Total wagers placed were ~$1,095. The only change post the pre-fight summary was an additional $120 on Gall at -400. I ended up losing $517 on the night, ~47% of bets placed and roughly ~7% of pre-card bankroll. Only two bets to payout were Bethe and Gall. Just one underdog on the entire card won. It’s worth noting, every card for the past six a simple underdog strategy has lost money, on average ~44% per card & ~262% in aggregate. It’s been tough, with my own bankroll (which I count based on net cumulative winnings) shrinking ~$7k or approximately in half over the same period… times like these force a guy to go back to the drawing board and rethink some core assumptions.
For a bit of background, I bet almost exclusively on underdogs. I do this because that’s what’s made the most money historically – I’m not interested in being right & high accuracy alone (like so many betting sites trying to sell you subscriptions are), I’m interested in being right relative to the odds & managing risk accordingly. This is how you make money over time. UFC underdog strategy on my math shows a ~2.6% edge over the past ~850 UFC fights, while more selective UFC underdog strategies have generated well over ~5% edge the past couple years betting over half of the total fights available. I believe this has been the case because consensus perception of how a bout should play out is often wrong – fighters are often inconsistent for unpredictable reasons, the margin between winning & losing is often much tighter than the odds imply, lines can reflect behavioural biases more than they reflect fundamental dynamics, judges & refs are inconsistent & biased, perception of how a fight should go is often rooted from biased & uneducated sources and the list goes on. This is all great and it’s cool betting on underdogs has worked in the past, but this recent losing streak is over a large sample and there’s some fundamental shifts taking place in concert which can’t be ignored… leads me to think betting on the underdog will not work as well in the future as it has in the past.
Some fundamental things are changing MMA betting making lines more efficiently priced & ultimately making it more difficult for an underdog strategy to be profitable – there are more educated eyes & dollars in the game today. Whether it’s UFC Fight Pass, Dan Hardy, Jack Slack or the countless other opinions available on potential fight outcomes via podcast or websites, it’s clearly much easier & quicker to understand key components of a match-up now than it has ever been. I also have reason to believe professional gamblers in general have turned more attention to MMA post some high-profile underdog wins (e.g. Diaz / McGregor 1, Holmes / Rousey), so the nature of the average bettor has also become more sophisticated. More bettors, smarter bettors and all with greater resources…
These dynamics can be seen in the prices. For example, at UFC 183 January 2015 the trailing 50-fight average underdog odds were over +290, while trailing 50-fight odds since Ultimate Fighter 23 July 2016 has been around +190, showing consistent trend for lower potential payouts over time. The UFC is probably doing a better job of making match-ups, but I also think bettors have gotten better at pricing fights. More concerning, today’s trailing 50-fight average underdog winning percentage & edge is at the lowest point it’s ever been since my database goes back (which is November 2014).
So what does this mean for the strategy and what do I need to change to adjust for this reality?
I still think mispricing will exist, I just think it will be less common and potentially shorter lived than in the past. This means I’m going to need to be more selective – more than ever I need to stick to where I have good reason to believe I have a competitive advantage, understand the risk, understand the other side of the bet & why they’re likely wrong and see a clear margin of safety to the Kelly Criterion. The second thing that comes to mind is I’ll potentially need to be more open to different betting return profiles. For example, I have a representative sample that shows I’ve picked the winner ~68% of the time (on over half the fights per card over the period, so I’m not cherry picking; also, this has shown an edge to the odds, which is ultimately what matters), which opens some doors to new strategies potentially…
This new landscape is going to be tough, and it may not be worth my time to continue… I love MMA & watching fights, but the opportunity cost to doing this at this level is real… I’ll give it to Christmas, then decide whether it’s worth continuing.