What peer-reviewed research says about NFL injuries, and where PlayerRiskIQ fits

Published Oct 17, 2026 · updated Oct 18, 2026 · PlayerRiskIQ

Site figures in this article are from the Oct 18, 2026 8:15 AM ET update. Research figures are quoted from the papers listed under Sources.

PlayerRiskIQ gives every listed player a chance of missing the next game and a usage or role impact score, built from public nflverse data. This article covers what peer-reviewed research says about which positions get hurt most, how often injuries come back, what "Questionable" and "Doubtful" really mean, and whether missing key players changes team results. For each topic, we say where our method matches the research, where it differs, and where it hasn't been tested yet.

How often NFL players get hurt, by position

All injuries: receivers, tight ends and defensive backs lead

The broadest public-data study we found is Lawrence, Hutchison and Comper (2015). They tracked official NFL injury reports over the 2012 to 2014 regular seasons and counted 4,284 injuries, including 301 concussions. The all-cause rate was 395.8 per 1,000 athletes at risk, and only 2.3% of team games were injury free. Wide receivers, tight ends and defensive backs had the highest rates of both injury and concussion. The knee was hurt most often, followed by the ankle, hamstring, shoulder and head.

Lower-body injuries from the league's own medical records

Mack and colleagues (2020) used the NFL's league-wide electronic health record for 2015 to 2018. They found an average of 2,006 time-loss lower extremity injuries per season, which they describe as "a 1-season risk of 41%" for an NFL player. 58.7% of those injuries happened in games. Hamstring strains were the single most common lower-body injury, ahead of lateral ankle sprains.

Pass plays carry more risk than run plays

Burke and colleagues (2021) studied every regular-season game from 2013 to 2016 using official game books. They found 2,721 in-game injuries, or 1.33 injuries per team per game. Passing plays were linked to more injuries than running plays (OR 1.4, 95% CI 1.3 to 1.5). The difference was largest for quarterbacks (OR 6.9), receivers (OR 5.0) and defensive backs (OR 2.3). OR is a ratio comparing how likely an injury was on passing plays versus running plays.

How this relates to PlayerRiskIQ

Our chance-of-missing model learns from each player's nflverse reports, snaps and stats. It has no hand-set position penalty. We haven't tested whether published position rates would improve it, so we don't claim they would.

Re-injury: the hamstring problem

What the studies found

• Elliott, Zarins, Powell and Kenyon (2011) reviewed 1,716 hamstring strains recorded in the NFL Injury Surveillance System from 1989 to 1998. More than half (51.3%) happened during the roughly seven-week preseason. Defensive backs (23.1%) and wide receivers (20.8%) accounted for the most injuries. In the full text, 16.5% of all hamstring injuries were re-injuries.
• Bodendorfer and colleagues (2021 online; 2023 print) used public data from the 2009 to 2020 seasons and found 2,075 hamstring injuries. Of these, 796 (38.4%) were recurrent, and 247 (11.9%) were same-season re-injuries. Returning sooner was a risk factor: recurrence risk was highest "when players returned within 2 weeks (13.4%)." Wide receivers were at particular risk of re-injury in the same season.

How this relates to PlayerRiskIQ

• Injury history is a model input. In our own held-out testing on 2014 to 2025 seasons, adding richer injury history improved the chance-of-missing model slightly in every held-out season. That is our internal result, not a peer-reviewed one.
• Player pages (open any card from the home page) show a "Returning from injury" flag when snap share drops after a listing, and a "How long could he be out?" history. For example, of 995 times a player was listed Out with a hamstring injury (2014 to 2025), 32% missed 1 game, 26% missed 2 and 42% missed 3 or more in a row. That is history, not a prognosis.
• Not modeled: the "returned within 2 weeks" effect, and any medical detail beyond the body part on the public report.

Questionable vs Doubtful: how often listed players actually play

What the research shows

The best peer-reviewed numbers we found come from Glazer, Binney and Seth (2025). For starters and situational players in 2015 to 2019, they report:

Most Questionable starters played, and almost no Doubtful starter did. (A blank status after 2016 is roughly the old "Probable.") We found no peer-reviewed study of more recent seasons.

What our own data shows (not peer-reviewed)

Our "How long could he be out?" history runs a similar check on nflverse reports. On held-out 2022 to 2025 listings, using all eligible players rather than only starters:

• Questionable (4,220 listings): 68.9% missed no games.
• Doubtful (387 listings): 0.8% missed no games.
• Out (3,014 listings): none played that game.

