Methodology

BEAST Challenge Rating Methodology

Last updated: 2026-05-22

Binary Elo Algorithm for Survivor Trials

BEAST is a companion rating system to SHALLOW that measures physical and competitive performance in Survivor challenges. While SHALLOW tracks strategic positioning through votes and survival, BEAST focuses exclusively on challenge outcomes.


What BEAST Measures

Important: Challenge counts include both immunity and reward challenges. A player's "Individual: 5/8" means they won 5 of 8 individual challenges—not 5 individual immunities. Reward challenges are weighted equally with immunity challenges.

Individual Challenges

When a player wins an individual challenge (immunity or reward), they are matched against every other competitor in that challenge. The winner gains rating points against each opponent; all losing competitors lose points to the winner.

This creates a proper zero-sum system: winning a challenge against 10 opponents generates 10 separate matchups, each contributing to the rating calculation.

Hybrid challenges (Team/Individual) are included when a player wins the individual component.

Redemption Island Duels

Seasons with Redemption Island (US22, US23, US27) feature duels where voted-out players compete to stay in the game. Duels are included in BEAST ratings but displayed separately from regular individual challenges because:

Example: Ozzy in Season 23 won 6 duels (beating 8 total opponents across all duels) for approximately +39 Elo points, while his single Final 5 immunity win (beating 4 opponents) earned +19 Elo points. Six duel wins yielded roughly the same Elo gain as two immunity wins at a typical merge challenge.

Duels are tracked separately in the UI (shown as "Duels: 6/6") so you can distinguish between a player who dominated Redemption Island versus one who won multiple merge immunities.

Team Challenges

Pre-merge tribal challenges (both immunity and reward) are included with reduced weight. When a tribe wins a challenge, each member of the winning tribe is matched against each member of the losing tribe(s). The K factor (rating volatility) for team challenges is lower than individual challenges, reflecting that team outcomes have more noise—a weak competitor can be carried by strong tribemates.


The Rating System

Like SHALLOW, BEAST uses the Elo methodology:

  1. Winners gain points; losers lose points
  2. Upsets cause larger swings - beating a higher-rated opponent earns more points than beating a lower-rated one

All players begin at 1500. Ratings are zero-sum.

K Factors

These values were optimized by testing prediction accuracy across all seasons.


Challenge Tiers

Players are assigned tiers based on z-scores, with names reflecting physical competition:

Tier Z-Score Name Interpretation
S ≥ 2.0 Beast Top ~2% - Elite challenge performers
A 1.0 to 2.0 Threat Top ~15% - Challenge threats
B 0 to 1.0 Competitor Above average
C -1.0 to 0 Average Middle of the pack
D < -1.0 Underdog Below average in challenges

What BEAST Does Not Measure


Challenge Classification Notes

Challenge counts in BEAST may differ from other sources due to how we classify hybrid challenges.

Hybrid challenges (where teams compete first, then individuals compete within the winning team) are classified based on the survivoR data package. Some challenges that appear individual on the show are classified as "Team/Individual" hybrids because they include a team component.

Example: Kyle Ostwald (Season 47) shows 4/6 individual challenges in BEAST. Some sources may count additional challenges, but those may have team components that our data classifies differently. Kyle's 67% win rate in pure individual competition reflects his actual 1v1 performance against other merge competitors.

Challenges are only counted when the player actively participated (not eliminated or sitting out).


Interpreting Individual vs. Duel Stats

The player cards show challenge statistics:

Both contribute to the BEAST rating, but a player with 5 immunity wins faced more opponents and earned more Elo than a player with 5 duel wins.

Note: Team challenge stats are shown in tribe snapshots where they're more relevant (showing how a player performed while on a specific tribe).


Case Studies

The Challenge Beast: Joe Anglim

Joe Anglim ranks #1 in BEAST with a rating of 1729 (S-tier). Across three seasons, he won 7 of 13 individual challenges (54%) and 24 of 32 team challenges (75%). His individual challenge dominance made him an immediate target post-merge in every season.

Joe's SHALLOW tells a completely different story: he ranks #947 with a rating of 1376. He was frequently on the wrong side of votes precisely because his challenge threat status made him a target. The two ratings capture fundamentally different dimensions of Survivor performance.

The Strategic Mastermind Who Can't Win Challenges: Cirie Fields

Cirie Fields ranks #8 in SHALLOW (1683) but #831 in BEAST (1365). Her individual challenge record is 1 win in 35 attempts—a 3% win rate. Her team record (18/45) reflects being on losing tribes.

Cirie demonstrates that challenge performance and strategic excellence are largely independent. She built one of the greatest Survivor legacies without ever winning individual immunity, relying entirely on social positioning to survive.

Recent Dominance: Kyle Ostwald (Season 47)

Kyle Ostwald from Season 47 ranks #3 in BEAST (1682) after just one season. He won 4 of 6 individual challenges (67%)—the highest win rate among players with 5+ individual challenges. His challenge dominance in the new era rivals the legends.

His SHALLOW is #511 (1481)—solid but unremarkable. Kyle shows that even in modern Survivor, challenge beasts emerge whose physical game far outpaces their strategic positioning.

The Complete Player: Rachel LaMont (Season 47)

Rachel LaMont won Season 47 and shows balanced excellence: #32 in SHALLOW (1625) and #45 in BEAST (1588). She won 4 of 11 individual challenges (36%) and 6 of 8 team challenges (75%).

Rachel demonstrates what a "complete game" looks like in the data—strong enough in challenges to win when needed, strategic enough to control her fate otherwise. Her profile mirrors Boston Rob's balanced approach in a single season.

Balanced Excellence: Boston Rob

Rob Mariano ranks #4 in BEAST (1663) and #10 in SHALLOW (1681)—elite in both dimensions. Across six seasons, he won 9 of 23 individual challenges (39%) and 23 of 44 team challenges (52%).

Boston Rob is one of few players to achieve top-tier status in both ratings, reflecting his complete game across physical, strategic, and social dimensions.


Relationship to SHALLOW

BEAST and SHALLOW measure different aspects of Survivor:

Aspect SHALLOW BEAST
Primary input Votes, survival, jury Challenge wins/losses
Measures Social-strategic positioning Physical competition
High correlation with Winning, FTC appearances Individual immunity

Some players excel at both (Boston Rob, Rachel LaMont, Mike Holloway). Others show stark divergence—Joe Anglim ranks #1 in challenges but #947 in strategy, while Cirie Fields ranks #8 in strategy but #831 in challenges.

Neither rating is "better"—they capture different skills that contribute to Survivor success.


Data Sources

Challenge data sourced from the survivoR package.


Methodology Notes

BEAST is under active development alongside SHALLOW. Current version tracks: - Individual immunity challenges - Individual reward challenges - Redemption Island duels (US22, US23, US27) - Hybrid challenges (Team/Individual, Tribal/Individual)

Fire-making challenges at Final Four are included in SHALLOW, not BEAST, as they determine elimination rather than reward/immunity.

Team/tribal challenges are processed in the model but not displayed in the main rankings (shown only in tribe snapshots).


See also: SHALLOW strategic rating methodology, the FAQ, and the full rankings.