CMS Just Cut Five-Star Medicare Plans From 22 to 15 for 2027

What CMS actually released on October 8

Every October, a week or so before the Annual Election Period opens, CMS recalculates the Star Ratings for every Medicare Advantage and Part D contract in the program and posts them to the Medicare Plan Finder. This year that happened on October 8, 2026, one week ahead of AEP's October 15 start, and the ratings now showing for 2027 will run through plan selections made between October 15 and December 7, 2026.¹ They'll also set 2028 quality bonus payments for the plans themselves, which is a separate story for plan sponsors and not the one this guide is about.¹

The headline number is a cut. Of 508 MA-PD contracts rated for 2027, 15 earned five stars.¹ In the same fact sheet's own historical table, CMS lists 22 five-star MA-PD contracts for 2026, which makes the year-over-year drop look like 7 contracts, down about 32%.¹ On the other end of the scale, 32 MA-PD contracts rated 2.5 stars or below for 2027, up from 23 the year before, and the enrollment-weighted average overall rating across every MA-PD contract slipped from 4.01 to 3.99.¹

None of that is unusual on its own. CMS recalculates cut points, the performance thresholds for each star level, every year based on how the whole program performed, and it says plainly that "changes in Star Ratings year-over-year are normal and expected and vary by measure."¹ A plan can hold its care quality completely flat and still move a half star because the rest of the field improved faster, or because CMS reweighted what counts. What's less normal, and worth sitting with before you get to the part where this affects your website, is what CMS's own number for last year did in the meantime.

CMS's own 2026 number changed twice

Pull the fact sheet CMS actually published for the 2026 Star Ratings, dated November 18, 2025, and its Table 1 shows 18 MA-PD contracts rated five stars out of 516 total, with a 3.98 enrollment-weighted average overall rating.² Roughly 40% of MA-PDs, 207 contracts, earned four stars or higher for 2026, and about 64% of MA-PD enrollees were weighted into those four-plus-star contracts.² That was the number every agent, every plan comparison page, and every AI assistant summarizing "2026 five-star Medicare Advantage plans" had to work with for most of the past year.

Now open the fact sheet CMS published eleven months later, on October 8, 2026, for the 2027 ratings. Its own Table 1 restates 2026 at 22 five-star MA-PD contracts and a 4.01 average.¹ That's not a typo and it's not a different data pull by a different team. CMS explains it directly in the same paragraph: "the 2026 overall rating data reflect the updates for MA contracts that had their 2027 Quality Bonus Payment rating updated in June 2026 and their 2026 overall rating updated in Medicare Plan Finder in July 2026."¹ CMS ran a mid-year correction on a plan year it had already published, rated, and let agents sell against for months, and the correction moved the headline five-star count up by four contracts and the average up three hundredths of a point.

This is the detail most coverage of the 2027 release skips, because most coverage of a Star Ratings release is written to answer "what changed this year," not "how stable is the number I already published." For an agent, the second question is the one that actually matters. If the federal source of record for a star rating can revise its own figure for a plan year that's already live, mid-cycle, with no announcement beyond a footnote in next year's fact sheet, then treating a star rating you published once as permanently true is a bet you're taking without realizing you're taking it.

Why a plan's rating moves without changing

Three separate mechanisms drive a star rating up or down in a given year, and an agent doesn't need to memorize CMS's full Technical Notes to understand the shape of each one.

Cut points move every year. For each of up to 43 quality and performance measures a MA-PD contract is rated on, CMS sets a threshold, called a cut point, that determines whether a given score earns one star or five.¹ Those cut points are recalculated annually based on how the whole program performed during the measurement period, which CMS describes as "the cut points are recalculated each year based on performance during the measurement period."¹ A plan can hold the exact same raw performance on a measure and still drop a star if the rest of the industry got better around it, because the bar moved, not the plan.

Measures get added, retired, and reweighted. For 2027, CMS added two entirely new measures: Concurrent Use of Opioids and Benzodiazepines, and Polypharmacy: Use of Multiple Anticholinergic Medications in Older Adults, both weighted at one.¹ It also tripled the weight of two existing measures, Improving or Maintaining Physical Health and Improving or Maintaining Mental Health, from a weight of one in 2026 to a weight of three in 2027, following what CMS calls "a substantive measure specification change."¹ A measure that carries three times the weight it carried last year can move a contract's overall rating on its own, independent of anything else in the plan's performance.

Mid-year corrections happen, and they land quietly. That's the mechanism behind the 18-to-22 revision covered above: a 2027 Quality Bonus Payment update CMS ran in June 2026, and a 2026 overall rating refresh it pushed to Medicare Plan Finder in July 2026.¹ Neither of those events generated the kind of press coverage the October release gets. They happened inside a plan year that had already been live for half the year, to a dataset most agency sites had already stopped checking.

The full 2024 to 2027 picture

Looking at one year in isolation makes the swing from 2026 to 2027 look sharper than the broader trend actually is. CMS's own Table 1 covers four plan years in one place, and the five-star count has swung in both directions before: 38 contracts in 2024, down to just 9 in 2025, up to 22 in the revised 2026 figure, now down to 15 for 2027.¹ None of those moves are small, and none of them read as a steady trend in either direction. That volatility is itself the point: a single year's rating, cited without the year attached, tells a reader almost nothing about how stable it is.

