IBM reported Q1 2026 on April 23. The filing was a clean beat across three pillars. The stock dropped -8.25% on 3.3x volume to $231.08 — the exact weighted-average price where eight insiders, including Krishna ($3.9M), Kavanaugh ($2.3M), Thomas ($1.9M), and Cohn ($1.75M), bought ≈$11.7M of stock February 18-25. They bought into the February panic; the tape has now returned them to cost.

What the filing says

Consulting signings: -13.3% FY2025 → +8.5% reported (+6.0% cc) Q1 2026. A 21-point swing. Book-to-bill TTM 1.04. Backlog $31.3B. Management cited "strength across our application and data transformation offerings." Q1 consulting revenue was only +0.9% cc — signings lag revenue 6-12 months, so H2 2026 is where this shows.

IBM Z: +50.9% / +48.3% cc. Infrastructure segment profit +111.7%, margin 15.8% (+7.2pts). "Continuing to outperform prior programs." Z17 launched June 2025 — month 9-10 of a typical 4-6 year cycle. AI spire cards being installed at large clients.

Software ARR $24.6B. Red Hat reaccelerated to +10% cc. OpenShift hit $2B ARR. Data +15.9% cc. Confluent closed March 17, six weeks ahead of schedule, $11.59B consideration. Q1 FCF $2.22B (+13.1% YoY), on track for $15.7B FY2026 guide. Management: "strong start... durability... confidence." No hedging language.

What the market thinks

$231.08, 10% of 52W range. RSI 37.2. Below 200DMA by 15.8%.

Sell-side bifurcated post-print: five Outperform calls at $310-340 (Wedbush, Evercore, RBC, Jefferies, Oppenheimer) vs Morgan Stanley Equal-Weight $225. Mean target $284.

ATM IV 104.8% on the May 15 (20d) expiry. 52W IV range: 17-59%. Current is 210th percentile — 2.2x realized vol 47.6%. Put delta at $175 implies ≈35% market probability of a -25% drawdown in 54 days. Our base-rate estimate: ≈10%. That's a 3-4x gap on tail pricing for a name with insider cluster at current price.

P-weighted scenario EV $253 (25% bear $195, 55% base $255, 20% bull $320). +9.5% to current.

Why the gap exists

The consulting inflection is sector-wide, not IBM-unique. ACN Q2 FY26 reported March 19 with record $22.1B bookings, book-to-bill 1.2; CTSH inflected a quarter earlier. Indian pure-plays bifurcated — INFY, HCLTech, Wipro guided 1-4% cc FY27 with stocks -5 to -10%. Attribution for IBM consulting: ≈30% idio (execution within sector) / 70% sector.

Factor decomposition: ≈45-55% idiosyncratic, below the 75% target. Z17 is the cleanest idio pillar — IBM owns 100% of mainframe TAM, and ACN's Julie Sweet validated mainframe modernization as "a major services market" on March 19. Software ARR is ≈40% idio / 60% sector (BofA called "SaaSpocalypse over").

The specific gap: the AI-displacement narrative that drove the February -22% crash has been contradicted by Q1 numbers, but the tape is still trading the old story. Five bullish analyst upgrades post-print; market ignored them.

Risks, ranked

1. Watson Code Assistant adoption unmeasured. The 10-Q named IBM Bob (GA March) but provided no COBOL-migration metrics. If external tools pull clients off-mainframe over 3-5 years, Z17 is the last cycle. Z17 +48% with Claude Code widely available is indirect evidence against this, not proof.
2. Morgan Stanley $225 outlier. Five firms upgraded; MS didn't move. What they see matters — the MS bear case hasn't been stress-tested here.
3. IV 104.8% may be informational. If something undisclosed is coming, the options market is right and we're early. May 15 expiry is the test.
4. 45-55% idio means sector drag. If the Western services recovery reverses, IBM doesn't escape.

Catalysts

• May 15, 2026 — front-month options expiry. IV crush to <60% without news implies anomaly was mechanical.
• ~Jun 12, 2026 — ACN Q3 FY26. Sector confirm/disconfirm.
• Jul 22, 2026 — IBM Q2 2026 earnings. Consulting revenue flow-through proof point.

What would change our mind

• Tape breaks Feb low $220.72 on volume — insider floor broken
• May 15 expiry: IV stays >100% — undisclosed catalyst more likely than dealer dislocation
• Q2 Z17 growth below 15% cc — cycle peaking early
• Q2 signings revert to decline — Q1 was noise
• Morgan Stanley or another shop publishes a bear thesis with a specific mechanism

Evidence