Overview
This is the raw stats document behind the Treat-ies. Two organizational views:
- By Category โ every stat shown as a ranked table across all managers, with definitions inline.
- By Manager โ one card per manager showing all their stats and where they rank in the league.
Sources: 03_League_History/Sleeper_2023-2025/sleeper_data.json (matchups, transactions, drafts, rosters, 2023โ2025) + 02_Chat_Corpus/treatment_room_corpus.csv (Treating thread, non-reaction messages). Cole and Curtis are excluded โ no Sleeper-era league history. Rodger and Will Holder are former members shown with reduced opacity for historical context.
Color legend (across all ranked tables):
#1 in league
#2 or #3
Second-worst
Worst
Former member (Rodger / Will Holder)
Rankings are computed across active managers (Mike, Hayden, Greg, Tyler, Ricardo, Ryland, Olivier, Ray, Will Purvis). Will Purvis only has a half-season in 2025 โ interpret his ranks with caution.
League-Wide Single-Game Records
| Record | Manager | Value | When |
| Highest single-week score | Hayden | 211.1 | 2024 wk 16 |
| Lowest single-week score | Ricardo | 30.4 | 2025 wk 18 |
| Largest blowout win (margin) | Mike | +117.3 | 2024 wk 7 |
| Largest blowout loss (margin) | Will Holder | โ117.3 | 2024 wk 7 |
| Closest game ever | Will Holder d. Ray | 136.3 โ 136.6 (0.34 margin) | 2024 wk 13 |
| Highest-scoring loss | Greg (lost to Hayden) | 175.9 (vs 177.1) | 2024 wk 17 |
| Biggest single-player start (1 week) | Jahmyr Gibbs for Mike | 55.9 fpts | 2025 wk 12 |
| Largest single FAAB bid | Ryland on Sam Darnold | $134 | 2024 wk 1 |
Most-Traded Players in League History (Sleeper era)
How it's measured: count of times each player changed hands in any completed trade, 2023โ25 (each appearance in a trade's "adds" dict = +1).
| Player | Times Traded |
| Derrick Henry | 15 |
| Geno Smith | 12 |
| Calvin Ridley | 10 |
| Jerry Jeudy | 9 |
| Matthew Stafford | 9 |
| Kirk Cousins | 9 |
| Terry McLaurin | 9 |
| Mark Andrews | 9 |
| Brandon Aiyuk | 9 |
| Michael Pittman | 9 |
By Stat Category
1. Record & Scoring
How it's measured: Wins / Losses, Points For (PF), Points Against (PA), and Points Per Week (PPW). Includes regular season + playoff games. PPW is total points scored รท games played. Sorted by PPW (higher is better).
2. Lineup Efficiency
How it's measured: For every week, the highest legal lineup is computed from the manager's entire roster (1 QB, 2 RB, 2 WR, 1 TE, 1 FLEX, 1 SUPER-FLEX, 1 DEF). Lineup Eff % = points actually started รท best possible from the roster, averaged across all weeks. Bench Pts / Wk = average points left unstarted on the bench each week. Higher Eff % = better lineup-setter.
3. Variance / Consistency
How it's measured: Standard deviation of weekly scores across the dataset. Lower stdev = more consistent scoring (closer to manager's average every week). CV % (coefficient of variation) = stdev รท mean ร 100 โ normalizes for scoring level so a high-scoring team and a low-scoring team can be compared directly. Sorted by stdev ascending (most consistent first).
4. All-Play / Median Wins
How it's measured: All-play record: for each regular-season week, every manager is hypothetically matched against every other manager. You "win" against anyone you outscored that week. Removes schedule luck (you can't blame an unlucky bad-week matchup). Median wins: for each week, you "win" if your score is above the league's median score that week. Median weekly rank: your average rank (1 to 10) among managers each week โ lower is better. #1 / Last weeks: regular-season weeks finished as top or bottom scorer.
5. Playoff Performance โ PPW Gap
How it's measured: Average points scored per week in playoff games minus average per week in regular-season games. Positive = scores more in playoffs (clutch). Negative = scores less (choke). Counts only games with a confirmed playoff-bracket matchup_id (excludes consolation rounds). Sorted by gap descending.
6. Trade Volume
How it's measured: Completed trades and the number of distinct trade partners across the 3 Sleeper seasons. Sorted by trade count descending.
7. Trade Quality (Hindsight)
How it's measured: For each completed trade: sum the fantasy points the acquired players have scored since the trade date, minus the points the given-up players have scored since the trade date. Both forward-looking, both running through the end of 2025. Excludes draft picks (those are captured separately in Pick Flow). Net Fpts / Trade = manager's career trade net รท number of trades. Caveat: trades made earlier in the dataset accumulate more "after fpts" than later trades, so volume traders are slightly penalized.
8. Pick Flow Through Trades
How it's measured: For each draft pick that changed hands in a completed trade, count one to the receiver, one to the giver. Weighted values weight R1 picks as 4, R2 as 3, R3 as 2, R4+ as 1. Net Picks = picks received minus picks given. Weighted Net = same with weighting. Positive = accumulated pick value. From Dead Teams = picks received specifically from Rodger or Will Holder (the league's two destroyed rosters who tanked and traded their futures).
9. Roster Churn / Waivers
How it's measured: Count of player adds and drops via waivers / free-agency / commissioner moves (not trades). FAAB Spent = total Free-Agent-Acquisition-Budget dollars spent on successful waiver claims, all 3 seasons combined.
10. FAAB Efficiency
How it's measured: Total fantasy points produced by players acquired via FAAB-waiver bids (from the week of acquisition through end of dataset), divided by total FAAB dollars spent. Higher = more value per dollar. Captures both bid discipline (didn't overpay) and roster scouting on free agents.
11. Player Retention
How it's measured: Median holding period: for every player a manager acquired and later dropped/traded, count days held. Median across all such holds (currently-rostered players excluded). Players 700+ days: count of currently-rostered players the manager has held continuously since the start of the dataset (capture window ~828 days). Higher = more loyal / less churn-prone.
12. Rookie Draft Performance
How it's measured: Rookie picks made across the 2023, 2024, and 2025 rookie drafts. Yr1 fpts = regular-season points the picked player scored in their first NFL season (the season immediately following the draft). Yr1 hits = picks that produced 100+ fpts in their rookie year. Yr1 fpts / pick = total yr-1 fpts รท picks made. Higher = better rookie evaluator.
13. Schedule Strength
How it's measured: For each regular-season game, the opponent's season-average points-per-week. Averaged across all the manager's regular-season opponents. Higher = harder schedule (faced stronger opponents). Lower = easier road to the playoffs.
14. Clutch โ Win % Above Own Avg vs Below
How it's measured: Take each manager's own career-average score. Bucket every game into "above avg" or "at-or-below avg." Report W-L in each bucket. Win % above = win rate when scoring above own avg. Win % below = win rate when scoring at-or-below own avg. A manager who wins below-avg games either has good schedule luck or wins close ones; one who never does is purely score-driven.
15. Chat Behavior
How it's measured: Messages and reactions in the main "Treating" group chat (2022โ2026). React:Msg = total tapback reactions given รท messages sent. >1 = reacts more than posts. Msgs / Trade = total chat messages รท completed trades โ a "talk-to-action" ratio (lower = more efficient communicator).