2026 figures shown as restated in CMS's October 8, 2026 fact sheet. Source: CMS 2027 Medicare Advantage and Part D Star Ratings Fact Sheet, Table 1.¹

Standalone Part D plans tell a slightly different story worth a sentence, since some agencies write both lines. For 2027, zero PDP contracts earned five stars, down from 2 in 2026, 2025, and 2024, and 12 of 41 rated PDPs fell at or below 2.5 stars.¹ But the enrollment-weighted Part D average actually rose, from 3.01 in 2026 to 3.22 in 2027, and the share of PDP enrollees in four-plus-star contracts jumped from about 2% to about 22%.¹ MA-PD quality dipped slightly this year. Standalone Part D quality, weighted by where people actually enrolled, improved. A page that only tracks one of the two is only telling half the story for an agency that writes both.

What a stale claim actually costs you

None of this is abstract once you connect it to what's actually sitting on an agency website right now. Picture the sentence a lot of county-level Medicare pages have carried since last AEP: "We offer [Carrier]'s 5-star rated plan, the highest rating available." If that sentence was written against the 2026 ratings and never revisited, it's now wrong in one of two directions. Either the plan dropped out of five stars entirely for 2027, in which case the claim is simply false, or it held five stars and the page is still citing 2026 data for a reader comparing 2027 options during AEP that opened October 15.

The compliance exposure here is narrower than the superlative-claim rule this blog covered in an earlier guide, because a specific star rating isn't a vague adjective like "best" or "top."³ It's a factual claim with a correct answer for a given plan and plan year, and CMS's general ban on misleading or confusing marketing material was never tied to the documentation requirement that rule loosened. A wrong star rating is a wrong fact on a page a prospect is reading to make an enrollment decision during the exact window, October 15 to December 7, when that decision gets made.

There's also a real business reason to get this specific fact right that has nothing to do with compliance. KFF's Medicare FAQ, published October 9, 2026, describes a Special Enrollment Period that lets a Medicare beneficiary switch to an available five-star Medicare Advantage or Part D plan in their area, usable once between December 8 and November 30 of the following year.⁴ That's an enrollment opportunity outside the normal AEP and lock-in rules, and it only works if the plan a prospect reads about on your site is actually the one holding five stars this year. A stale claim doesn't just risk being wrong. It risks quietly costing you the one piece of information that would have told a prospect they had an extra door open.

A real audit of your own site

This takes under fifteen minutes and needs nothing but your own site's search function and the CMS fact sheet linked in the sources below.

First, find every page that names a star rating. Search your own site for "star," "5-star," "rated," and the name of any specific plan or carrier you write often. Check county and city landing pages first, they're where this claim tends to live longest between revisions, along with any comparison chart, PDF rate sheet, or seasonal AEP page built fast and not touched since.

Second, check the rating against CMS's current table, not your memory of it. CMS's 2027 fact sheet and Medicare Plan Finder are both live right now.¹ A plan's contract ID doesn't change year to year even when its star rating does, so look the specific contract up rather than assuming a rating carried over.

Third, attach the plan year to every rating you publish. "5-star rated" with nothing else is a claim with an invisible expiration date. "The only 5-star-rated Medicare Advantage plan available in [county] for 2027, per CMS's October 2026 Star Ratings release" is a claim a reader, a regulator, or an AI assistant can all verify against the same public document.

Fourth, update your page's dateModified when you update the claim. Google's own documentation on Article structured data describes dateModified as something you add "if you want to provide more accurate date information to Google," separate from datePublished.⁵ A page that updates its star-rating sentence in October but leaves a dateModified timestamp from the prior AEP is sending a crawler the same mixed signal a reader would get from the two different numbers.

How we keep this current for you

This is the same thing every guide on this blog argues for applied to our own build: a figure doesn't go on a page unless it's sourced, dated, and current, and we re-checked our own pricing before writing this so we weren't about to repeat the mistake this guide is warning against.

Digital Foundation's Starter tier runs $247 a month and ships 7 new pages a month, 2 blog posts, 2 location pages, 2 answer pages, and 1 custom page, alongside the full 60-point SEO/AEO checklist.⁶ Pro runs $497 a month and adds up to 15 to 18 new pages a month across blog, location, and answer content, plus a 24/7 AI voice receptionist.⁶ Scale runs $997 a month and ships up to 76 new pages a month, roughly one blog post, one location page, and one answer page every business day.⁶ Every plan except the one-time Full Clone includes a 14-day free look before billing starts.⁶

The reason that matters for a guide about a number that moves every October: a county or plan comparison page we ship isn't a one-time publish. It's a page that gets revisited on the same cadence CMS updates the underlying data, with the source and the plan year stated in the text and the dateModified property updated to match, not a static asset nobody looks at again until next AEP forces someone to notice it's wrong.

We build on Astro, which ships zero JavaScript by default, so a page like this one arrives as finished HTML rather than a runtime the browser has to boot before anything loads.⁶ Our builds score 90+ on Lighthouse performance on both mobile and desktop, and some score 100, which matters here for a plain reason: a plan comparison page a prospect can't load fast enough on a phone during AEP never gets read closely enough for an accurate star rating to matter in the first place.

What changes for your agency

This doesn't promise a ranking position, a citation, or a specific enrollment outcome, because none of those can be promised honestly. What it changes is narrower and more concrete.

The underlying lesson here isn't really about Star Ratings specifically. It's that federal data you treat as a fixed, one-time fact can move under you, sometimes within the same plan year, with nothing louder than a footnote to announce it. CMS just did that to its own 2026 numbers. The only real defense is a page that gets looked at again on a real schedule, not a page that's technically correct the day it ships and silently wrong every day after.

Questions agents ask