LitFantasy

Methodology & Data Sources

LitFantasy combines public fantasy football data sources into a single, research-friendly view. We believe in giving credit where it is due and being transparent about where our numbers come from.

Where our data comes from

  • Sleeper — Player identity and metadata (player IDs, names, positions, teams, age, years of experience, and roster status), in-season injury designations, and trending add/drop counts come from the Sleeper API. We also use Sleeper to resolve the current NFL season and week. The “Trend” column on our player tables is Sleeper waiver-wire data — the most-added and most-dropped players of the last 24 hours among Sleeper users, which is not the same as activity across all fantasy leagues. The column credits Sleeper in place, as their documentation requests. Learn more at sleeper.com.
  • nflverse — Weekly and season statistics, historical production, depth charts, and schedule/bye-week data come from the nflverse-data project, which also provides the identifier crosswalk we use to join players across sources. Data from nflverse is licensed under CC-BY 4.0. LitFantasy modifies the source data by aggregating, joining, and normalizing it across sources for use in our ranking and research features.
  • Fantasy Football Calculator — We use FFC's public ADP REST API for Average Draft Position data across PPR, half-PPR, standard, 2-QB, and dynasty formats. Used with FFC's ADP REST API terms. Learn more at fantasyfootballcalculator.com.
  • ESPN — We use ESPN's public endpoints for news headlines and to match players to their ESPN identifiers, with team schedule data as a fallback when other sources are unavailable.
  • News publishers — Player news headlines are aggregated from public RSS and syndication feeds published by Pro Football Rumors, Yahoo Sports, ESPN, FOX Sports, USA TODAY, and CBS Sports. We link back to the original article and never reproduce full stories — headlines and short excerpts only, with credit to the publisher.

Our projections are our own

The projections you see on LitFantasy are produced by our own projection engine, not licensed or resold from a third party. We build them from the historical and weekly statistics described above — usage and role signals such as target share and depth chart position, efficiency and volume trends across multiple seasons, opponent strength, and adjustments for injury and depth chart news. Where a player has little history to work from, such as rookies, we shrink our estimates toward positional baselines rather than overfitting a small sample.

Our projections are anchored to the market. Purely historical baselines carry biases that drafters have already priced in: a hot finish to last season extrapolates too far, and young players with small samples stay pinned below what the market expects of them. So we anchor each baseline to a fitted curve relating ADP to per-game points for that position and scoring format, and keep our own signal as a bounded deviation from that anchor. The effect fades out for players drafted well outside the fitted range, and players with no ADP at all are left untouched by this step.

This is worth knowing when you compare our numbers to ADP: the two are not independent, so a player projected close to their draft position is partly the anchor doing its job rather than confirmation from a separate signal. The deviations are where our model is actually speaking. Projections are estimates, not predictions.

How often data refreshes

Core player data — rosters, statistics, injuries, ADP, and trending — refreshes every six hours. Projections are recomputed every two hours, and player news is polled hourly during the season. This keeps rankings and player context current without hammering upstream APIs; we cache upstream responses and refresh on a fixed schedule rather than fetching per page view.

How we present the data

LitFantasy doesn't roll everything into a single ranking number. Instead, we show projections, ADP, positional context, and other signals side-by-side, with color-coded percentile gradients on each column so you can quickly see how a player compares to their peers on any given metric. You weigh the signals; we just make them easy to read.

LitFantasy is not affiliated with or endorsed by Sleeper, ESPN, the NFL, or any of the publishers listed above. All trademarks are the property of their respective owners.