Skip to research content
National Running Club Database NRCD

Prototypes for Humanity · University of Notre Dame

University of Notre Dame · Prototypes for Humanity

National Running Club Database

Fair performance data for every collegiate club runner

Built by the Notre Dame Running Club so collegiate club runners finally have the shared results website NCAA Division I programs always had, plus advanced analytics those programs still do not. Live today for 190 U.S. college clubs in 39 states, with open science anyone can reproduce.

CRC Coaches Poll members at 2026 NIRCA Track and Field Nationals
CRC Coaches Poll · 2026 Track and Field Nationals 190 teams in the database · 35 universities in the coaches poll · 5 research universities · 1 database

How collaboration changes culture

Before there was a national database, there was one club that refused to stay average. Campus culture at the University of Notre Dame Running Club is why NRCD exists: shared training, shared standards, and the belief that club runners deserve infrastructure worth racing for.

Around 2020, athletes at Notre Dame looked outward to clubs like the University of Michigan Running Club, whose public presence made expectations clear, while Notre Dame’s own site lagged. The fix was never only a better webpage. Notre Dame had to embrace team culture: train together, stick together, and treat club racing as a community with standards, not a collection of individuals who happen to wear the same kit.

Notre Dame Running Club

These milestones are for one campus club. The national platform numbers appear later under Reach.

  • 5 → 11 ND officers, 2022 to 2026

    A wider board, including roles such as sprint coordinator, so practice quality and event coverage scaled with the roster.

  • 2 → 25 ND All-Americans through 2020 vs spring 2026

    Honors that used to feel rare became the product of shared training, better meets, and belief that the club belonged on the national stage.

  • 2022 ND starts varsity-team meets

    Notre Dame Running Club began competing at varsity team meets, raising the bar for preparation instead of waiting only for club-only calendars.

From “every four years” to expected

The old line was that Notre Dame might reach nationals once every four years. Through culture, that shrug disappeared. Qualifying stopped being a pleasant surprise and became something the team planned for together: workouts, travel, and accountability that treat the podium as the default ambition, not a coin flip.

Results follow the culture

Stronger practice design, specialized officer roles, and racing against sharper fields showed up in denser meet calendars, more teammates racing together, and All-American depth that mirrors the evidence that community tracks improvement better than isolated training.

From grassroots to national presence

So they built the missing infrastructure. Athletes at Notre Dame Running Club turned that culture into shared tools: durable results history, better interclub communication, and a national platform that still races NIRCA nationals while serving 190 colleges nationwide.

How it started

  1. Aug 2022

    Club history sheet and summaries

    Jonathan Karr created a Google Sheet for Notre Dame Running Club to track the club’s historical results, and began writing season summaries so marks and stories would not disappear after each meet.

  2. Fall 2022

    Nationals qualification scouting

    Jonathan manually tracked which teams looked likely to qualify for nationals and shared those reads inside the club. It was useful, and it made the larger gap obvious: every club was guessing from incomplete data, with no common place to compare.

  3. Jan 2023

    Notre Dame club Substack

    The club launched its Substack to publish club stories and race coverage, before the national coaches poll existed.

  4. Aug 2023

    CRC Coaches Poll

    Jonathan launched the CRC Coaches Poll so clubs across the country could rank the best NIRCA cross country and track programs and begin talking to each other in one channel.

  5. Sep 2023

    CRC Substack

    Ben Darden (Virginia Tech) launched the CRC Coaches Poll Substack so national club racing had a public newsletter, not only private spreadsheets.

  6. Fall 2023

    A shared national Google Sheet

    To compile poll inputs, Ben and other poll members helped build the first national results sheet. For the first time, club marks from many colleges lived in one place instead of scattered meet PDFs.

  7. 2023-24

    Classroom prototype

    In Professor Ramzi Bualuan’s Advanced Database course, Jonathan, with Kayla Ambrose and Evan Hall, turned that sheet into a National Running Club Database prototype.

  8. Summer 2024

    Built in the open

    Jonathan kept shipping features through the summer so clubs could use a real product, not a class demo, when the season returned.

  9. Aug 12, 2024

    Public launch

    NRCD went live at nationalrunningclubdatabase.com. Clubs began uploading historical and current meets; the corpus has grown into a national research and competition infrastructure.

  10. Aug 2025

    Club Coordination Council Discord

    Ninawa Odicho (University of Illinois) launched the Club Coordination Council Discord as a shared communication platform between clubs and NIRCA, closing the interclub gap that started this whole story.

