One thing that’s very important to me is to be accountable – not just to the readers of this blog, but to myself as well. It’s time to level with everybody. My “degenerate gambling” picks for the UFC and the NFL have basically been terrible for two months now. I hoped that UFC 178 would be the beginning of a turnaround but I was dead wrong.
Eddie Alvarez -110: Was simply out-matched at standing distance against Donald Cerrone. I thought Alvarez would be able to tee off on Cerrone with punches, but that only really happened during that one flurry in the first round.
Dustin Poirier +235: Basically got smoked by Conor McGregor.
Tim Kennedy +125: I could complain about Yoel Romero’s antics between rounds if I wanted to, but I miscalculated on this fight. I thought Kennedy could control Romero with takedowns, but it looks instead like Romero’s takedown issues of the past are behind him.
Patrick Cote +305: Looked like the 3-1 underdog that he was. 0/8 in takedowns and most of his striking offense was in the form of leg kicks.
Brian Ebersole +190: Winner!
I’m frustrated but I think it’s a mistake to try to “get it all back” in a short period of time. The worst thing to do is go on “tilt” and just make a tremendous number of questionable bets. So I’m going to do the opposite – I’m going to really narrow down my betting selections and try to focus in on what I believe will be the best picks.
Part of that will be to rely on a more stats-heavy approach. I’ve gradually been straying farther and farther from my metrics, to the point where my picks have been based mostly on what I believe from a scouting perspective. That approach may work for 99% of intelligent MMA folks out there, but it’s not my intellectual strength. My strength is working with numbers, and while there are plenty of people in the betting community who cackle at the idea that numbers can be meaningful in MMA, I strongly believe it can work.
If I had strictly adhered to what FPR told me, I would have made the following decisions:
-Bet on Ebersole (+1.84 FPR) to beat John Howard (-0.16).
-Pass on a bet on Cote (+0.08) to beat Stephen Thompson (+3.27).
-Pass on a bet on Kennedy (+6.98), due to a small data sample on Yoel Romero.
-Pass on a bet on Poirier (+3.77), due to a small data sample on Conor McGregor.
-Pass on a bet on Alvarez due to a small data sample on him against Donald Cerrone (+4.30).
Obviously it’s easy to say these things in retrospect. FPR would have also indicated betting value on Chris Cariaso (-0.04) against Demetrious Johnson (+3.49), who is criminally underrated by FPR for some reason. I’m not sure if I’ll go back to using strictly FPR for betting… but it can’t be as bad as my picks have been the last couple months.
Part of the problem is that, with the UFC running so many fight cards, the UFC is featuring many fighters who simply haven’t been in the promotion for very long. Next week’s fight card in Stockholm is a great example of this. The only fighters on that card with a data sample I consider sufficient (60 minutes) are Rick Story, Max Holloway, and Dennis Siver. It becomes pretty much necessary to do scouting to determine where there is betting value.
I’ll have to think about this further. It could be that I never had a strategy that was a long-term winner, and recent events were simply regression to the mean. There are reasons that I always make sure to recommend people don’t follow my picks. For now, I’m in a process of re-evaluation and hopefully that will lead to better things ahead.