David Williams · @dwilliamsmma
23rd Sep 2016 from TwitLonger
UFC Fight Night: Cyborg vs. Lansberg - Statistical Model Estimates
This card has a lot of talent for a UFC Fight Night, and I'm excited to watch it, BUT... I don't know what happened to the matchmaking here. The UFC has been doing such a good job of matchmaking on such a consistent basis that it's a bit jarring to see fights like Thiago Santos vs. Eric Spicely on the main card.
Note that these estimates make a home-cage advantage adjustment for Brazilian fighters.
Cris Cyborg 83.2% Lina Lansberg 16.8%
Renan Barao 84.5% Phillipe Nover 15.5%
Antonio Silva 54.2% Roy Nelson 45.8%
Francisco Trinaldo 68.0% Paul Felder 32.0%
Thiago Santos 85.2% Eric Spicely 14.8%
Godofredo Pepey 51.4% Mike De La Torre 48.6%
Michel Prazeres 64.8% Gilbert Burns 35.2%
Rani Yahya 62.9% Michinori Tanaka 37.1%
Jussier Formiga 70.2% Dustin Ortiz 29.8%
Luan Chagas 75.6% Erick Silva 24.4%
Steven Ray 63.6% Alan Patrick 36.4%
Vicente Luque 74.5% Hector Urbina 25.5%
Glaico Franca 67.5% Gregor Gillespie 32.5%
The model always likes Roy Nelson and Erick Silva a lot less than the betting public, and this event is no exception. Nelson would be the model's pick over Antonio Silva in a neutral setting, and the model NEVER likes Nelson to beat anybody. Naturally, home-cage advantage makes Bigfoot the favorite anyway. I'll be honest - with Bigfoot losing by knockout routinely in recent fights, I'll be surprised if he survives the first round against Nelson.
The other things to note are +160 underdog Michel Prazeres being the model's pick to beat Gilbert Burns, and Cris Cyborg having "just" an 83 percent chance of winning in the main event. Regarding Cyborg, I think the model just doesn't have enough data to properly reflect how dominant she is in the cage. The only statistics I include in the model are statistics from UFC fights, so right now the model only has Cyborg's win over Leslie Smith to work with.