  11. July 2026

    Meet registration and live results

    NRCD launched on-platform meet registration and live results so clubs can run race day without a separate ops stack. That goes beyond a D-I-style results archive into tools NCAA Division I public browsers typically do not provide together.

  12. Today

    National presence

    190 clubs in 39 states, 141,928 approved marks, open papers and Zenodo releases, and community coverage that treats NRCD as the club sport’s shared results home. Full history also lives on our About page.

What the build unlocked

  • Advanced analytics for clubs

    Thousands of athletes can track PRs, progression, and meet data that used to live in PDFs and private sheets.

  • Standardized performance data

    Public algorithms adjust XC times for course grade, weather, wind, and elevation so marks on different courses become comparable.

  • Open scientific research

    The same corpus powers peer-facing papers on standardization and schedule analytics for non-varsity collegiate athletics.

  • National officer community

    The CRC Coaches Poll still connects hundreds of schools with shared rankings and friendly rivalry.

Meet the athletes behind the prototype at the Running Club of Notre Dame, follow their Substack, or read the full product history on About.

Visit ND Running Club

The gap we solve

NRCD looks familiar if you know NCAA Division I results sites: meets, marks, teams, and history in one place. Club running never had that floor. We built the floor, then went beyond what D-I athletes usually get from public tools alone.

Before / elsewhere

  • Results trapped in PDFs, private sheets, and closed websites
  • No bulk export from Athletic.net, MileSplit, or TFRRS
  • Outside research often stuck at about 500 race results, with no public downloadable dataset
  • Course weather, distance, and elevation rarely comparable
  • NCAA D-I browsing tools still lack open meet registration, live results, and research-grade exports for clubs

With NRCD

  • One moderated live platform for NIRCA clubs that matches D-I-style coverage and then exceeds it
  • 141,928 approved performances and growing
  • Gender-stratified corpus (39.4% women)
  • Unified standardization for weather, elevation, course, and venue
  • Same formulas on the website, in nrcd, and in Zenodo

Beyond NCAA Division I tooling

Matching a D-I results archive was the start. Club athletes also needed operations and analytics that public NCAA D-I stacks typically do not put in one open product.

  • July 2026 · Meet registration and live results

    Clubs can run roster registration and live meet-day results on the platform itself (upcoming meets). That ops layer is not what NCAA Division I programs get from a public results browser alone.

  • Advanced analytics NCAA sites do not ship together

    Power rankings, predicted matchups, head-to-head comparisons, and weather-aware standardization so marks stay comparable across courses and days.

  • Research-scale open data

    Without a public downloadable dataset, prior club-running analyses often stopped near about 500 race results. NRCD publishes a moderated corpus and annual Zenodo releases so the next paper can start from tens of thousands of marks, not a private scrape.

How the prototype works

A production system already in use by clubs, with an open scientific layer that updates yearly.

  1. 1

    Community submits

    Any club can upload meets and results through the live website.

  2. 2

    Admins approve

    Moderators review submissions before marks become public competition data or research corpus.

  3. 3

    Times standardize

    Weather, elevation, course length, and venue factors produce comparable marks.

  4. 4

    Science opens

    Annual PII-safe Zenodo export and pip install nrcd match the site.

Who benefits

  • Athletes get a durable history of every mark in one place.
  • Coaches get Automark registration, live results, records, and fairer comparisons.
  • Researchers get reproducible data and the same formulas used in production.
  • The sport gets evidence that includes women and the full club ability band.

National reach

These numbers update from the live NRCD database: teams with at least one approved result.

190 Club teams
39 U.S. states
31,081 Athletes
141,928 Approved marks
1,386 Approved meets
39.4% Women

Club athletes are real athletes

NIRCA club runners compete for their universities without being limited to an exclusive varsity roster. Programs must serve the full band of ability on campus, from first-time collegiate racers to athletes who would place on many NCAA, NAIA, or NJCAA teams.

That breadth is why open data matters. Scheduling, equity, and progression research should reflect the athletes who actually race, not only an elite subset.

Day to day, clubs use Automark registration, live results with national badges, records, and analytics. Once a year, researchers get a PII-safe export of the same community.

Teams by NIRCA region

Live distribution of 190 clubs with approved results.

Corpus dashboard

Same approved-entry rules as the public Database Summary: named athletes with timed results. Unique runners are split by athlete gender.

