Technical faults around live sports ad breaks made three times as many viewers rate the ads intrusive, according to research from FreeWheel's Viewer Experience Lab that the Comcast-owned video advertising technology company sent to subscribers on November 12, 2026. The report, titled "Delivering Ads Without Disrupting the Game", combines a lab experiment with 282 viewers, run with the research company MediaScience in June 2026, and a Dynata survey of 8,750 sports viewers.

The findings point one way. Played without visible faults, the same ads were rated 13% less intrusive and drew more viewers' eyes. When a "we'll be right back" holding screen filled the break, negative emotions measured through facial expressions rose 170% against a seamless stream. Ads from the same brands that were matched to the moment in the game were rated relevant by 22% more viewers. FreeWheel, which sells the ad serving and buying technology through which much of this advertising is delivered, presents the result as a shared job for publishers, advertisers and technology vendors.

In Short

FreeWheel tested how glitches around live sports ad breaks change the way people react to ads, using 282 viewers in a lab and a survey of 8,750 sports fans. When the stream stuttered, lagged or showed a "we'll be right back" screen, viewers found the ads more intrusive, looked at them less and showed more negative emotion, even though the ads themselves did not change. For broadcasters and brands, it means the technical handoff into a break shapes how an ad lands, although the study did not measure whether any brand was remembered better or sold more.

How the study was built

The research had two parts, according to the report. In June 2026, 282 viewers took part in a controlled viewing session conducted with MediaScience, each watching 45 minutes of MLB, NBA or WNBA content. The lab used physiological measures alongside a post-viewing survey: eye movements to record where viewers looked, facial coding to classify expressions of emotion, and electrodermal activity, a reading of changes in the skin's electrical conductance.

Inside those sessions, ad breaks were delivered under four conditions. An uninterrupted stream played ads without a visible delivery issue. Transition artifacts introduced abrupt cuts or frame stutters around the break. Latency put loading symbols or delays into the break. A slate showed a "we'll be right back" screen instead of an ad. Viewers were also exposed to two kinds of creative: contextual ads, from a brand connected to the game or the sports moment, and standard ads, from the same brand with no such connection.

The second part was a Dynata survey of 8,750 sports viewers commissioned by FreeWheel "to understand broader expectations for advertising during live sports." The report dates it two ways. Its study design section says FreeWheel commissioned the survey in August 2026, while its source notes date it October 29, 2026. Both may be true if the survey was commissioned in one month and closed in another; the report does not say.

A good deal is left out. The report does not name the country of either sample, although all three leagues are North American. It gives no age or gender profile, no device type, no description of how participants were assigned to the four conditions, no count of ads per session and no list of the brands used. Several results carry the note "Differences are statistically significant, p < 0.05." The note does not accompany every figure, including the attention results.

Why live breaks are hard to deliver

Live sport strains ad technology in two ways, according to FreeWheel. The first is scale. "Large audiences may be watching simultaneously, with different ads delivered to individual viewers," the report states, and that scale "can lead to buffering, failed ad calls, and inconsistent video quality." The second is timing. Breaks have to line up with play that can change in seconds, and poor timing, the report says, "can create dead air, black screens, or missed moments."

"When these challenges become visible, an ad delivery issue becomes a viewing issue," according to the report.

As the industry's response, FreeWheel points to the Live Event Ad Playbook, or LEAP, a set of IAB Tech Lab specifications intended "to improve how systems handle massive viewer concurrency and live-event monetization." IAB Tech Lab announced LEAP on August 5, 2025, with Amazon, NBCUniversal, FreeWheel and Index Exchange, and opened the LEAP implementation guide for public comment on October 28, with feedback due on November 23, 2026. That guide sets default limits of 1,000 requests an hour and 100 a minute, and suggests polling every 15 to 30 seconds during live peaks so that buyers and sellers can track how many streams are open at a given moment. Those mechanics sit far from the viewer. The FreeWheel study is an attempt to show what happens on screen when they fail.

Same ad, harsher verdict

"Even when the ad itself does not change, a smoother viewing experience can make the ad feel less disruptive," the report states. Ads were rated 13% less intrusive when played within an uninterrupted stream. Three times as many viewers rated ads intrusive when technical issues disrupted the stream.

The report does not publish the underlying shares behind the threefold figure. A move from 5% to 15% of viewers and a move from 25% to 75% would both be described as "3X," and the report gives no way to tell which kind of gap it measured.

Fewer eyes on the ad

Attention fell as well. According to the report, 7-8% more viewers fixated on the ads when there were no interruptions than when slates or transition artifacts were present. FreeWheel defines a fixation as a period when the eyes remain focused on a specific location for at least 100 milliseconds, a basic unit of eye tracking research and one of the raw inputs behind many attention metrics. The report does not say whether the 7-8% is a relative change or a gap in percentage points.

More negative emotion

The facial coding results were larger. Compared with a seamless stream, the report finds:

  • 97% more negative response when transition artifacts were present
  • 170% more negative response when slates were present
  • 84% more negative response when latency was present

Negative emotions in this measure include sadness, anger, disgust and fear. "Not only did fewer people spend time fixated on the ads, technical issues triggered negative audience response," according to the report.

