The Ultimate Fighter: Team Hughes vs. Team Serra Finale: Predictions & Analysis

Saturday, December 8, 2007·Las Vegas, Nevada, USA

The Ultimate Fighter: Team Hughes vs. Team Serra Finale lands on Saturday, December 8, 2007 in Las Vegas, Nevada, USA with 9 bouts on the card. 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

MatchupPickConfidenceProb
Roger Huerta vs Clay GuidaLightweightRoger HuertaConfident70%
Mac Danzig vs Tommy SpeerWelterweightTommy SpeerLean61%
War Machine vs Jared RollinsWelterweightWar MachineToss-up51%
George Sotiropoulos vs Billy MilesWelterweightGeorge SotiropoulosToss-up52%
Ben Saunders vs Dan BarreraWelterweightBen SaundersToss-up52%
Troy Mandaloniz vs Richie HightowerWelterweightTroy MandalonizToss-up50%
Matt Arroyo vs John KolosciWelterweightMatt ArroyoLean57%
Roman Mitichyan vs Dorian PriceWelterweightDorian PriceToss-up55%
Jonathan Goulet vs Paul GeorgieffWelterweightJonathan GouletLean63%

Fight-by-Fight Breakdown

Roger Huerta vs Clay Guida

Lightweight
70%
Roger Huerta
Huerta
6-1
Elo 1257
Wrestler
VS
Guida
18-18
Elo 926
All-Rounder

The Lightweight matchup features Roger Huerta (6-1) taking on Clay Guida (18-18).

Huerta is rated at 1257 — 331 points above Guida's 926. Gaps this large usually mean one fighter has been consistently beating better opponents.

Stylistically this is Huerta's striker game against Guida's all-rounder approach. Huerta brings a versatile approach, while Guida 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. Huerta throws significantly more leather — a 2.0 sig. strike per minute gap. Huerta is far more active with takedowns, averaging 0.8 more per 15 minutes. Guida has tighter striking defense, making opponents miss more often.

**The Pick: Roger Huerta over Clay Guida.** We're leaning Huerta here at 70%, a solid but not overwhelming edge.

Mac Danzig vs Tommy Speer

Welterweight
61%
Tommy Speer
Danzig
5-7
Elo 920
All-Rounder
VS
Speer
0-1
Elo 780

The Welterweight matchup features Mac Danzig (5-7) taking on Tommy Speer (0-1).

There's a real Elo separation here: Danzig at 920 versus Speer at 780. That 140-point gap typically reflects a meaningful difference in recent quality of competition and results.

A few statistical edges stand out. Speer throws significantly more leather — a 0.0 sig. strike per minute gap. Speer is far more active with takedowns, averaging 0.0 more per 15 minutes. Speer has tighter striking defense, making opponents miss more often.

**The Pick: Tommy Speer over Mac Danzig.** The model gives Speer a slight nod at 61% — this could easily go either way.

War Machine vs Jared Rollins

Welterweight
51%
War Machine
Machine
1-0
Elo 978
VS
Rollins
0-0
Elo 873

The Welterweight matchup features War Machine (1-0) taking on Jared Rollins (0-0).

There's a real Elo separation here: Machine at 978 versus Rollins at 873. That 106-point gap typically reflects a meaningful difference in recent quality of competition and results.

A few statistical edges stand out. Rollins throws significantly more leather — a 0.0 sig. strike per minute gap. Rollins is far more active with takedowns, averaging 0.0 more per 15 minutes. Rollins has tighter striking defense, making opponents miss more often.

**The Pick: War Machine over Jared Rollins.** This is essentially a pick'em. The model nudges toward Machine at 51%, but there's almost nothing separating these two.

52%
George Sotiropoulos
Sotiropoulos
7-3
Elo 918
Wrestler
VS
Miles
0-0
Elo 890

The Welterweight matchup features George Sotiropoulos (7-3) taking on Billy Miles (0-0).

The Elo gap here is razor-thin — Sotiropoulos at 918, Miles at 890. On paper, this is as close to a coin flip as it gets.

A few statistical edges stand out. Miles throws significantly more leather — a 0.0 sig. strike per minute gap. Miles is far more active with takedowns, averaging 0.0 more per 15 minutes. Miles has tighter striking defense, making opponents miss more often.

