Discourse, not appearance
MOGCHARTS does not rank how people look. We rank how much the community is talking about how people look. That distinction is everything.
The Discourse Dominance Index (DDI) measures appearance-focused discourse across social platforms. Think of it like a Billboard chart — Billboard doesn't judge whether a song is good, it measures how much people are listening. We don't judge whether someone is attractive, we measure how much — and how — people are discussing their appearance.
Every score on the chart is derived from public, observable data: posts, replies, quotes, threads, and comments across multiple platforms. No editorial opinion, no subjective ratings, no human scoring. The algorithm decides rank — not us.
Charts are generated daily. Each cycle scores the latest 24-hour window with momentum smoothed across prior cycles using exponential moving averages.
Five weighted signals
The composite DDI score is built from five distinct signals, each capturing a different dimension of discourse presence. Together they produce a single score from 0 to 100.
The core signal. Measures what percentage of discourse about a person is specifically about their appearance — face, physique, bone structure, looksmaxxing techniques, mog comparisons. This is what separates MOGCHARTS from a generic attention tracker. A sentiment modifier of ±25% adjusts the weight based on whether discourse is positive or negative.
Raw conversation volume. How many people are mentioning this person at all, regardless of context. This captures general buzz and cultural presence within the ecosystem.
Depth of interaction — replies, quote tweets, thread discussions, comment chains. High engagement means people aren't just mentioning someone, they're having conversations about them. Scored as a 50/50 blend of total engagement volume and per-mention engagement depth.
Rate of change. How fast is discourse growing or declining? A person going from 200 to 800 mentions in a week scores higher on momentum than someone steady at 1,000. Calculated using exponential moving averages across consecutive chart windows.
Specific comparison events — "X mogs Y," direct face-offs, side-by-side discussions. These are the moments when the community explicitly compares appearance between ranked individuals. Weighted lower because they're rarer but very high-signal.
Beyond the base score
After the five signals are combined into a raw composite, three additional multipliers adjust the final DDI score. These modifiers reward organic, high-quality discourse and penalize manufactured or shallow engagement.
Additive bonus. Measures connections to other high-ranking individuals. If a person is discussed alongside other top-ranked profiles, their prestige score increases. This captures community centrality.
Multiplier. Distinguishes between shallow "mention spam" and deep narrative discourse. Single-word mentions score lower; extended discussions, analysis threads, and detailed comparisons score higher.
Multiplier. Analyzes post timing patterns. Organic discourse distributes naturally across time zones. Coordinated bursts (bot farms, pods) cluster unnaturally and get penalized.
Putting it together
The final DDI score is computed once per chart cycle. Here's the simplified formula:
appearance × 0.40 // ± sentiment modifier
+ attention × 0.20
+ engagement × 0.15
+ momentum × 0.15
+ mog_events × 0.10
) + prestige_bonus
× gravity_mult
× temporal_mult
All sub-signals are independently normalized to a 0–100 scale within the tracked roster for each cycle, then weighted and clamped. This means scores reflect relative position within the ecosystem, not absolute thresholds.
Multi-platform collection
Data is collected daily from multiple platforms using a matrix of hundreds of targeted queries. No single platform dominates the overall signal — each person's platform distribution is tracked and displayed alongside their score.
X / Twitter
Primary discourse hub. Posts, replies, quote tweets, and threads. Largest source of mog comparisons and appearance discourse.
TikTok
Video-native discourse. Comments, duets, and stitch videos. Key source for visual reactions and younger demographic signals.
Photo-first platform. Story mentions, reel comments, and post engagement. Lower volume but high-signal for appearance-specific content.
Long-form discussion. Subreddit posts and comment threads. Highest average engagement depth per mention across all platforms.
Multi-language lexicons (English, Portuguese, Spanish) ensure non-English discourse is captured. Collection uses intelligent sampling strategies to balance depth with cost efficiency.
Built to resist manipulation
Any ranking system worth its name must defend against gaming. Our anti-manipulation stack runs multiple layers of defense before any score is finalized.
Content Deduplication
Identical or near-identical posts are detected and collapsed. Copy-paste spam, repost bots, and template-based engagement are filtered before scoring.
Temporal Fingerprinting
Post timing is analyzed against natural distribution patterns. Coordinated posting (pods, bot farms) creates detectable clusters that get penalized via the temporal multiplier.
Reply-Spam Clamping
Excessive replies from single accounts or low-follower networks are clamped. One account replying 50 times doesn't count as 50 mentions.
Anomaly Detection
Sudden spikes that deviate from established baseline patterns trigger review flags. Nuclear thresholds prevent single-event manipulation from distorting scores.
Integrity Scoring
Each data point receives an integrity score based on source quality, account age, and behavioral signals. Low-integrity data is down-weighted before it enters the scoring pipeline.
The 5% threshold
This is the most important constraint in the system. General attention without appearance relevance contradicts the chart's core purpose.
Any individual whose appearance-related discourse falls below 5% of their total mentions is excluded from the chart entirely. This gate ensures MOGCHARTS remains an appearance discourse index, not a generic popularity contest. The threshold is enforced before ranking — no exceptions.
This gate was established after early analysis showed that some highly-discussed individuals had virtually zero appearance-focused discourse. Including them would dilute the chart's meaning and undermine the entire indexing thesis.
What DDI is not
DDI is not a beauty score. We don't evaluate how anyone looks. We measure how much — and how — the community discusses appearance. A high DDI means people are talking about you, not that you're attractive.
We use public data only. Every data point comes from public posts, comments, and threads. No private messages, no private accounts, no scraping behind login walls.
We don't claim endorsement. Being ranked on MOGCHARTS doesn't imply that an individual is aware of, approves of, or participates in this project.
Sentiment adjusts intensity, not eligibility. Negative discourse still counts — but the appearance gate and integrity layers prevent brigading from distorting scores.
From raw data to rank
Each chart cycle follows a fixed pipeline. No manual intervention touches the scoring once collection begins.
Collect
Hundreds of targeted queries run across all active platforms. Video content is transcribed. Raw data is stored unmodified.
Clean & Deduplicate
Duplicate posts, repost bots, and malformed data are filtered. Each record receives an integrity score.
Enrich
Text is classified using a multi-language lexicon. Person detection matches mentions to tracked individuals. Mog events are extracted.
Gate
The appearance threshold is enforced. Individuals below the minimum appearance ratio are excluded before scoring begins.
Score & Rank
The five signals are computed, modifiers applied, and final DDI scores calculated. Momentum is derived from the prior chart. Rankings are published.