UFC 36: Worlds Collide: Predictions & Analysis
UFC 36: Worlds Collide lands on Friday, March 22, 2002 in Las Vegas, Nevada, USA with 8 bouts on the card. The card is headlined by a championship fight. Below is our fight-by-fight breakdown, combining Elo ratings, rolling statistical trends, style matchup data, and betting market context into a pick for every bout.
Quick Picks
| Matchup | Pick | Confidence | Prob |
|---|---|---|---|
| Josh Barnett vs Randy CoutureHeavyweight | Josh Barnett | Toss-up | 54% |
| Pedro Rizzo vs Andrei ArlovskiHeavyweight | Pedro Rizzo | Confident | 66% |
| Matt Hughes vs Hayato SakuraiWelterweight | Matt Hughes | Strong | 79% |
| Matt Lindland vs Pat MiletichMiddleweight | Matt Lindland | Lean | 62% |
| Evan Tanner vs Elvis SinosicLight Heavyweight | Evan Tanner | Strong | 76% |
| Frank Mir vs Pete WilliamsHeavyweight | Frank Mir | Lean | 63% |
| Matt Serra vs Kelly DullantyLightweight | Matt Serra | Confident | 66% |
| Sean Sherk vs Jutaro NakaoWelterweight | Sean Sherk | Toss-up | 55% |
Fight-by-Fight Breakdown
Josh Barnett vs Randy Couture
The Heavyweight matchup features Josh Barnett (6-3) taking on Randy Couture (16-7). Barnett is the bigger frame at 6'3" with a 3-inch reach advantage.
Barnett is rated at 1454 — 207 points above Couture's 1248. Gaps this large usually mean one fighter has been consistently beating better opponents. Couture has won 3 straight.
The style clash matters here: Barnett is comfortable adjusting on the fly, mixing strikes and grappling as openings appear, while Couture looks to close distance and put the fight on the mat. In our database, wrestlers own a 59% win rate against all-rounders, giving Couture the stylistic edge.
A few statistical edges stand out. Couture throws significantly more leather — a 0.1 sig. strike per minute gap. Couture is far more active with takedowns, averaging 0.4 more per 15 minutes. Barnett has tighter striking defense, making opponents miss more often.
The Pick: Josh Barnett over Randy Couture. This is essentially a pick'em. The model nudges toward Barnett at 54%, but there's almost nothing separating these two.
Pedro Rizzo vs Andrei Arlovski
The Heavyweight matchup features Pedro Rizzo (8-5) taking on Andrei Arlovski (23-17).
Rizzo is rated at 1268 — 410 points above Arlovski's 858. Gaps this large usually mean one fighter has been consistently beating better opponents.
Stylistically this is Rizzo's striker game against Arlovski's all-rounder approach. Rizzo brings a versatile approach, while Arlovski is comfortable adjusting on the fly, mixing strikes and grappling as openings appear. Historically these archetypes are dead-even when they collide.
A few statistical edges stand out. Rizzo throws significantly more leather — a 0.3 sig. strike per minute gap. Arlovski is far more active with takedowns, averaging 1.0 more per 15 minutes. Rizzo has tighter striking defense, making opponents miss more often.
The Pick: Pedro Rizzo over Andrei Arlovski. We're leaning Rizzo here at 66%, a solid but not overwhelming edge.
Matt Hughes vs Hayato Sakurai
The Welterweight championship matchup features Matt Hughes (18-6) taking on Hayato Sakurai (0-0).
Hughes is rated at 1273 — 331 points above Sakurai's 942. Gaps this large usually mean one fighter has been consistently beating better opponents.
A few statistical edges stand out. Hughes throws significantly more leather — a 4.0 sig. strike per minute gap. Hughes is far more active with takedowns, averaging 6.8 more per 15 minutes. Sakurai has tighter striking defense, making opponents miss more often.
The Pick: Matt Hughes over Hayato Sakurai. The model is firm on this one: Hughes at 79%.
Matt Lindland vs Pat Miletich
The Middleweight matchup features Matt Lindland (8-3) taking on Pat Miletich (8-1).
There's a real Elo separation here: Lindland at 1239 versus Miletich at 1150. That 89-point gap typically reflects a meaningful difference in recent quality of competition and results.
