I wrote a computer program called Zoltar that predicts the results of NFL football games. Unlike many prediction programs that use classical statistics techniques, Zoltar uses an information-theoretic approach.
Zoltar’s point spread predictions for week 16 of the 2015 NFL season are:
Zoltar favorite = raiders by 2 underdog = chargers Zoltar favorite = eagles by 6 underdog = redskins Zoltar favorite = panthers by 5 underdog = falcons Zoltar favorite = bills by 4 underdog = cowboys Zoltar favorite = lions by 6 underdog = fortyniners Zoltar favorite = chiefs by 15 underdog = browns Zoltar favorite = colts by 0 underdog = dolphins Zoltar favorite = patriots by 5 underdog = jets Zoltar favorite = texans by 5 underdog = titans Zoltar favorite = steelers by 4 underdog = ravens Zoltar favorite = buccaneers by 6 underdog = bears Zoltar favorite = saints by 6 underdog = jaguars Zoltar favorite = cardinals by 6 underdog = packers Zoltar favorite = seahawks by 10 underdog = rams Zoltar favorite = vikings by 6 underdog = giants Zoltar favorite = broncos by 2 underdog = bengals
Notice that Zoltar predicts a tie for the Colts vs. Dolphins game. When this happens, just in order to calculate a head-to-head winning percentage, Zoltar predicts the team with the higher internal rating (Zoltar computes ratings for each team) will win.
Zoltar’s predictions are very close to the Las Vegas point spreads, but there are 4 games where Zoltar sees a theoretical betting opportunity:
* Zoltar says the Raiders will beat the Chargers by only 2 points but Vegas says the Raiders will win by 5.5 points so Zoltar would recommend betting on the Vegas underdog Chargers.
* Zoltar says the Texans will beat the Titans by 5 points but Vegas says the Titans will win by 4.5 points so Zoltar would recommend betting on the Vegas underdog Texans. Both team have been hit hard by player injuries so Zoltar think Vegas is overreacting
* Zoltar says the Steelers will win by only 4 points over the Ravens but Vegas say the Steelers will win by 10.5 points so Zoltar would recommend betting on the underdog Ravens. (I saw both teams play last week and The Steelers looked unbeatable and the Ravens looked horrible so Vegas is using the “eye test”.)
* Zoltar says the Seahawks will beat the Rams by just 10 points but Vegas says the Seahawks will crush the Rams by 13.5 points so Zoltar would (barely) recommend betting on the Vegas underdog Rams.
For the season, through week 15, in head-to-head predictions, meaning just predicting the winner of a game (which isn’t particularly useful), Zoltar has gone 164-60 for 73% accuracy. Bing Predictions is 142-82 for 63% accuracy. By the way, if you simply picked the home team to win, you’d only get about 55% accuracy.
Update: Tuesday, 12/29/2015: It was good news and bad news for Zoltar in week 16. On the positive side, Zoltar was a perfect 4-0 against the Vegas point spread. Zoltar picked 4 underdogs (Chargers, Texans, Ravens, Rams). The Raiders didn’t cover the 5.5 spread against the Chargers, and the Texans, Ravens and Rams all won outright. On the downside, in just predicting the winners of games (which isn’t useful except for parlay bets), Zoltar was a weak 10-6 (62.5% correct). Bing Predictions also went 10-6 for the week. In fact, Zoltar and Bing had the exact same winner predictions for all 16 games, something that has never happened before.