UFC Analytics

TL;DR

After filtering for data quality and removing fighters with limited fight history (1-2 fights of data), two plays emerge with strong model consensus:

The Parlay: Norma Dumont (-166) + Talita Alencar (-245)

Norma Dumont vs Ketlen Vieira

  • WT5 Legacy EV: +13.8 (exceptional)
  • Strong consensus: WT5=16, Plain=13
  • WT6 shows model conflict (-17.1% EV) but legacy models strongly favor Dumont
  • Both fighters have extensive data (10+ fights)

Talita Alencar vs Ariane Carnelossi

  • WT5 Legacy EV: +8.2 (strong)
  • Strong consensus: WT5=22, Profit=18, Plain=23
  • WT6 shows negative EV (-5.7%) but legacy models converge
  • Alencar’s grappling dominance (82.8 vs 55.2) creates clear path
Norma Dumont
Win
-166
Strength: 16.0 points
EV: 13.8%
Talita Alencar
Win
-245
Strength: 22.0 points
EV: 8.2%
Total Odds
1.88x
Return on $10.00 Bet
$18.80

Combined odds: 1.88x payout ($10 returns $18.80)

Card Overview:

  • 12 fights total, focused on quality data (excluded fighters with fewer than 3 fights)
  • Legacy models show strong conviction on both parlay legs
  • WT6 model disagreement noted but legacy consensus overrides

Model Predictions Summary

Below is a complete breakdown of all predictions from both our WT6 (current ML) and legacy WolfTickets models. The table shows win probabilities, expected value, odds, and recommendations.

Model Key:

  • WT6 ML: Current generation machine learning model (AutoGluon ensemble with enhanced features)
  • WT6 EV: Expected value from WT6 model (percentage)
  • WT5: WolfTickets legacy model confidence percentile
  • Profit: Profit Model confidence percentile
  • Plain: Plain Model confidence percentile
  • Legacy EV: Expected value calculated by legacy WT5 system (shown as flat number, e.g., 10.7)
  • # of Fights: Total UFC fights for each fighter

Main Card

FighterWT6 MLWT6 EVWT5ProfitPlainLegacy EV# of FightsOdds
Steve Garcia64.3%+11.0%613164.29-138
David Onama35.7%8+108
Waldo Cortes-Acosta57.0%+10.1%61510.79-106
Ante Delija43.0%51-120
Jeremiah Wells52.6%+13.8%646+118
Themba Gorimbo47.4%1-4.36-150
Isaac Dulgarian67.9%-4.4%13172.73-250
Yadier del Valle32.1%61+190
Charles Radtke65.2%+6.7%399-6.15-158
Daniel Frunza34.8%1+124
Allan Nascimento60.1%+7.1%31510-1.04-128
Rafael Estevam39.9%3+100

Preliminary Card

FighterWT6 MLWT6 EVWT5ProfitPlainLegacy EV# of FightsOdds
Timmy Cuamba60.3%+11.5%71110.03-118
ChangHo Lee39.7%12-108
Ketlen Vieira48.0%313+130
Norma Dumont52.0%-17.1%161313.810-166
Alice Ardelean70.6%-11.8%272020-0.83-400
Montserrat Conejo Ruiz29.4%4+285
Phil Rowe45.3%27+154
Seokhyeon Ko54.7%-18.1%1468.51-200
Abdul-Kareem Al-Selwady57.6%+5.9%15131020.61-120
Matheus Camilo42.4%1-106
Talita Alencar67.2%-5.7%2218238.23-245
Ariane Carnelossi32.8%5+186

Key Insights:

  • Norma Dumont shows exceptional WT5 Legacy EV (+13.8) with strong consensus (WT5=16, Plain=13)
  • Talita Alencar shows strong WT5 Legacy EV (+8.2) with triple consensus (WT5=22, Profit=18, Plain=23)
  • Parlay opportunity: Both fighters have 10+ and 3+ fights of reliable data
  • Fighters excluded: Al-Selwady, Camilo, Del Valle, Frunza, Ko (all have 1-2 fights of data)
  • Legacy vs WT6 conflict: Both parlay legs show negative WT6 EV but strong legacy consensus

PARLAY LEG #1: Norma Dumont vs Ketlen Vieira

Preliminary Card • Women’s Bantamweight

Norma Dumont WT6 ✓
vs
Ketlen Vieira
8-2 (10 UFC) | 9-4 (13 UFC)
MODERATE-DISAGREE
WT6 ML Model Current
Norma Dumont 52.0%
Expected Value: -17.1% Odds: -166
Older Models Legacy
Norma Dumont
WT5 (16) Plain (13)
vs
Ketlen Vieira
Profit (3)
Recommendation: BET

Exceptional Legacy EV With Model Conflict

Dumont shows +13.8 WT5 Legacy EV - the highest among fighters with reliable data (10+ fights). Strong consensus from legacy models (WT5=16, Plain=13) suggests historical betting patterns strongly favor Dumont.