1,386 Meets with results
141,925 Entries
31,079 Unique runners
18,842 Men 12,237 Women

Unique runners by gender

Named athletes with at least one approved timed result.

Historical vs current

Era uses meet start date: historical before Aug 2023, current from Aug 2023 onward (same cutoff as coverage tools).

Meets and results by state

Venue state on approved meets, ordered by results. Click a bar to search NRCD for that state.

Results across the U.S.

Darker states have more approved results. Click a state to search NRCD.

Loading map…

Sorted by results (high to low). Click a column to reverse or change sort.

Approved meets and results by venue state
Pennsylvania PA 92 22,451
Virginia VA 112 17,896
Indiana IN 97 16,736
Michigan MI 96 14,842
Illinois IL 66 12,318
North Carolina NC 180 11,788
Massachusetts MA 59 9,177
Ohio OH 63 5,158
Kentucky KY 23 3,402
New Jersey NJ 23 3,139
Wisconsin WI 79 3,132
Maryland MD 17 2,886
California CA 95 2,750
New York NY 42 2,587
Oregon OR 111 1,951
Connecticut CT 15 1,827
District of Columbia DC 6 1,453
Iowa IA 30 1,275
Tennessee TN 23 1,019
Rhode Island RI 6 1,013
Minnesota MN 15 852
South Carolina SC 14 845
Florida FL 23 734
Georgia GA 19 727
Arizona AZ 8 539
West Virginia WV 11 472
Missouri MO 24 313
Delaware DE 4 204
D.C. 1 144
Arkansas AR 9 115
Texas TX 5 60
Nebraska NE 3 51
Alabama AL 4 31
Washington WA 2 19
Colorado CO 3 10
Kansas KS 3 7
Hill 1 3
Nevada NV 1 1
Poland 1 1

Findings

Evidence from the methods paper and XC analytics paper. Live charts use the website database; paper charts use published table values. Associations are observational, not causal.

Live corpus

A national, multi-sport archive

The live site now holds 141,928 approved performances across 1,386 meets. Gender-stratified reporting keeps women visible in the evidence base (39.4% of athletes with approved results).

Corpus by sport

Approved performances on the live site, by sport and gender.

Methods paper

Standardization makes meets comparable

Full XC standardization lowers median within-athlete cross-meet variability by 51.0% for women and 34.4% for men versus raw times. Distance-only conversion still overstates mean first-to-last seasonal gains by about 14 s (women) and 20 s (men).

Standardization impact

Percent drop in median within-athlete XC variability vs raw times.

XC analytics paper

An inverted-U, not “more is always better”

Median first-to-last Standardized gains rise into the mid-season race window, then fall at 5+ races. Mixed models confirm that inverted-U shape without a significant sex difference. Separately, race-result features still do not forecast which athlete improves next season (best held-out men’s R² ≈ 0.04; women’s negative).

Inverted-U by race volume

Median first-to-last seasonal gain (sec) by race count, Standardized times.

XC analytics paper

Team race frequency tracks nationals placement

While individuals are hard to forecast, programs whose athletes race more often show higher nationals placement rates (pooled RR = 2.09 at ≥4 races; GEE OR = 2.56 per SD). Treat this as an associational signal for hypothesis generation, not proof that adding meets causes place.

Nationals placement association

Pooled risk ratio for nationals placement by team race-frequency threshold.

Papers & open stack

Methods, schedule analytics, and the artifacts needed to reproduce them.

  • Methods · dataset CIKM submission · 2026

    NRCD: An Open Database of Collegiate Running with Unified Performance Standardization

    Jonathan A. Karr Jr., Ryan M. Fryer, Ben Darden, Nicholas Pell, Kayla Ambrose, Evan Hall, Ramzi K. Bualuan, Nitesh V. Chawla

    Introduces the open NRCD dataset at scale and a unified standardization framework that lowers within-athlete cross-meet variability on XC versus raw times. Recommends gender-stratified modeling.

  • Analytics · XC arXiv · 2026

    Faster Results from a Smarter Schedule: Reframing Collegiate Cross Country through Analysis of the National Running Club Database

    Jonathan A. Karr Jr., Ryan M. Fryer, Ben Darden, Nicholas Pell, Kayla Ambrose, Evan Hall, Ramzi K. Bualuan, Nitesh V. Chawla

    On comprehensive-era XC results with Standardized times: individual improvement is hard to forecast from race features alone, while team race frequency associates with nationals placement. Converted Only (distance only, no weather) overstates seasonal gains relative to Standardized.