That lines up closely with Glazer and colleagues, despite different seasons, players and definitions.

How this relates to the cards

The red bar on each card normally shows our model's estimate, which is made on a weekly schedule, often before the official game status is posted. Once this week's report lists a player Out or Doubtful for the game the estimate covers, the card switches to the history for that tag above and shows the earlier model estimate next to it. The lineup call is "Bench or hold" for both tags. For Questionable and untagged players the model estimate stays, and it is only a rough guide at the top end. Our player pages are open about this: of 598 past listings where an earlier, out-of-fold version of our model gave a 75% or higher chance of missing, 38% missed no games. The card guide explains each part of the card.

Does missing high-usage players change team results?

What the research shows

• LaPlaca and Elliott (2021) looked at all 32 teams from 2010 to 2019. Teams whose starters missed fewer games won more. Ranking teams from fewest starter games missed (1) to most, the authors report Spearman correlations of -.212 with wins per season and -.226 with winning percentage. The five healthiest teams averaged 9.42 wins and a 0.553 winning percentage. The five least healthy averaged 7.28 wins and 0.434. These are correlations, not proof of cause.
• Glazer, Binney and Seth (2025) asked which way of counting injuries best tracks the gap between a team's preseason projection and its actual season. Weighting injuries by position helped. Correlations with the win gap were -0.35 for a simple count of games missed, -0.38 for their position-weighted wAGL, and -0.40 when quarterback injuries got double weight. A player-level wins-above-replacement weighting did worst (-0.19).
• Gregory-Smith (2021), treating injuries as a natural experiment, concluded that "Injuries are shown to play a major role in franchise success."
• Shapiro (2026) studied 351 team-seasons from 2012 to 2022 using a salary-weighted injury measure. The paper found that "NFL teams are more resilient to injury than is commonly assumed": after accounting for expected team quality, a seven-percentage-point drop in "Percent Cash Wasted" (the share of payroll tied to players who missed games through injury) went with about one extra win over a 17-game season. Franchise quarterbacks were the main exception, with much larger effects.

Over a season, then, losing important players (quarterbacks above all) goes with fewer wins, though for most positions the effect is smaller than fans tend to assume.

Where PlayerRiskIQ agrees, and where it differs

Agrees in spirit: usage impact weights players by importance: his share of team targets and carries (pass attempts for a QB) over the team's last 3 games, with games he missed counted as zero, times his chance of missing. A 30% share times a 40% chance is 12.0 points. High starts at 3.7 (top 5% of 2017 to 2021 player-seasons), Medium at 0.7 (top 20%), and the bold High badge also needs at least a 25% chance of missing. Team Injuries sums these into "impact at risk," much like weighted burden measures such as wAGL. The handcuffs and waiver wire pages show where that work would go.

Differs: our weights come from usage and snaps, not salaries or wins above replacement, and our tiers are percentiles not tied to wins. Because a QB's share uses pass attempts, one quarterback can dominate a team total. In the Oct 18, 2026 (8:15 AM ET) update, Philadelphia's league-high 167.9 came mostly from one QB card: Jalen Hurts, listed Out, at 98.9.

Tested once against game results, and the test came back negative: in a test we set up before seeing results (1,673 games, 2018 to 2025), adding team availability to a game model did not improve predictions of who wins or by how much. That fits the season-level studies: they explain drift from preseason expectations, while week-to-week expectations already react to injury news. So our impact scores are descriptive. They show how much work could change hands, not who will win. Grading of this season's chance-of-missing estimates hasn't started yet.

A data caveat that applies to everyone, including us

Inclan and colleagues (2021 online; 2022 print) compared 42 studies built on publicly obtained NFL injury data with the league's own medical database, using ACL tears as the test case. On average, the public-data studies captured 66% (range 31% to 90%) of ACL injuries. Capture was 86% for skill players, 72% for "midskill" players and 61% for linemen, and just 35% for special teams injuries. They warn that public data is biased "toward capturing injuries in more popular players."

PlayerRiskIQ uses public nflverse reports, so our histories likely undercount some injuries, most of all for linemen, where role impact matters. See Data sources.

Key takeaways

• Receivers, tight ends and defensive backs have the highest injury rates in public data, and passing plays carry more risk than running plays.
• Hamstrings come back often. One public-data study found 38.4% were recurrent, with the highest risk when players returned within two weeks.
• Most starters listed Questionable play, and almost no one listed Doubtful does. Our own 2022 to 2025 check agrees.
• Over a season, losing important players (quarterbacks above all) goes with fewer wins. Our test found that availability did not improve game-by-game predictions, so treat the impact scores as descriptive.