WT6 shows severely negative EV (-17.1%), creating a classic model disagreement case. However, both fighters have extensive fight data (Dumont 10 fights, Vieira 13 fights), eliminating data quality concerns. This is a true model philosophy difference, not a sample size issue.

When legacy models show this level of consensus (WT5=16, Plain=13) on fighters with 10+ fights each, historical accuracy has been strong. The -166 odds create value according to legacy projections despite WT6’s contrary assessment.

Radar Chart Analysis

Close matchup with key differentiators:

Vieira’s Advantages:

  • Grappling Defense (94.8): Elite takedown defense - 90% TDD
  • Experience (88.7): 13 fights vs 10, more octagon time
  • Durability & Cardio (78.4): Better conditioning for 3 rounds
  • Physical Edge (80.4): Size advantages

Dumont’s Advantages:

  • Experience (97.9): 5-fight win streak, superior form
  • Striking Defense (92.8): Excellent defensive striking - 65.5% defense
  • Striking Offense (72.2): More offensive output
  • Grappling Defense (80.4): Strong 72% TDD

Key Technical Points: Both fighters show low finishing threat (Vieira 42.3, Dumont 40.2), meaning this fight almost certainly goes to decision. Dumont’s 5-fight win streak and superior striking defense (92.8 vs 58.8) create edges in a decision fight. Vieira’s grappling defense is elite, but Dumont’s wrestling is adequate (64.9 grappling offense) to compete.

The Fight

Dumont is riding a 5-fight win streak with excellent striking defense (65.5%). Vieira has lost 2 of her last 3. In decision fights (which this will be), Dumont’s current form and striking advantages accumulate over 15 minutes.

Vieira’s elite grappling defense (94.8) keeps the fight standing where Dumont wants it. Dumont’s striking offense (72.2) and defense (92.8) create systematic advantages in the standup battle.

The Bet

BET Dumont (-166) as Parlay Leg #1.

The +13.8 WT5 Legacy EV is exceptional for fighters with extensive data. Strong consensus (WT5=16, Plain=13) overrides WT6’s negative assessment. Both fighters have 10+ fights, eliminating data quality concerns.

This is a pure legacy model play where historical betting patterns see clear value that modern ML misses.

Risk Factors:

  • WT6 severely negative (-17.1% EV) - model disagreement
  • Vieira’s grappling defense (94.8) is elite
  • Close radar metrics suggest competitive fight
  • Decision fight (low finishing threat) increases variance

PARLAY LEG #2: Talita Alencar vs Ariane Carnelossi

Preliminary Card • Women’s Strawweight

Talita Alencar WT6 ✓
vs
Ariane Carnelossi
2-1 (3 UFC) | 3-2 (5 UFC)
STRONG CONSENSUS
WT6 and all legacy models agree on the winner
WT6 ML Model Current
Talita Alencar 67.2%
Expected Value: -5.7% Odds: -245
Older Models Legacy
Talita Alencar 3 models support
WT5: 22
Profit: 18
Plain: 23
Recommendation: BET

Strong Legacy Consensus Overrides WT6

Alencar shows +8.2 WT5 Legacy EV with exceptional triple consensus (WT5=22, Profit=18, Plain=23). When all three legacy models converge at these percentiles, historical accuracy is strong.

WT6 shows negative EV (-5.7%), but the legacy consensus is too strong to ignore. Alencar has 3 fights of data (minimum threshold), Carnelossi has 5 fights. Both have sufficient data for reliable predictions, unlike many fighters on this card.

The -245 odds appear steep, but legacy models suggest Alencar’s grappling dominance creates a larger skill gap than the market recognizes.

Radar Chart Analysis

Alencar holds systematic grappling advantages:

Alencar’s Advantages:

  • Durability & Cardio (100.0): Perfect conditioning - won’t fade
  • Grappling Defense (98.3): Elite 100% TDD, submission defense
  • Grappling Offense (82.8): Strong wrestling and control
  • Striking Defense (81.0): Solid defensive fundamentals

Carnelossi’s Advantages:

  • Finishing Threat (94.0): Dangerous finisher - 60% finish rate
  • Physical Edge (50.9 vs 0.9): Significant size advantages
  • Striking Offense (25.0 vs 9.5): More offensive output (both low)

Key Technical Points: Alencar’s grappling profile (82.8 offense, 98.3 defense) creates a massive mismatch against Carnelossi’s poor grappling defense (10.3 percentile). Carnelossi’s finishing threat (94.0) is real, but she needs to land clean strikes. Alencar’s durability (100.0) and striking defense (81.0) make finishing her extremely difficult.

The Fight

Alencar is a BJJ world champion with elite grappling. Her 98.3 grappling defense (100% TDD) means Carnelossi won’t take her down. Her 82.8 grappling offense vs Carnelossi’s 10.3 grappling defense creates a clear path to control and decision victory.

Carnelossi’s only path is knockout. Her 94.0 finishing threat (60% finish rate, 3 finishes in 5 UFC fights) is legitimate. But against Alencar’s 81.0 striking defense and 100.0 durability, landing a finish is extremely difficult.

Alencar’s 100.0 durability percentile means she won’t fade in Round 3. Carnelossi’s 12.1 durability creates a massive gap. If this fight goes deep, Alencar dominates.

The Bet

BET Alencar (-245) as Parlay Leg #2.

The +8.2 WT5 Legacy EV with triple consensus (WT5=22, Profit=18, Plain=23) is exceptional. Both fighters have adequate data (3+ fights), eliminating sample size concerns.

Alencar’s grappling dominance (82.8 vs 10.3 defense) creates a clear technical path to victory. Legacy models see a larger skill gap than -245 odds reflect.

Risk Factors:

  • WT6 negative (-5.7% EV) - model disagreement
  • Carnelossi’s finishing threat (94.0) is dangerous
  • Alencar only has 3 fights of data (minimum threshold)
  • If Carnelossi lands clean early, upset potential exists

Other Fights Overview

Quick recaps of remaining fights with radar chart analysis:

Steve Garcia vs David Onama

Main Card • Featherweight

Garcia (9 fights) shows elite finishing threat (98.9) and striking defense (88.8), but terrible cardio (29.5). Onama (8 fights) has exceptional striking offense (96.2) and durability (87.2). Garcia needs an early finish; if Onama survives to Round 3, his cardio advantage dominates. Both models show positive EV (WT6 +11.0%, WT5 +4.2) with Profit Model support (13).


Jeremiah Wells vs Themba Gorimbo

Main Card • Welterweight

Wells (6 fights) vs Gorimbo (6 fights) features elite grappling on both sides (96.8 vs 98.1). Wells has massive finishing threat advantage (92.5 vs 59.1) but poor physical profile (40.5) at age 39. Gorimbo holds physical edge (93.8) and striking advantages. Both show terrible cardio (21.0 vs 23.3). WT6 favors Wells (+13.8% EV, 52.6% win probability) despite underdog odds (+118), but legacy models slightly favor Gorimbo.


Allan Nascimento vs Rafael Estevam

Main Card • Flyweight

Nascimento (4 fights) vs Estevam (3 fights). Estevam dominates radar metrics: exceptional durability (99.3), elite grappling offense (99.3), and physical edge (95.4). Nascimento has elite striking defense (94.8) but catastrophic grappling defense (14.4). WT6 favors Nascimento (+7.1% EV) with strong Profit Model (15), suggesting Nascimento’s striking defense keeps fights standing where his grappling weakness is neutralized.


Timmy Cuamba vs ChangHo Lee

Preliminary Card • Bantamweight

Cuamba (3 fights) vs Lee (2 fights - limited data). Cuamba shows exceptional durability/cardio (89.8) vs Lee’s poor conditioning (54.5). Lee has catastrophic striking defense (17.3) despite elite finishing threat (96.6). Dual positive EV (WT6 +11.5%, WT5 +10.0) with Profit Model support (11). Lee’s limited data (2 fights) creates statistical uncertainty.


Key Insights

Data Quality Filtering

After reviewing all model predictions and filtering for data quality, fighters with 1-2 fights of data were excluded: Al-Selwady, Camilo, Del Valle, Frunza, Ko (all have limited fight history creating unreliable statistical profiles).

Legacy Model Consensus

Both Dumont and Alencar show strong legacy model consensus:

  • Dumont: WT5=16, Plain=13 (exceptional consensus), +13.8 WT5 EV
  • Alencar: WT5=22, Profit=18, Plain=23 (triple consensus), +8.2 WT5 EV

Despite WT6 showing negative EV on both (-17.1% and -5.7%), the legacy model agreement is too strong to ignore when both fighters have reliable data (10+ and 3+ fights respectively).

The Model Conflict Question

When WT6 and legacy models disagree, it creates uncertainty. However, both parlay fighters have sufficient data to trust predictions:

  • Dumont (10 fights) vs Vieira (13 fights) - both extensive records
  • Alencar (3 fights) vs Carnelossi (5 fights) - both above minimum threshold

The legacy consensus on fights with adequate data historically shows strong accuracy.


Final Recommendation

PARLAY: Norma Dumont (-166) + Talita Alencar (-245)

  • Combined odds: 1.88x payout ($10 returns $18.80)
  • Unit sizing: 1 unit
  • Both legs show strong legacy model consensus
  • Both matchups feature fighters with reliable data samples
  • WT6 disagreement noted but legacy consensus overrides

Risk Factors:

  • WT6 shows negative EV on both legs (model conflict)
  • Dumont fight likely goes to decision (low finishing threat)
  • Carnelossi has dangerous finishing threat (94.0)
  • Parlay requires both legs to hit

What’s Next

This Fight Night card required careful data quality filtering. After excluding fighters with 1-2 fights of data, the legacy models showed strong conviction on Dumont and Alencar despite WT6 disagreement.

Good luck!

-Chris@WolfTickets.AI