Last week, the UNU Football met on Thursday ( as we do every week ) to take a stab at predicting the outcomes of all 15 NFL games. Of course, when you’re harnessing a weapon as power as Swarm Intelligence, it’s not enough just to pick winners, so we’ve challenged the UNU Football group to pick every single game Against the Spread.
A quick primer: if you’re unfamiliar with gambling jargon, the Spread is essentially a betting handicap given to the team that’s supposed to win. For example, on Thursday the Cardinals are favored by 3.5 points over the 49ers. That means that the Swarm can’t simply pick either the Cardinals or 49ers to win, it needs to decide if the Cards will win by more than 3.5 points. If the 49ers win, or even if they lose by less than 3.5 points, the 49ers would be the winner Against the Spread.
As you can imagine, the actual number of the Spread is crucial. Experts in Las Vegas set the spread so that, in their ideal world, 50% of the money is wagered on the Favorite and 50% on the Underdog, so that Vegas doesn’t really care who wins the game. In this ( very simplified ) way, sports betting is often thought of as a 50-50 proposition, a coin flip. And, in Week 4, that’s exactly how the Swarm did. If you take a look at the Swarm’s picks, you’ll see a perfectly average 6-6 record. The 4 games picked with High Confidence? They went 2-2. The two times the Swarm picked the Underdog? They went 1-1. Like I said, Even Steven.
Here, by the way, is the Swarm’s best pick, the Broncos -3 over Tampa Bay. As I mentioned last week, this was the Swarm’s highest confidence pick ( rated as High Confidence and with High Brainpower ), and the Broncos delivered. In a game they were favored to win by 3 points, they won by 20, easily covering the spread. Nice work, Denver…and UNU Football.
But let’s back up a second. I wrote earlier that the Swarm had only picked 2 Underdogs out of the 12 predictions made last week. Now, I personally don’t know enough about handicapping to say which ( if any ) of those games were bad picks, but that’s sort of like predicting that a coin would land on heads 10 out of 12 times. And, we’ve seen that Favorite bias a few times in UNU Football, which is using a slightly different methodology than we used last year. ( If you remember, we had a fairly remarkable run last year in picking NCAA and NFL Playoff games). So, it’s worth wondering why the Swarm is leaning so heavily into the favorites.
In short, I believe there’s something inherent to the current format that is leading the Swarm to pick favorites nearly ever time. And, because I believe ( that is, I’ve seen time and again ) that the Swarm can do far better than a coin flip, we’re going to be tweaking the format for UNU Football slightly. Now, instead of revealing the Spread to the Swarm, we’ll be asking the Swarm to effectively set the line itself, like this:
This will allow the Swarm a bit more time to think about the game in question, and determine for itself who should be favored and by how much. With that knowledge in hand, we’ll then ask the Swarm to make a recommendation using the official betting line in the standard format shown above in the Broncos game.
The UNU Football Week 5 Swarm will take place on Thursday. Help us show the world the power of Swarm Intelligence…and bust up Vegas in the process!
And here’s the full rundown of the Week 4 picks: