FightIQ's model flagged six of the 11 UFC 330 bouts it modeled, including four at LOCK, the highest tier the model issues.
The card runs Saturday at Xfinity Mobile Arena in Philadelphia and lists 12 fights, including two title fights. FightIQ has a model read on 11 of them.
The exception is Charles Johnson vs. Eduardo Chapolin, a catchweight bout. It carries no model read, and FightIQ is not publishing a number for it.
Four LOCKs on one card is the most the model has ever produced. An internal check found the same count has occurred six times before, so the card sits at the top of the model's range rather than beyond it.
It is unusual. It is not unprecedented, and it is not a glitch. The count was checked before publication, because an unusual number of top-tier picks is the kind of output that should be questioned.
FightIQ flags two tiers. A pick is LOCK or HIGH, or it is not flagged. Everything else is published with its probability and left there.
The four LOCKs:
Mansur Abdul-Malik, 80.4% over Dustin Stoltzfus, most likely by KO/TKO. It is the highest number on the card.
Donte Johnson, 77.2% over Eric McConico, by decision.
Myktybek Orolbai, 76.6% over Jeremiah Wells, by decision.
Esteban Ribovics, 73.5% over Edson Barboza, by decision.
Two fights are flagged HIGH. Joel Alvarez sits at 76.8% over Chidi Njokuani, by KO/TKO. Islam Makhachev sits at 68.3% over Ian Machado Garry in the five-round main event, by decision.
All four LOCKs are on the undercard. The flagged main event carries the lowest of the six flagged numbers. The model's conviction has again landed away from the biggest fights of the night.
The two title fights sit at opposite ends of the board. One is flagged, at the lowest flagged number on the card. The other is not flagged, and lands closer to even than all but one of the modeled fights.
Method reads follow the same split. Four of the six flagged picks project a decision, and two project a KO/TKO.
Across a validated cohort of 80 flagged LOCK picks, the LOCK label has hit 91.2%, by FightIQ's own measurement. That describes past results on that cohort. It is not a forecast for Saturday.
Five of the modeled fights are not flagged.
Vicente Luque is at 66.5% over Tresean Gore, by KO/TKO. Lucas Fernando is at 58.6% over Rafael Tobias, by KO/TKO. Neil Magny is at 57.4% over Ramiz Brahimaj, by submission.
The second title fight sits here. Mackenzie Dern is at 53.5% over Gillian Robertson across five rounds, close enough to even that the model barely separates them.
Kaue Fernandes is at 52.1% over Jalin Turner, by KO/TKO, the other near coin flip on the card.
One read carries a caveat. Lucas Fernando has no prior UFC fight in FightIQ's dataset, so his 58.6% leans on the betting market rather than on fight history.
All 11 modeled fights have market odds attached. FightIQ publishes probabilities and tiers and makes no recommendation about wagering.