Kevin Holland, No Such Thing as "Liking" a Horse, March Madness, and BMF Champion Usman
Kevin Holland, No Such Thing as "Liking" a Horse, March Madness, and BMF Champion Usman
Horses, BMFs and Baylor Bears (oh my!)
I’m writing this in the evening, and the sun is out. Daylight Savings Time rocks. It also means I’m in a good mood which means I’m going to ramble more than usual, so let’s keep the intro short.
By the way I’m also starting work on a new book.
Here’s this week’s newsletter!
Holland vs Brunson:
Kevin Holland has 3 UFC fights in which he faced 5+ takedowns, and 7 in which he did not.
When his opponents don’t attempt 5+ takedowns, he is 6-1 with a +166 significant strike differential.
When his opponents do, he is 2-1 (with both of those wins being by split decision) with a -4 significant strike differential, spending 51% of that fight time with his opponents in control positions.
Derek Brunson has attempted 7+ takedowns in four straight fights.
It’s obviously not as simple as “Brunson will attempt 5 takedowns and automatically win,” but the fact that those split decisions were against lesser competition (Darren Stewart and Gerald Meerschaert) is not a good sign for how Holland will fare if Brunson commits to his wrestling again.
Replace “Horse” With “Fighter”:
"The issue is not which horse in the race is the most likely winner, but which horse or horses are offering odds that exceed their actual chances of victory. This may sound elementary, and many players may think they are following this principle, but few actually do. Under this mindset, everything but the odds fades from view. There is no such thing as “liking” a horse to win a race, only an attractive discrepancy between his chances and his price."
- Steven Crist
This is why it should immediately give you pause when someone says they will “never again” bet on any given fighter.
But this is also important to reinforce to yourself often. Even with this concept at the core of my philosophy when I’m handicapping, I find myself falling into the trap too — analyzing a fight, deciding who I like in the fight, then looking at the odds and shaping my opinion around the odds.
An easy way to avoid this pitfall is to completely avoid the odds and consciously try to avoid “liking” or “not liking” a fighter when you’re first analyzing a fight.
Analyze the fight, assign a range of win probabilities for each fighter, and then check which side (if either) your win probabilities suggest the odds are offering value on. You should be entirely unemotional at this stage — who cares how badly Bethe Correia burned you in her last fight, if you have her at 25-30% to win and she’s a +400 underdog, that’s a great bet, even if you don’t think it’s on a great fighter.
Kamaru Usman vs Jorge Masvidal II …
is one of the most non-sensical title fights of all time, from a “sport” perspective. I get the business behind it, but in terms of the actual sport/rankings/merit side it’s a real head-scratcher.
On the plus side, maybe if he wins again people will start giving Usman his proper status as the Welterweight/BMF double-champ. Surely they’ll fly The Rock out to Jacksonville for this one, right?
Winning March Madness Brackets With No College Basketball Knowledge
March Madness bracket office pools might have the biggest edge of any common betting opportunity —and I say this as someone who can't name more than 2-3 college basketball players before March Madness each year.
Winning a March Madness pool is not about being good at picking college basketball winners.
You're obviously not here for my college basketball takes, but the edge in March Madness brackets is entirely betting theory related. People don't understand game theory.
The absolute easiest edge you're going to find is figuring out which No. 1 seeds *not* to pick to win the tournament.
To do this, you don't need to know a single thing about any of the four teams. Most sites have "public pick" percentages listed, which will give you a decent idea of how popular a team is as a pick to win in any given round.
This year, the clear popular pick is Gonzaga, at 39% (on CBS). That's compared to 17% for Illinois, 9% for Baylor and 8% for Michigan.
Now we can look at projection sources. "Good enough" is all we need here, and I like to use FiveThirtyEight, but your mileage may very. The process is the same regardless.
As has been the case every year since I've started paying attention, the public is over-confident on the favorite. FiveThirtyEight gives Gonzaga just a 27% chance to win. Even Illinois (15%) is projected less likely to win than they're picked, while Baylor (13%) is more likely than are being picked.
Because of the disproportionate way pools pay out (most/all the prize money for first place, while finishing 20th out of 30 and 29th out of 30 give you equal pay) and disproportionate way games are scored (more points for a championship win than a Final Four win), simply fading that over-confidence puts you in a highly favorable spot.
That's thanks to our old friend "conditional probability." Let's say there are 100 other brackets in your pool (101 entries total, including yours).
If you have Gonzaga as the winner and they do win, you're competing with 39 other brackets.
If you have Baylor as the winner and they do win, you're competing with only 9 other brackets.
Assuming you and everyone else who has the same winner have a roughly similar chance of finishing 1st when your champion wins:
27% of the time Gonzaga wins, and 2.5% (1/40) of that 27% of the time you win it all — that's 0.68% overall (0.025 * 0.27)
13% of the time Baylor wins, 10% (1/10) of that 13% sees you win — that's 1.3% overall.
Your baseline expectation before the contest (1/101) is to win 0.99% of the time. All other things equal, picking Gonzaga hurts your chances while picking Baylor helps your chances.
This edge is even bigger in years with a more popular favorite. In 2019 Duke was picked at something like a 40-50% rate, which obviously worked out very well for Duke faders.
And as far as the "all other things equal" qualifier goes? Just pick the betting/projection favorites. If anyone in your office could actually be a long-term profitable college basketball better (unlikely), they're still only going to have a couple percentage point edge — which would work out to maybe 5 games across the whole tournament.
If you're just blindly following projections you can expect to be as accurate as anyone else in your pool, and by combining that with some sound game theory you can leverage yourself very easily into a profitable spot.
42% of Kamaru Usman's career fights have ended in KO/TKO wins …
33% of Jorge Masvidal's career fights have ended in KO/TKO wins.
Giveaway Winner!
Congratulations to Twitter user @Dr_Beans17 for winning this week’s giveaway!