The slate result stands out. In that condition a holding card appeared where an ad should have been, and it produced the strongest negative reaction of the three faults. Why would a holding card provoke more than a stutter? The report offers no explanation. Nor does it say whether the slate replaced a single spot within a break or the whole break, or which ads, if any, viewers saw around it.

In the message announcing the research, FreeWheel compressed these results. It said that "7% more viewers" spent time focused on ads without technical interruptions, and that technical issues during the ad experience "increased negative emotions in viewers by 2x." The report's own range runs from 84% to 170% depending on the fault, so the doubling describes the transition artifact result most closely and understates the slate result.

What bothers viewers most

The report ranks five frustrations with streaming. Its source notes attribute the list to the 282-person lab group rather than to the larger survey.

  1. Slowness, buffering or delays - 83%
  2. Poor video quality or glitches - 82%
  3. Seeing the same ad too many times - 77%
  4. Too many ads in general - 71%
  5. Abrupt cuts or choppy transitions around ad breaks - 68%

Two of the five concern the stream itself, two concern the volume and repetition of advertising, and one sits on the seam between programme and break. "Viewers' biggest frustrations span the entire experience, from buffering and video glitches to repetitive or excessive advertising," according to FreeWheel.

Formats that leave the game on screen

According to the Dynata survey, 62% of viewers "say the priority for live sports ads is simple: Don't miss the action, and get back to the game fast."

FreeWheel asked those viewers about three in-game formats that run while play stays visible. In the report's illustrations, the lower third places a banner across the bottom of the live picture; picture in picture sets the ad and the game side by side in two boxes; and the frame ad wraps an L-shaped border around a shrunken game feed. The results:

  • Lower third: 55% said the ad works well in live sports, 24% were neutral, 21% said it does not work well
  • Picture in picture: 50% works well, 20% neutral, 30% does not work well
  • Frame ad: 47% works well, 26% neutral, 28% does not work well (the shares add to 101% because of rounding)

"Across all three in-game formats we tested, more viewers said the format works well than said it does not," according to the report. The margins differ widely, though. The lower third's lead was 34 points, by PPC Land's calculation; picture in picture, which shrinks the action the most in the illustration, drew a "does not work well" verdict from nearly a third of respondents. And the conventional spot holds its own. According to the survey, 55% of viewers say 15-second and 30-second ads work well, the same share as the best-rated in-game format.

These are stated preferences from the survey. The lab experiment, as the report describes it, tested delivery faults and contextual creative, not the three formats.

The formats map onto standards work already under way. IAB Tech Lab released six CTV ad format standards for public comment on December 11, 2025 - pause, menu, squeezeback, overlay, in-scene and screensaver - and finalized how those formats are signaled in bid requests on July 22, 2026. In an opinion post on IAB Tech Lab's site, the chief executive of InTheGame, a company that sells such units, argued that moving part of the ad load into non-break formats could shorten streaming ad breaks, without publishing revenue or viewer data to support the case.

Betting apps, everywhere

Repetition is the other side of the problem. Among viewers who find sports advertising disruptive, 45% point to seeing the same ads repeatedly, according to the report, and across all viewers 39% want to see a wider variety of brands. The report does not say what share of viewers find sports advertising disruptive in the first place.

Asked which categories they see too often, respondents named:

  • Sports betting apps - 40%
  • Insurance - 30%
  • Pharmaceuticals - 27%
  • Telecommunications - 27%
  • Alcohol - 24%

The categories they would like to see more often were local attractions (29%), small businesses (27%), travel and tourism (24%), health and wellness (20%) and fast food (18%). No category on the wish list reaches the 40% who complained about betting apps.

Repetition is usually treated as a frequency capping problem, in which a system limits how often one viewer sees one ad. The category lists suggest a second layer that caps do not touch: the same few categories turning up break after break, at least as viewers perceive it. FreeWheel's prescription is to widen the market. "Expanding the demand pool can improve the viewer experience while opening live sports to a broader range of advertisers," the report states. For publishers, the recommended route is wider programmatic access: the report calls on them to "expand programmatic access and package inventory to bring more advertisers into live sports."

That route also runs through FreeWheel's own products. Its first-half 2026 Video Monetization Report recorded programmatic ad views up 25% in the US and 44% in Europe, year on year.

Matching the ad to the moment

The third strand of the research concerns relevance. "Break-level contextual ads in sports use real-time signals from the game broadcast to match an ad to the moment it follows so the break feels connected to the action rather than an interruption of it," according to the report - a live form of contextual advertising.

The comparison held the brand constant. Against the same brands' ads with no link to the game, the contextual versions were rated relevant to the game by 22% more viewers, and emotional engagement, measured by electrodermal activity, was 37% higher. Both differences are reported as statistically significant. "The same brands' ads landed better simply by matching the moment," the report states.

