What Wall Street's deal-day verdict got wrong about pharma M&A — measured across 830 graded deals.
Here's what's actually happening in pharma M&A: every deal gets judged twice. Once by the Street on announcement day, and once by history — and the two verdicts disagree far more often than the industry admits. The Pharma Closeout grades both: a Critic Score built from announcement-week coverage, and an Outcome Score assessed five years after the deal closes. The gap between them is the most expensive blind spot in the business. Across 830 deals where both grades exist, deal-day consensus missed the eventual outcome by 25 points or more on 43.1% of them — and the biggest misses ran 84 points, in both directions. The Street isn't wrong at the margin. It's wrong about which deals mattered.
| 1 | Merck & Co. Inc. / Insmed Inc. (Follow-on Biologics platform)2009 | $130M | 100 → 16 | -84 |
| 2 | Takeda Pharmaceutical Company Ltd. / Shire plc2018 | $62.0B | 88 → 8 | -80 |
| 3 | Gilead Sciences Inc. / Cell Design Labs, Inc.2017 | $567M | 100 → 21 | -79 |
| 4 | AstraZeneca PLC / Amplimmune Inc.2013 | $500M | 100 → 21 | -79 |
| 5 | Celgene Corporation / OncoMed Pharmaceuticals, Inc.2013 | $3.3B | 88 → 10 | -78 |
| 6 | Hospira Inc. / Orchid Chemicals & Pharmaceuticals (Generic Injectables Business)2010 | $400M | 100 → 23 | -77 |
| 7 | Watson Pharmaceuticals (Actavis Inc.) / Uteron Pharma SA2013 | $305M | 92 → 16 | -76 |
| 8 | Novartis AG / Ionis Pharmaceuticals, Inc. / Akcea Therapeutics, Inc.2016 | $1.4B | 100 → 26 | -74 |
| 9 | AstraZeneca PLC / Arrow Therapeutics Ltd.2007 | $150M | 86 → 13 | -73 |
| 10 | Endo Health Solutions Inc. / Boca Pharmacal, LLC2013 | $225M | 92 → 19 | -73 |
| 1 | Abbott Laboratories / Alere Inc.2016 | $5.8B | 25 → 95 | +70 |
| 2 | Pfizer Inc. / Warner-Lambert Company2000 | $90.0B | 30 → 83 | +53 |
| 3 | Quintiles Transnational Holdings Inc. / IMS Health Holdings, Inc.2016 | $17.6B | 38 → 91 | +53 |
| 4 | Astra AB / Zeneca Group plc1998 | $35.0B | 42 → 94 | +52 |
| 5 | AstraZeneca PLC / Bristol-Myers Squibb Company / Amylin diabetes alliance2012 | $3.5B | 25 → 77 | +52 |
| 6 | Otsuka Pharmaceutical Co. Ltd. / Astex Pharmaceuticals2013 | $886M | 36 → 79 | +43 |
| 7 | Shire plc / Baxalta Inc.2016 | $32.0B | 36 → 77 | +41 |
| 8 | Gilead Sciences Inc. / Pharmasset Inc.2011 | $11.2B | 58 → 98 | +40 |
| 9 | Horizon Pharma plc / Crealta Holdings LLC2015 | $510M | 50 → 87 | +37 |
| 10 | Abbott Laboratories / Knoll Pharmaceutical Company (BASF Pharma)2001 | $6.9B | 63 → 100 | +37 |
Computed across 830 deals with both scores (54.5% of the 1522-deal database), data as of 2026-07-31.
In 2016 the Street scored Abbott's $5.8B bid for Alere a 25 out of 100 — and Abbott itself spent a year in court trying to walk away from it. History scored the same deal a 95. Here's what the consensus missed: the question was never whether Alere's rapid-testing platform worked — it was whether Abbott could build a franchise around it. It could. That platform became the engine of Abbott's COVID-era diagnostics business, turning the industry's least-wanted deal into arguably its most valuable diagnostics acquisition. A 70-point miss — and it came down to commercialization, not the science.
Read the full grade →Takeda's $62.0B takeover of Shire was celebrated on announcement — a Critic Score of 88, the kind of number the Street reserves for deals it's sure about. Five years past close, the Outcome Score reads 8: crushing debt, crown jewels sold off to service it, and a pipeline that never justified the premium. At 80 points, it's the widest megadeal overhype in the database. The enthusiasm was real. The diligence, apparently, was optional.
Read the full grade →Conventional wisdom says pharma got better at M&A as the science matured. Let's put that in context. Deals announced in the 2000s delivered on their rationale 46.7% of the time (255 graded deals). The 2010s — the decade of megamergers and record premiums — managed 37.2% across 390 graded deals. More capital, more scale, a worse hit rate. The boom bought more deals, not better ones.
1990s: n=19 (directional — small sample) · 2000s: n=255 · 2010s: n=390 · 2020s: too early to grade — most assessment windows still open.
Volume does not equal skill. Among the most-graded serial acquirers in the database, outcome hit rates split into clear tiers — and the ranking does not follow reputation.
| Acquirer | Graded deals | Hit rate | Median outcome |
|---|---|---|---|
| Johnson & Johnson | 25 | 68.0% | 72 |
| AstraZeneca PLC | 38 | 60.5% | 62 |
| Roche Holding AG | 47 | 48.9% | 59 |
| Merck & Co. Inc. | 35 | 48.6% | 58 |
| Pfizer Inc. | 44 | 43.2% | 52 |
| Novartis AG | 40 | 42.5% | 56 |
Hit = Outcome Score ≥ 60. Only acquirers with ≥8 graded deals and ≥60% grading coverage shown; full methodology below.
How the scores work. The Critic Score measures announcement-week reception: it is computed from sentiment-tagged coverage across a minimum of four independent sources per deal. The Outcome Score grades what actually happened: strategic rationale, financial impact, and pipeline outcome, assessed once a deal's five-year post-close window opens. Deals too recent to grade are excluded rather than guessed. The Hype Gap is simply Outcome minus Critic.
What this study includes. 830 of the database's 1522 deals carry both scores (54.5%). Decade and acquirer figures are shown only where grading coverage clears 60% and sample size clears 8 graded deals; blocks that fail those gates are excluded and noted, not estimated.
Limitations, honestly. Critic Scores proxy consensus through press sentiment, which under-samples deals that drew little coverage. Coverage is deepest for large, recent-decade transactions. Grades are our own editorial judgments from sourced facts — every deal page lists its sources, and every number on this page links to the underlying record.
Cite this data. Free to cite with attribution and a link. Suggested: “The Pharma Closeout Deal Intelligence Database, July 2026 — 830 graded pharma deals.” For questions about the data or press inquiries, see About.
Every figure on this page regenerates nightly from the live database.