About
How the ratings work · How the predictor works · What's coming next
What is CageDex?
CageDex is an independent ratings and prediction engine for UFC fighters. It's not affiliated with the UFC, and it's not trying to replace the official rankings — it's asking a different question: what does the fight record actually say?
Every fighter in the database has a rating built entirely from results. Who they beat, who beat them, how decisive those wins were, and how recently. The system has processed every UFC event from 1993 to the present — over 8,701 fights — and uses that data to rank fighters, compare them across divisions, and power a machine learning predictor for upcoming matchups.
No committees. No narratives. Just the math.
How ELO Ratings Work
Every fighter in the database starts at 1500 — a neutral baseline meaning "we have no information on you yet." From there, ratings move based on results. Beat a highly-rated opponent and you gain a lot. Grind out a split-decision over a low-rated fighter and you gain much less.
The core idea: your rating reflects the quality of your wins and losses, not just the number of them. A 20-fight veteran feasting on journeymen might be rated lower than a 7-fight prospect who's already beaten two ranked opponents.
[ Finish Type Matters ]
A knockout or submission is worth more than a unanimous decision, and a unanimous decision more than a split. Finishing a fight is a stronger signal of dominance. Fighters who consistently finish get more credit.
[ New Fighters Move Faster ]
Early ratings move faster while the system is still learning a fighter. The PROV tag on rankings means 3 or fewer actual fights in that displayed division, including fighters who just moved weight classes. Treat those rankings as useful, but still small-sample.
[ Title Fights Carry Extra Weight ]
Championship bouts have bigger impact. The stakes are higher, the competition certified elite, and the system treats those results accordingly.
[ Inactivity Decay ]
A rating earned two years ago and undefended isn't worth as much as one earned last month. Division ratings inactive for 12+ months get a current freshness adjustment toward 1500. Fight results still update from fight-night skill signals, not a pre-fight layoff penalty.
[ Confidence Scores ]
Every rating has a confidence score. High confidence = recently active with enough fights. Low confidence flags provisional fighters or those deep in a layoff. The number is still informative — treat it with appropriate skepticism.
[ P4P Movers ]
Changing divisions does not erase a fighter's history, but current ranking surfaces use current-division Elo. Transfer logic can seed the new rating, then P4P compares that current-division rating against the new division's field. A moved fighter can appear in the new division before 3 actual fights there, with the PROV tag carrying the small-sample warning.
[ P4P Head-To-Head ]
Pound-for-pound starts with cross-division Elo, then checks direct results between fighters on the list. If one ranked fighter has beaten another and never lost back to him, the winner cannot sit below him. Split series and circular MMA math go back to the normal ranking order.
How the Fight Predictor Works
A machine learning model trained on historical UFC fights. Given two fighters, it outputs a win probability for each — not a gut feeling or a narrative, just what the data says has historically happened when fighters with similar profiles match up.
[ Training Data ]
Trained on UFC fights through end of 2022, tested on 2023 onward — fights it had never seen before. That temporal separation is how you check whether a model learned something real versus memorizing history.
[ 37 Input Signals ]
Striking output (SLpM), striking absorption (SApM), opponent quality, Elo rating, recent form (last 5), finish rate, days inactive, age, reach, stance, takedown stats, rolling knockdown rate, and rolling submission attempt rate. Public matchup predictions are model-only; Vegas odds are not used as a live input.
[ Accuracy ]
Predictor v6 is tested on a 822-fight temporal holdout from 2023 to present. It picks the right winner 69.3% of the time, versus 57.5% for the Elo-only baseline.
On the current 572-fight Best Fight Odds benchmark, the model is 71.7% accurate versus 71.0% for the opening-line favorite. It called 72 of 166 market upsets.
Real numbers, not cherry-picked. The model is solid. It's also not infallible — MMA is chaotic, and no model captures a fighter's chin, their corner, or whether they had a bad camp.
[ Vegas Policy ]
Public matchup predictions do not use Vegas odds as an input. Best Fight Odds lines are used for benchmarking, event context, and betting-edge analysis where available. That keeps CageDex predictions independent from the market, while still letting Events and betting analysis compare the model against Best Fight Odds when lines exist.
[ What the % Means ]
When Fighter A wins 67%, it means: historically, when fighters with profiles like these matched up, the Fighter A analog won ~⅔ of those fights. A statement about distributions — not a guarantee about Saturday night.
What's Coming
The site is in active development. Post-launch pipeline:
Surface picks from top MMA analysts alongside the model probability and Vegas line. Three signals at once: the data, the market, and the minds who watch this sport for a living.
When two fighters have history, it should be visible. A fight timeline that gives every rematch the context it deserves.
The model knows every fighter's Elo and stats at any point in history. GSP vs. Khabib at their respective peaks. Fedor vs. Ngannou in 2009. It's a natural extension of the data.
Every great fighter has a great corner. CageDex will rate coaches the same way it rates fighters — by results. Every time a fighter wins, their coach's rating goes up. Every time they lose, it goes down. A ranking of the most proven corners in MMA.
Fighter profile pages currently use data-only layouts. The plan is to bring in real headshots for every fighter in the database, so the numbers have a face.
Key Terms
- ELO Rating
- Starts at 1500. Reflects performance quality weighted by opponent level and finish type. Top P4P contenders typically sit 1650–1720+.
- ELO Win Probability
- Raw prediction from the Elo gap alone — one of 37 signals in the full model. Layering in striking stats meaningfully improves over Elo alone.
- Z-Score (P4P)
- Standard deviations above divisional mean. Fixes the cross-division comparison problem — raw Elo ratings can't be compared across weight classes. Movers are scored from their current-division Elo.
- PROV Tag
- Small-sample warning on active rankings. Applies to non-champions with 3 or fewer actual fights in the displayed division.
- Confidence Score
- 0.0–1.0. High = recently active, enough fights. Low = new, inactive, or provisional. Doesn't affect the math, just tells you how much to trust the number.
- Recent Form
- Win rate over last 5 fights. Recency matters — a fighter on a streak gets a bump regardless of career Elo.
- Finish Rate
- % of wins by KO/TKO/sub. High finish rate = more volatility. Factored into matchup dynamics.
- Days Inactive
- Days since last bout. Used as a direct predictor feature and for current-rating freshness. Ring rust is real.