UFC 20: Battle for the Gold: Predictions & Analysis

Friday, May 7, 1999·Birmingham, Alabama, USA
Published February 27, 2026
Predictions are for entertainment purposes only and do not constitute financial advice. Please gamble responsibly.

UFC 20: Battle for the Gold lands on Friday, May 7, 1999 in Birmingham, Alabama, USA with 7 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
Bas Rutten vs Kevin RandlemanHeavyweightBas RuttenLean55%
Pedro Rizzo vs Tra TelligmanHeavyweightPedro RizzoStrong82%
Pete Williams vs Travis FultonHeavyweightPete WilliamsConfident74%
Wanderlei Silva vs Tony PetarraMiddleweightWanderlei SilvaConfident71%
Marcelo Mello vs David RobertsLightweightMarcelo MelloLean58%
Laverne Clark vs Fabiano IhaLightweightLaverne ClarkStrong81%
Ron Waterman vs Chris CondoHeavyweightRon WatermanLean61%

Fight-by-Fight Breakdown

55%
Bas Rutten
Rutten
1-0
Elo 1256
VS
Randleman
3-3
Elo 1189
Wrestler

The Heavyweight matchup features Bas Rutten (1-0) taking on Kevin Randleman (3-3).

Rutten carries a modest Elo edge (1256 to 1189), the kind of gap that reflects a slightly better run of form rather than a talent chasm.

A few statistical edges stand out. Rutten throws significantly more leather — a 2.2 sig. strike per minute gap. Randleman is far more active with takedowns, averaging 7.5 more per 15 minutes. Rutten has tighter striking defense, making opponents miss more often.

The Pick: Bas Rutten over Kevin Randleman. The model gives Rutten a slight nod at 55% — this could easily go either way.

82%
Pedro Rizzo
Rizzo
8-5
Elo 1268
Striker
VS
Telligman
1-3
Elo 848

The Heavyweight matchup features Pedro Rizzo (8-5) taking on Tra Telligman (1-3).

Rizzo is rated at 1268 — 420 points above Telligman's 848. Gaps this large usually mean one fighter has been consistently beating better opponents.

A few statistical edges stand out. Rizzo throws significantly more leather — a 1.9 sig. strike per minute gap. Telligman is far more active with takedowns, averaging 0.0 more per 15 minutes. Rizzo has tighter striking defense, making opponents miss more often.

The Pick: Pedro Rizzo over Tra Telligman. The model is firm on this one: Rizzo at 82%.

74%
Pete Williams
Williams
3-4
Elo 853
Wrestler
VS
Fulton
0-1
Elo 1043

The Heavyweight matchup features Pete Williams (3-4) taking on Travis Fulton (0-1).

Fulton is rated at 1043 — 190 points above Williams's 853. Gaps this large usually mean one fighter has been consistently beating better opponents.

A few statistical edges stand out. Williams throws significantly more leather — a 2.4 sig. strike per minute gap. Williams is far more active with takedowns, averaging 1.0 more per 15 minutes. Fulton has tighter striking defense, making opponents miss more often.

The Pick: Pete Williams over Travis Fulton. We're leaning Williams here at 74%, a solid but not overwhelming edge.

71%
Wanderlei Silva
Silva
4-7
Elo 1282
Knockout Artist
VS
Petarra
0-0
Elo 859

The Middleweight matchup features Wanderlei Silva (4-7) taking on Tony Petarra (0-0).

Silva is rated at 1282 — 423 points above Petarra's 859. Gaps this large usually mean one fighter has been consistently beating better opponents.

A few statistical edges stand out. Silva throws significantly more leather — a 1.4 sig. strike per minute gap. Petarra is far more active with takedowns, averaging 0.0 more per 15 minutes. Petarra has tighter striking defense, making opponents miss more often.

The Pick: Wanderlei Silva over Tony Petarra. We're leaning Silva here at 71%, a solid but not overwhelming edge.

58%
Marcelo Mello
Mello
0-0
Elo 1128
VS
Roberts
0-0
Elo 873

The Lightweight matchup features Marcelo Mello (0-0) taking on David Roberts (0-0).

Mello is rated at 1128 — 255 points above Roberts's 873. Gaps this large usually mean one fighter has been consistently beating better opponents.

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

The Pick: Marcelo Mello over David Roberts. The model gives Mello a slight nod at 58% — this could easily go either way.

81%
Laverne Clark
Clark
4-0
Elo 1095
VS
Iha
3-3
Elo 895
Submission Artist

The Lightweight matchup features Laverne Clark (4-0) taking on Fabiano Iha (3-3).

Clark is rated at 1095 — 200 points above Iha's 895. Gaps this large usually mean one fighter has been consistently beating better opponents. Clark rides a 4-fight win streak into this one.

A few statistical edges stand out. Clark throws significantly more leather — a 1.9 sig. strike per minute gap. Clark is far more active with takedowns, averaging 3.7 more per 15 minutes. Iha has tighter striking defense, making opponents miss more often.

The Pick: Laverne Clark over Fabiano Iha. The model is firm on this one: Clark at 81%.

61%
Ron Waterman
Waterman
1-1-1
Elo 1032
VS
Condo
0-0
Elo 873

The Heavyweight matchup features Ron Waterman (1-1-1) taking on Chris Condo (0-0).

Waterman is rated at 1032 — 160 points above Condo's 873. Gaps this large usually mean one fighter has been consistently beating better opponents.

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

The Pick: Ron Waterman over Chris Condo. The model gives Waterman a slight nod at 61% — 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.