  • Public research dataset

    Annual PII-safe export of approved performances. Updated once per year.

    Zenodo record
  • nrcd Python package

    The same distance, elevation, heat, and venue adjustments used on the live site.

    pip install nrcd
  • Live platform

    Teams, meets, Automark, records, and analytics in production for NIRCA clubs.

    nationalrunningclubdatabase.com

Team

Researchers and community builders behind NRCD. Interested in joining the effort? Email jkarr@nd.edu.

  • Jonathan A. Karr Jr.

    Creator · Lead researcher

    University of Notre Dame · National Running Club Database

  • Ryan M. Fryer

    Graduate researcher

    University of Notre Dame · University of Virginia

  • Ben Darden

    Results contributor · Writer

    Uploaded over 50,000 results and written club-running articles that grow the national corpus.

    Virginia Tech

  • Nicholas Pell

    NRCD Vice President · Results contributor · Writer

    Uploaded over 50,000 results and written articles that help clubs find and trust the data.

    University of North Carolina at Chapel Hill

  • Owen Christou

    CRC Coaches Poll Vice President

    Penn State University

  • Kayla Ambrose

    Advanced Database project partner

    Built the first NRCD prototype with Jonathan and Evan in Professor Bualuan’s Advanced Database course.

    University of Notre Dame

  • Evan Hall

    Advanced Database project partner

    Built the first NRCD prototype with Jonathan and Kayla in Professor Bualuan’s Advanced Database course.

    University of Notre Dame

  • Ramzi K. Bualuan

    Faculty collaborator

    University of Notre Dame

  • Nitesh V. Chawla

    Faculty advisor

    University of Notre Dame · Lucy Family Institute for Data & Society

Want to join the team?

Students, coaches, and researchers who want to help grow the open club-running evidence base are welcome. Tell us briefly about your background and how you’d like to contribute.

Email jkarr@nd.edu

UN Sustainable Development Goals

Open gender-stratified performance data and research on training under environmental stress align with these global goals.

Next steps

Finish times tell us who raced and how fast under fair conditions. The next gap is everything that happens before the gun: training load, recovery, and health signals that athletes already capture on devices. We are filling that gap by linking opt-in sync to the race results they claim on NRCD.

The next gap to fill

  • Environmentally standardized marks still do not encode individual training plans, mileage, intensity, or taper.
  • Meet dates alone leave recovery out of view: heart-rate variability, sleep, and other freshness markers from wearables.
  • Older historical rows often lack complete course and weather metadata, and much of the variance work so far focuses on XC.
  • Race-result features alone do not forecast who improves next. Training-aware features are the natural next model class.

How we fill it

  1. Opt-in device sync. Athletes authorize Garmin and related health platforms through the product’s Connected Health flow (consent is collected at connect time; research rides with the connection while it stays active).
  2. Link workouts to race results. Synced training load and recovery markers join the same claimed athlete profile that already holds meet marks, so coaches and researchers can study the week before a race, not only the finish line.
  3. New models. Gender-stratified progression and readiness models that combine training signals with Standardized race features.
  4. Health stats for athletes. Personal improvement trends and training insights on the live platform (roadmap), with annual PII-safe research exports remaining consent-gated.

NRCD keeps the open meet corpus as the competition backbone while we add how athletes prepared, recovered, and arrived at each mark. Researchers who want to collaborate on Connected Health models or consent UX can join the team.

Empower club leaders and future research

NRCD exists to empower the people who run clubs and the people who study them: officers who need trustworthy tools on Tuesday night, and researchers who need open methods that outlast one season.

Empower club leaders

Officers and coaches with the full picture

Moderated results, meet registration and live results, power rankings, and the CRC Coaches Poll belong in one place so presidents, training chairs, and meet hosts can plan with the clarity Division I staff expect, plus ops tools public D-I browsers still lack.

Empower future research

Open science beyond one campus

The same corpus should empower the next paper and classroom project. Annual Zenodo releases, pip install nrcd, and published methods let scholars reproduce standardization and extend the work without rebuilding the database from scratch.

Force for good in the world

A culture that started in a Notre Dame club room now aims to lift collegiate club running, open science, and athlete health as a force for good in the world.

Research & team inquiries: jkarr@nd.edu · support: nationalrunningclubdatabase@gmail.com