Two limits apply. The report does not describe which game moments were used, how the creative was matched to them or whether the matching was automated, so the lab result tests the idea of a moment-matched ad rather than any live signal system. And electrodermal activity registers arousal, not whether a feeling is pleasant or unpleasant; the report labels the 37% figure "emotional engagement" without separating the two.

PPC Land has covered other claims for sports context. Integral Ad Science said contextual campaigns around the 2025 Super Bowl delivered a 102% ROI increase against overall averages, and PubMatic launched a live sports marketplace with real-time targeting in July 2025.

FreeWheel's division of labour

The report closes with a table assigning work to each side of the market under three goals.

To protect the game, publishers should, according to FreeWheel, validate ad delivery workflows, creative readiness, capacity and contingency plans before the event. Advertisers should have creative uploaded and pre-approved at least 48 hours before the event.

To keep viewers connected, the report calls on publishers to expand programmatic access and package inventory, and to offer formats that minimise disruption and make them easy to buy and activate. Advertisers, in FreeWheel's account, should rotate multiple creatives, messages and products throughout the season, and select formats suited to the campaign objective and the moment within the game.

To make advertising fit the moment, publishers are asked to enable real-time contextual signals and break-level targeting, and advertisers to develop creative tailored to sports moments, teams, leagues or game context.

The 48-hour creative deadline is the only hard number in the table. The report does not explain how it was set or link it to any result in the study.

What the research does not show

FreeWheel's web page for the report goes further than its data. It says "a poor stream or disruptive ad experience can frustrate fans and impact their perception of a brand." The study measured intrusiveness ratings, eye fixations, facial expressions, skin conductance and stated preferences. It reports no brand recall, brand favourability, purchase intent or sales figures. Whether a stuttering stream damages the advertiser, rather than the viewer's mood during the break, is not tested here.

That gap is notable because FreeWheel has measured brand outcomes before. A FreeWheel and MediaScience study of 436 participants, covered by PPC Land on September 18, found unaided brand recall of 41% under a lighter ad load against 26% under a heavier one. An earlier lab study from January 2024, with more than 700 viewers, linked breaks of two minutes or less and limits on repeated spots to better recognition and purchase intent.

The report also gives no data on how often slates, latency or transition artifacts occur in real live sports streams. The lab induced each fault deliberately. Without a measure of their frequency in actual broadcasts, the study shows how much a fault hurts when it happens, not how much it costs the market.

Finally, the source is a vendor. FreeWheel's business is the technology through which these breaks are served, sold and filled. Its recommendations - more programmatic access, contextual signals, packaged in-game formats - align with what it sells. That does not make the measurements wrong, but the method details that would let outsiders check them are absent.

Why marketers are watching live sport

The stakes have grown through 2026. Magnite reported in September that global live sports ad spend on its platform rose 56% year over year from January to July, with more than 5,800 advertisers buying live sports inventory they had not bought the year before. Amazon said its live sports inventory sold out in the second quarter. A Magnite survey of 835 US adults found that live streamers accept an average of 8.7 minutes of ads per hour.

More money chasing the same breaks means more demand per break and more requests hitting ad servers in the same second. That is the pressure LEAP is meant to absorb, and the FreeWheel study puts numbers on what viewers notice when the system buckles. Its category data adds a second warning: more spending does not by itself bring more variety, and viewers already say a handful of categories appear too often.

FreeWheel's own position may change. PPC Land reported on October 25 that, according to tipsheet, Comcast plans to move FreeWheel to NBCUniversal as part of the NBCUniversal spin-off, a plan Comcast had not confirmed. The report itself still carries the line "FreeWheel, A Comcast Company."

The question the study raises but does not answer is how much of a live sports budget is lost to the seams - the stutter before a break, the slate where an ad should be - and who carries that loss. FreeWheel frames the fix as everyone's job. Its data, so far, shows only who notices: the viewer.

Timeline

Summary

Who: FreeWheel, the Comcast-owned video advertising technology company, through its Viewer Experience Lab, with the research company MediaScience running the lab study and Dynata fielding the survey. The findings are addressed to publishers, advertisers and technology vendors selling and buying live sports inventory.

What: A study, "Delivering Ads Without Disrupting the Game," which found that technical faults around live sports ad breaks made three times as many viewers rate ads intrusive, and raised negative emotions by 84% to 170% depending on the fault. In a clean stream, 7-8% more viewers fixated on the ads than when slates or transition artifacts appeared. Ads matched to the game moment were rated relevant by 22% more viewers. A survey of 8,750 sports viewers found majorities favouring in-game formats and 40% saying they see sports betting app ads too often. The study measured no brand outcomes.

When: FreeWheel sent the report to subscribers on November 12, 2026. The lab sessions ran in June 2026; the survey is dated August 2026 in one part of the report and October 29, 2026, in another.

Where: The lab study used MLB, NBA and WNBA content. The report does not state the country of either sample. It is published on FreeWheel's website.

Why: Live sports draw large simultaneous audiences that strain ad delivery, and advertiser demand for those breaks has been rising. FreeWheel argues that ad delivery is part of the viewing experience and that publishers, advertisers and vendors share responsibility for it, a case that also supports the programmatic and contextual products it sells.