NoPunt API Review 2026
A developer review of the NoPunt public API, a free read-only feed of one NFL prediction model: this week's picks with win probabilities, every graded pick back to 2020, EPA power rankings, stat leaders and a read-only MCP server, all with no key.
NoPunt at a Glance
Green = available ยท amber = unverified or varies by plan ยท crossed out = not available. Pricing: ๐ฒ โ $20/mo up to ๐ฒ๐ฒ๐ฒ๐ฒ๐ฒ = $1,000+/mo. From our directory research.
What NoPunt Is
NoPunt runs an NFL prediction model and publishes everything on its site as JSON. The pitch is accountability: live picks are captured before kickoff where NoPunt can prove the time, graded on a public ledger, and every change to a pick has been recorded since 24 September 2026. The 2020 to 2025 seasons are a walk-forward backtest that NoPunt rebuilt in May 2026, and its own llms.txt says so (checked November 2026). The free API exposes that ledger, the current week's picks, team analytics and player stat leaders. Three models vote on each game, and the site publishes the method and the calibration alongside the results.
It is not an odds API. The picks feed carries no spread or moneyline fields (checked November 2026), and the product is the model's own number: a win probability and a confidence tier for the side it backs. Think of it as a second opinion to put next to a price feed, not a replacement for one.
NoPunt is NFL only, regular season only. The model has never made a postseason pick, so weeks 19 and up return a 404.
Coverage and Endpoints
The docs list 18 endpoints. The JSON ones need no authentication, allow 60 requests a minute, and send open CORS headers, so a browser widget can call them directly. The free set splits into picks, analytics and plumbing:
- Picks.
/api/picks.jsonreturns the latest published slate./api/picks/{season}/{week}.jsonreturns any regular-season week from 2020 with the final score and a graded result joined to each pick. - Performance.
/api/stats.jsoncarries the all-time record, hit rate by confidence tier and a per-season breakdown./api/teams.jsonand/api/team/{abbr}add the pick history, EPA profile, coaching tendencies and playoff simulation odds for each of the 32 teams. - Analytics.
/api/power-rankings.jsonranks teams by net EPA per play./api/players.jsonreturns stat leaders by position./api/nextgen.jsonand/api/advanced-stats.jsonserve Next Gen Stats and Pro Football Reference style leaderboards./api/game-wp.jsonreturns a finished game's win-probability curve, or the season's most dramatic games. - Plumbing. A health check, an RSS feed of the last 50 graded games, and the MCP server covered below.
Five endpoints sit behind the Pro subscription: fantasy projections, player prop projections and three CSV exports of the full pick history, weekly player stat lines and team EPA composites. They return a 403 without an active session.
The Picks Feed
Start with the current slate:
curl https://www.nopunt.com/api/picks.json
Each pick names the game by its stable nflverse id, the two teams, the side the model backs, a tier from S down to C, the picked side's win probability as a decimal and the projected margin. Kickoff times are ISO 8601 in UTC. Historical weeks add a result object with both scores, a correct flag and the flat one-unit profit at the spread-implied price.
Two field conventions matter, and the docs are blunt about them. predicted_margin is home-relative, so flip the sign when pick_side is AWAY. And result.correct is null on a tie, not false, so test for null before you count a loss.
Read slate_state before you treat a slate as upcoming. It is upcoming before the first kickoff, in_progress during the week and final after the last game, and between a week's end and Wednesday's new picks the newest slate is a completed one. The current slate is cached for five minutes and historical weeks for a day.
The Public Record
A picks feed is worth exactly as much as its track record, and NoPunt puts the whole thing behind one endpoint. When we called /api/stats.json on 16 October 2026 it reported about 1,700 graded picks across seven seasons, a straight-up hit rate around 66 percent overall and around 84 percent for the small S tier, with every season between 61 and 70 percent. Six of those seasons, 2020 to 2025, come from the walk-forward backtest that NoPunt rebuilt in May 2026. Only the 2026 picks are live calls.
The by-tier object is the part to read twice. The docs note that accuracy rises with tier but profit does not: the S and A+ tiers hit most often and sit near flat at the spread-implied price, while the B and C tiers carry the profit. That is the sort of detail a picks site usually buries, and here it is in the JSON.
One caveat the docs state themselves: the profit series is priced at a vig-free moneyline derived from the market spread, which no sportsbook offers. Treat it as a comparable series, not a bankroll you could have had.
The MCP Server
NoPunt ships a read-only MCP server at https://www.nopunt.com/api/mcp. It uses the Streamable HTTP transport with no auth and no session, and it serves the same free data as the JSON endpoints. A tools/list call returned seven tools when we tested it: the week's picks, a plain-English weekly briefing, the survivor board, the pick'em board, the confidence pool, the ledger summary by season, and a single matchup with its pick, line movement and injuries.
An agent-skills discovery document sits at /.well-known/agent-skills/index.json. The paid Edge signal, Kelly sizing and subscriber data are kept off the server, the same as everywhere else on the site.
NoPunt Pricing and Attribution
The JSON API is free: no key, no signup, 60 requests a minute. Pro costs $19 a month, billed until you cancel, or $69 a year, with no seasonal pause, and new members get 7 free days (checked November 2026). Pro unlocks the fantasy and player prop projections, the three CSV exports and the Edge fields on the site. Pro is sold as a membership for bettors and pool players, and the API extras come with it.