**The Pick: George Sotiropoulos over Billy Miles.** This is essentially a pick'em. The model nudges toward Sotiropoulos at 52%, but there's almost nothing separating these two.

Ben Saunders vs Dan Barrera

Welterweight
52%
Ben Saunders
Saunders
9-9
Elo 812
All-Rounder
VS
Barrera
0-0
Elo 915

The Welterweight matchup features Ben Saunders (9-9) taking on Dan Barrera (0-0).

There's a real Elo separation here: Barrera at 915 versus Saunders at 812. That 103-point gap typically reflects a meaningful difference in recent quality of competition and results.

A few statistical edges stand out. Barrera throws significantly more leather — a 0.0 sig. strike per minute gap. Barrera is far more active with takedowns, averaging 0.0 more per 15 minutes. Barrera has tighter striking defense, making opponents miss more often.

**The Pick: Ben Saunders over Dan Barrera.** This is essentially a pick'em. The model nudges toward Saunders at 52%, but there's almost nothing separating these two.

50%
Troy Mandaloniz
Mandaloniz
1-0
Elo 1025
VS
Hightower
0-0
Elo 873

The Welterweight matchup features Troy Mandaloniz (1-0) taking on Richie Hightower (0-0).

Mandaloniz is rated at 1025 — 152 points above Hightower's 873. Gaps this large usually mean one fighter has been consistently beating better opponents.

A few statistical edges stand out. Hightower throws significantly more leather — a 0.0 sig. strike per minute gap. Hightower is far more active with takedowns, averaging 0.0 more per 15 minutes. Hightower has tighter striking defense, making opponents miss more often.

**The Pick: Troy Mandaloniz over Richie Hightower.** This is essentially a pick'em. The model nudges toward Mandaloniz at 50%, but there's almost nothing separating these two.

Matt Arroyo vs John Kolosci

Welterweight
57%
Matt Arroyo
Arroyo
1-1
Elo 887
VS
Kolosci
0-0
Elo 890

The Welterweight matchup features Matt Arroyo (1-1) taking on John Kolosci (0-0).

The Elo gap here is razor-thin — Arroyo at 887, Kolosci at 890. On paper, this is as close to a coin flip as it gets.

A few statistical edges stand out. Kolosci throws significantly more leather — a 0.0 sig. strike per minute gap. Kolosci is far more active with takedowns, averaging 0.0 more per 15 minutes. Kolosci has tighter striking defense, making opponents miss more often.

**The Pick: Matt Arroyo over John Kolosci.** The model gives Arroyo a slight nod at 57% — this could easily go either way.

55%
Dorian Price
Mitichyan
1-0
Elo 983
VS
Price
0-0
Elo 890

The Welterweight matchup features Roman Mitichyan (1-0) taking on Dorian Price (0-0).

There's a real Elo separation here: Mitichyan at 983 versus Price at 890. That 94-point gap typically reflects a meaningful difference in recent quality of competition and results.

A few statistical edges stand out. Price throws significantly more leather — a 0.0 sig. strike per minute gap. Price is far more active with takedowns, averaging 0.0 more per 15 minutes. Price has tighter striking defense, making opponents miss more often.

**The Pick: Dorian Price over Roman Mitichyan.** This is essentially a pick'em. The model nudges toward Price at 55%, but there's almost nothing separating these two.

63%
Jonathan Goulet
Goulet
4-4
Elo 997
Submission Artist
VS
Georgieff
0-0
Elo 860

The Welterweight matchup features Jonathan Goulet (4-4) taking on Paul Georgieff (0-0).

There's a real Elo separation here: Goulet at 997 versus Georgieff at 860. That 137-point gap typically reflects a meaningful difference in recent quality of competition and results.

A few statistical edges stand out. Goulet throws significantly more leather — a 1.0 sig. strike per minute gap. Goulet is far more active with takedowns, averaging 3.3 more per 15 minutes. Georgieff has tighter striking defense, making opponents miss more often.

**The Pick: Jonathan Goulet over Paul Georgieff.** The model gives Goulet a slight nod at 63% — this could easily go either way.

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.