Stat Categories Used in Fantasy History Data
Stat categories are the atomic units of fantasy sports — the specific measurable outputs that platforms convert into points, rankings, and roster decisions. Understanding which categories appear in historical datasets, how they're structured, and where their definitions diverge across sports and platforms is foundational work for anyone doing serious research with fantasy history data.
Definition and Scope
A stat category is any discrete, officially recorded player performance metric that a fantasy platform assigns a point value to or uses as a roster-eligibility filter. Passing yards, strikeouts, blocked shots, power play points — these are stat categories. "Good vibes" and "team leadership" are not.
The scope varies considerably by sport. Fantasy football datasets typically track 20–30 standard stat categories per player per game, ranging from passing attempts to two-point conversions. Fantasy baseball sits at the opposite extreme: a single pitcher's game log can involve 15 or more countable categories, including quality starts, holds, blown saves, and WHIP — the last of which is itself a derived ratio (walks plus hits divided by innings pitched), not a raw counting stat.
That distinction between counting stats and ratio/rate stats matters for historical analysis. Counting stats accumulate — they can be summed across a season, a career, or a dynasty dynasty window. Rate stats like batting average, on-base percentage, or yards-per-carry require weighted aggregation; simply averaging them across games produces distorted results.
For a detailed breakdown of how scoring systems translate these categories into point values, fantasy points scoring systems explained covers the conversion mechanics across formats.
How It Works
Historical fantasy databases store stat categories in one of two structural models:
- Game-level granularity — each row represents a single player-game, with columns for every tracked category (e.g., REC, REC_YDS, REC_TD, FPTS). This is the most flexible format for trend analysis and year-over-year consistency metrics.
- Season-level aggregation — each row represents a full season, with cumulative totals or averaged rates. Faster to query, but it obscures within-season variance that matters for identifying hot streaks, injury effects, or role changes.
Platforms like ESPN, Yahoo, and Sleeper each maintain proprietary stat feeds that map to their own category identifiers. ESPN might label a receiving touchdown as receivingTouchdowns; the same data from an NFL Next Gen Stats feed might appear as rec_td. Cross-platform historical research almost always requires a normalization layer — a mapping table that reconciles these identifiers before any aggregation can happen. Platform-specific historical data for ESPN, Yahoo, and Sleeper addresses those reconciliation challenges directly.
The NFL's official stats, published through its data partnerships, serve as the upstream source for most fantasy platforms. MLB's Statcast system (operated by Baseball Savant, part of MLB Advanced Media) similarly supplies the raw tracking data that flows into fantasy baseball stat columns. Hockey fantasy categories derive from NHL RTSS (Real Time Scoring System) feeds.
Common Scenarios
Scenario 1 — Scoring format research. An analyst comparing historical scoring formats — standard, PPR, half-PPR needs reception count as a separate category from receiving yards. In a standard-scoring dataset that predates the PPR era, receptions may not have been tracked at all, making historical PPR projections impossible without supplementary box score data.
Scenario 2 — Target share and usage tracking. Target share and snap count history requires categories that weren't universally logged before roughly 2012 in publicly accessible datasets. Pre-2012 historical records frequently lack targets, air yards, and snap percentages — categories that are now standard in any serious receiver analysis.
Scenario 3 — Daily fantasy sports (DFS) research. DFS platforms score differently from season-long leagues and sometimes include categories that season-long formats ignore entirely, such as "bonus points for 300+ passing yards" or "double-double bonuses" in basketball. Daily fantasy sports historical data explores how those category variations affect historical lineup research.
Decision Boundaries
Not every stat category belongs in every analysis. Three boundaries are worth keeping clear:
Included vs. excluded categories by platform. Yahoo Fantasy omits certain defensive categories that ESPN tracks (and vice versa). Before pulling historical positional rankings, confirm which categories were active in that platform's scoring system during the target season. A 2015 Yahoo quarterback ranking built on 6-point passing touchdowns will not match ESPN's 4-point equivalent from the same year.
Official stats vs. fantasy-adjusted stats. Some platforms apply proprietary adjustments — fumble recovery attribution, fractional point thresholds, or "quality start" definitions that differ slightly from the BBWAA's standard. These adjustments can affect historical totals by 5–15% for players near scoring thresholds, according to structural descriptions published in platform help documentation.
Raw categories vs. derived metrics. Historical databases increasingly blend raw stat columns with derived fantasy-specific metrics: historical average draft position (ADP) data, positional scarcity scores, and opportunity rates. These are analytical outputs layered on top of raw categories — useful, but they should never be treated as source data in the same way that official box score figures are. Mixing the two in a regression without labeling them separately is a common methodological error in fantasy historical research.
For a broader map of what other dimensions organize historical fantasy datasets beyond stat categories, key dimensions and scopes of fantasy history data provides the structural context.