The underlying play-by-play, schedules, rosters and injuries come from nflverse under CC BY 4.0, so anything you build on them needs the same attribution. NoPunt asks for a credit and a link back as well, and offers to link to production projects from its footer.
What It Does Not Cover
- Other sports. NFL only, regular season only. For college football or anything else, see the sports data API guide.
- Sportsbook odds. The picks feed carries no spreads or moneylines (checked November 2026), and the terms bar redistributing odds or lines as a feed. Pair it with an NFL odds API for prices by book.
- Live data. The win-probability curves are historical play-by-play from nflfastR, not a live feed, and there is no WebSocket. The model's numbers are pregame.
- Schedules, rosters and standings. The picks feed carries kickoff times for the current week, and the team endpoints carry EPA rankings and playoff odds, but there is no season schedule, roster or standings endpoint.
- Box scores. Final scores appear inside graded picks. There are no box scores or play-by-play.
Getting an API Key
NoPunt has no API key to request. The JSON endpoints answer any caller, so a curl command, a script or a browser tab works as soon as you have the URL. The one thing that needs a login is NoPunt Pro, and it works through a website session, not an API token. Without a signed-in Pro account, the Pro endpoints (fantasy projections, player props and the CSV exports) return 403 with {"error":"Pro subscription required"}.
Read the terms before you ship. The public endpoints are free for personal, research and non-commercial use. The terms ask you to email hello@nopunt.com for written permission before you put them in a paid product or service. The API docs list the calls and the fields (checked November 2026).
A First Request in Python
This sample reads the current slate from /api/picks.json. It sends no key, because the endpoint needs none (from the docs, checked November 2026).
import requests
resp = requests.get("https://www.nopunt.com/api/picks.json", timeout=15)
resp.raise_for_status()
slate = resp.json()
print(slate["season"], slate["week"], slate["slate_state"])
strongest = sorted(slate["picks"], key=lambda p: p["confidence"], reverse=True)
for pick in strongest[:3]:
print(pick["game_id"], pick["pick_team"], pick["tier"], round(pick["confidence"], 2))
The response carries season, week and slate_state, then a picks list. Each pick has game_id, pick_team, tier and confidence, the model's win probability for the picked side as a decimal. The sample sorts by that number and prints the three strongest. Check slate_state first, because between a week's last game and Wednesday's new picks the newest slate is a finished one.
Rate Limits, Docs and SDKs
The JSON endpoints allow 60 requests a minute. The docs do not describe a 429 response, so treat any non-200 status as a cue to back off. Caching makes the limit easy to live with. NoPunt caches the current slate for 5 minutes, a historical week for 24 hours and the RSS feed for 30 minutes, so polling faster returns the same data. The terms add that heavy pollers should mirror the data into their own store (checked November 2026).
The docs list no SDK, no OpenAPI file and no changelog. The docs page is the reference, with field notes and an example response for the endpoints. /llms.txt gives AI tools a short site summary. For uptime, GET /api/health returns {"status":"ok"} with a 200, or a 503 when the service reports degraded or stale. Questions go to hello@nopunt.com.
Who NoPunt Is For
- NFL value-finder tools that already carry book prices and want an independent win probability to rank spreads against.
- Survivor, pick'em and confidence pool apps that want a ranked board with win probabilities, through the JSON or the MCP tools.
- Discord bots, dashboards and widgets. Open CORS, no key and five-minute caching make it a low-friction embed.
- Model builders who want a public, graded benchmark to measure their own NFL model against, week by week since 2020.
- AI agents that need an NFL picks source with a hosted MCP server and no credentials to manage.
Verdict
NoPunt does one narrow thing and documents it well. The feed is free and keyless, the picks come with a probability instead of a hunch, every result back to 2020 is a GET away, and the docs flag their own gotchas before you hit them. The seven-tool MCP server is a bonus most free feeds do not have.
Use it as the model side of an NFL comparison, not as a price feed. Pull /api/picks.json, check /api/stats.json for the tier you plan to trust, and read the docs for the field conventions before you compute anything from the margins.
Last updated: ยท How providers get listed
Frequently Asked Questions
/api/picks/{season}/{week}.json returns any regular-season week with the final score and the graded result joined to each pick, and /api/stats.json summarises the record by tier and by season./api/stats.json. When we called it in October 2026 it reported about 1,700 graded picks since 2020, most of them from the 2020 to 2025 backtest, with a straight-up hit rate around 66 percent, higher in the top tiers. Check the tier you plan to use, because accuracy and profit move independently.POST https://www.nopunt.com/api/mcp is a read-only MCP server over Streamable HTTP with no auth. It exposed seven tools when we tested it: the week's picks, a weekly briefing, the survivor, pick'em and confidence pool boards, the ledger summary and a single matchup.pick_side is AWAY, multiply it by -1 to see it from the pick's point of view.NoPunt Alternatives
Similar providers from the directory worth shortlisting alongside NoPunt.
Want a model's read on every NFL game?
No key and no signup. One curl to /api/picks.json returns the week's picks with a win probability for each. The docs list all 18 endpoints with field notes and example responses.
Are You the Team Behind NoPunt?
Show developers you're independently listed on SportsAPIs.dev. Copy this badge onto your site. It links straight back to this profile.