FAQ

What NFL position gets injured the most?

In 2012 to 2014 injury-report data, wide receivers, tight ends and defensive backs had the highest injury and concussion rates (Lawrence and colleagues, 2015).

How often do players listed Questionable actually play?

In Glazer and colleagues' 2015 to 2019 data, 28% of Questionable starters missed the game, so about 72% played. In PlayerRiskIQ's own held-out 2022 to 2025 check of all eligible players, 68.9% of Questionable listings missed no games.

Does Doubtful mean a player will sit?

Almost always. Glazer and colleagues found that 1 of 653 Doubtful starters played from 2015 to 2019. In our 2022 to 2025 check, 0.8% of Doubtful listings missed no games.

How common are hamstring re-injuries in the NFL?

A 2009 to 2020 public-data study found 38.4% of 2,075 hamstring injuries were recurrent. A 1989 to 1998 surveillance study reported 16.5%, using different data and definitions.

Does a high usage impact score mean a team will lose?

No. It shows how much of a team's offense is at stake if a player sits. In our pre-set test on 1,673 games, adding availability did not improve predictions of game results. See the guide.

Has PlayerRiskIQ's chance-of-missing model been published in a journal?

No. It is our own frozen model, logged before games and graded afterwards. Nothing here is medical advice, and no outcome is certain. See the legal disclaimer.

Sources (peer-reviewed)

1. Lawrence DW, Hutchison MG, Comper P. Descriptive Epidemiology of Musculoskeletal Injuries and Concussions in the National Football League, 2012-2014. Orthopaedic Journal of Sports Medicine. 2015;3(5). https://doi.org/10.1177/2325967115583653
2. Mack CD, Kent RW, Coughlin MJ, et al. Incidence of Lower Extremity Injury in the National Football League: 2015 to 2018. The American Journal of Sports Medicine. 2020;48(9):2287-2294. https://doi.org/10.1177/0363546520922547
3. Burke J, Geller JS, Perez JR, Naik K, Vidal AF, Baraga MG, Kaplan LD. Effect of Passing Plays on Injury Rates in the National Football League. Journal of Strength and Conditioning Research. 2021;35(12S):S1-S4. https://doi.org/10.1519/JSC.0000000000004172
4. Elliott MCCW, Zarins B, Powell JW, Kenyon CD. Hamstring Muscle Strains in Professional Football Players: A 10-Year Review. The American Journal of Sports Medicine. 2011;39(4):843-850. https://doi.org/10.1177/0363546510394647
5. Bodendorfer BM, DeFroda SF, Newhouse AC, et al. Recurrence of Hamstring Injuries and Risk Factors for Partial and Complete Tears in the National Football League: An Analysis From 2009-2020. The Physician and Sportsmedicine. 2023;51(2):139-143 (online 2021). https://doi.org/10.1080/00913847.2021.2013106
6. Glazer A, Binney Z, Seth T. Weighted Approaches to Assessing Injury Burden in Team Sports. Journal of Sports Analytics. 2025;11. https://doi.org/10.1177/22150218241304941
7. LaPlaca DA, Elliott JM. The Relationship Between Injury Rates and Winning in the National Football League. International Journal of Strength and Conditioning. 2021;1(1). https://doi.org/10.47206/ijsc.v1i1.62
8. Gregory-Smith I. Wages and Labor Productivity: Evidence From Injuries in the National Football League. Economic Inquiry. 2021;59(2):829-847 (online 2020). https://doi.org/10.1111/ecin.12960
9. Shapiro J. Percent Cash Wasted: Quantifying the Relationship Among Player Injury, Payroll, and Team Performance in the NFL. Sports Economics Review. 2026;16:100080. https://doi.org/10.1016/j.serev.2026.100080
10. Inclan PM, Chang PS, Mack CD, et al. Validity of Research Based on Public Data in Sports Medicine: A Quantitative Assessment of Anterior Cruciate Ligament Injuries in the National Football League. The American Journal of Sports Medicine. 2022;50(6):1717-1726 (online 2021). https://doi.org/10.1177/03635465211015435

PlayerRiskIQ figures come from our own analysis of nflverse data (CC-BY-4.0). See Data sources.

More reading: Fantasy Injury Guide: Start/Sit & Waivers · How to read the cards · All articles