The style clash matters here: Lindland looks to close distance and put the fight on the mat, while Miletich is most dangerous on the ground, constantly threatening chokes and joint locks. In our database, wrestlers own a 56% win rate against submission artists, giving Lindland the stylistic edge.
A few statistical edges stand out. Lindland throws significantly more leather — a 1.5 sig. strike per minute gap. Lindland is far more active with takedowns, averaging 2.0 more per 15 minutes. Miletich has tighter striking defense, making opponents miss more often.
The Pick: Matt Lindland over Pat Miletich. The model gives Lindland a slight nod at 62% — this could easily go either way.
Evan Tanner vs Elvis Sinosic
The Light Heavyweight matchup features Evan Tanner (11-5) taking on Elvis Sinosic (1-5). Sinosic is the bigger frame at 6'3" with a 3-inch reach advantage.
Tanner is rated at 1034 — 214 points above Sinosic's 820. Gaps this large usually mean one fighter has been consistently beating better opponents.
The style clash matters here: Tanner is most dangerous on the ground, constantly threatening chokes and joint locks, while Sinosic looks to close distance and put the fight on the mat. In our database, wrestlers own a 56% win rate against submission artists, giving Sinosic the stylistic edge.
A few statistical edges stand out. Tanner throws significantly more leather — a 2.9 sig. strike per minute gap. Tanner is far more active with takedowns, averaging 2.0 more per 15 minutes. Tanner has tighter striking defense, making opponents miss more often.
The Pick: Evan Tanner over Elvis Sinosic. The model is firm on this one: Tanner at 76%.
Frank Mir vs Pete Williams
The Heavyweight matchup features Frank Mir (16-10) taking on Pete Williams (3-4).
Mir is rated at 1252 — 399 points above Williams's 853. Gaps this large usually mean one fighter has been consistently beating better opponents.
The style clash matters here: Mir is most dangerous on the ground, constantly threatening chokes and joint locks, while Williams looks to close distance and put the fight on the mat. In our database, wrestlers own a 56% win rate against submission artists, giving Williams the stylistic edge.
A few statistical edges stand out. Mir throws significantly more leather — a 2.0 sig. strike per minute gap. Williams is far more active with takedowns, averaging 0.6 more per 15 minutes. Williams has tighter striking defense, making opponents miss more often.
The Pick: Frank Mir over Pete Williams. The model gives Mir a slight nod at 63% — this could easily go either way.
Matt Serra vs Kelly Dullanty
The Lightweight matchup features Matt Serra (7-6) taking on Kelly Dullanty (0-0).
Serra is rated at 1201 — 303 points above Dullanty's 898. Gaps this large usually mean one fighter has been consistently beating better opponents.
A few statistical edges stand out. Serra throws significantly more leather — a 0.6 sig. strike per minute gap. Serra is far more active with takedowns, averaging 2.5 more per 15 minutes. Dullanty has tighter striking defense, making opponents miss more often.
The Pick: Matt Serra over Kelly Dullanty. We're leaning Serra here at 66%, a solid but not overwhelming edge.
Sean Sherk vs Jutaro Nakao
The Welterweight matchup features Sean Sherk (7-4) taking on Jutaro Nakao (1-0).
Sherk is rated at 1371 — 297 points above Nakao's 1074. Gaps this large usually mean one fighter has been consistently beating better opponents.
A few statistical edges stand out. Nakao throws significantly more leather — a 0.3 sig. strike per minute gap. Sherk is far more active with takedowns, averaging 6.1 more per 15 minutes. Sherk has tighter striking defense, making opponents miss more often.
The Pick: Sean Sherk over Jutaro Nakao. This is essentially a pick'em. The model nudges toward Sherk at 55%, but there's almost nothing separating these two.
Methodology
Predictions are generated by our ensemble model combining LightGBM (65%) and CatBoost (35%), trained on every UFC fight since 1994. The model uses 23 features including Elo ratings, rolling 5-fight statistical averages, style matchup history, physical attributes, and market odds when available.
On our held-out test set (402 fights from January-September 2023), the model achieves 63.4% accuracy with a log-loss of 0.626. High-confidence picks (>75% probability) hit at 82.7%. For full model transparency, visit our Model page.