither is a vote against the company.

Director open-market purchases are different in kind. They are discretionary, personal, small enough to be meaningful only as a signal, and legally awkward to make unless you genuinely believe the price is wrong. Four of them doing it inside 48 hours remains the rarest bullish tell available.

But the honest summary is not “insiders are buying.” It is: insiders trimmed heavily near the top, and then several of them put personal money to work near the bottom. Both halves are information.

And a longer memory

These directors buy fairly often, and not always well. Robert Eddy bought at $194.99 and $185.31 in June 2025 and is well underwater on both. Larry Fitzgerald bought at $215.05 in December 2024. William Colombo bought at $110 in September 2023 and has been adding at intervals ever since. A cluster buy from people who buy regularly is a weaker signal than a cluster buy from people who never do.

6 · What would make us wrong

Foot Locker has to work, and right now it is not. Comparable sales fell 3.6% last quarter, inventory is high, and management has guided the unit to a full-year operating loss. This is the whole thesis and the whole risk in one line.

The debt. The acquisition took cash from roughly $2.6 billion to under $1 billion and net debt to about $7 billion. Much of it is store leases, which is normal for big-box retail, but the cushion is thinner and the Altman reading of 2.51 sits in the grey zone rather than the safe one.

The dividend. The ~3.7% yield is only a gift if it holds, and there is a cut in the history. With cash tighter and debt higher, a bad year would pressure it.

The written test

What would tell us we are wrong: Foot Locker comparable sales failing to stabilise, or the dividend being reduced. Either one breaks the case.

What would tell us we are right: Foot Locker reaching operating breakeven, and the debt beginning to come down.

7 · Considered and rejected

ConsideredRejected because
Leading Issue #4 on the app’s “43% undervalued” verdictThe blended figure leans on growth formulas fed by pandemic-era earnings. The conservative Graham Number is the defensible anchor, and the gap between them is the actual question.
Framing it as a Graham bargainThe price sits above the Graham Number and net current assets are negative. It is a turnaround bet, and the issue said so.
Publishing the Beneish flag in the emailSpace, and a high likelihood of a false positive caused by the acquisition. On reflection this was the wrong call — it belonged in one line. Recorded here.
Omitting the insider sellingDone, and it should not have been. The email mentioned the CEO selling near $228 but not the $41.6 million from the Executive Chairman. That is the correction this file exists to make.

8 · Addendum — September 8, 2026

Issue #4 was published on September 6 quoting the September 4 close of $139.15. Two sessions later the shares closed at $132.37, down 4.87% on the day, against a market that fell only 0.55%. Retail was sold specifically.

At that price the stock sits about $12 above its 52-week low of $120.40 and 46% below its $244.38 high — and, notably, back at roughly the level the four directors paid in August.

Nothing in the thesis has changed. The test set out in section 6 is unaffected by a two-day price move, and neither Foot Locker’s comparable sales nor the dividend has been revisited since publication.

9 · Position and disclosure

The author owns Dick’s Sporting Goods, bought near $126 in late August 2026, and is modestly ahead at the date of value. The position was disclosed in Issue #4 at the time of publication.