Friday, April 19, 2013

This Will Be Wrong

It's playoff time again, and since mine are the only opinions on the internet, I decided to roll out some predictions

Most of these are based on a model I built for a statistics class in college that I have tweaked since then. It's not necessarily the most complex thing in the world, but I still won't get too much into detail. Basically it uses some regressions, t-distributions, and probabilities in order to determine which team is better and how long a series is likely to last. That being said, on to the predictions.

East First Round
1. Miami v. 8. Milwaukee- Miami in 4
This is pretty easy. Milwaukee was the worst team to qualify for the playoffs and ranked 18th overall.

4. Brooklyn v. 5. Chicago- Chicago in 7
This was the hardest matchup. The model only gives Chicago a 51-49 edge in the series, so this one could easily go the Nets way. However, Brooklyn plays very slow and that only helps out a stingy Bulls' defense.

3. Indiana v. 6. Atlanta- Indiana in 5
It's a shame Atlanta couldn't have held on to the 5 seed, because they would have been favored to beat either Chicago or Brooklyn. Alas, they will run into a potentially underrated Pacers team.

2. New York v. 7. Boston- New York in 6
I know it seems volatile with the 3-point shooting and a tough Boston opponent, but I think the Knicks will pull it out.

West First Round
1. Oklahoma City v. 8. Houston- Oklahoma City in 5
The model saw this as a sweep, but I think Houston will be able to steal a game. Even 5 games feels too short with the quality of this Rockets team.

4. Los Angeles Clippers v. 5. Memphis- Los Angeles Clippers in 7
This will be an awesome rematch of last year. I was a little surprised that the Grizzlies came out as slightly overrated despite heavy factoring of certain defensive stats.

3. Denver v. 6. Golden State- Denver in 5
I watched a lot of Nuggets games and so I'm a little biased. The model had them slightly overrated, which I'm fine with due to the injuries (which weren't factored in). Still, the numbers had this as a sweep which I think was too optimistic. I still think Denver is able to put it away quickly.

2. San Antonio v. 7. Los Angeles Lakers- San Antonio in 4
Does this make me a hater? Maybe a sweep is too much, but there are bound to be a couple in the early rounds and this one was one of the more likely four-gamers.

East Semifinals
1. Miami v. 5. Chicago- Miami in 5
This was predicted as a sweep, but I cannot see Chicago going down in 4 games, especially not after how hard they played the Heat during the regular season.

2. New York v. 3. Indiana- Indiana in 6
The Pacers are just too good at taking away what teams do well and if the 3-point line is choked off, I'm not sure the Knicks can score enough.

West Semifinals
1. Oklahoma City v. 4. Los Angeles Clippers- Oklahoma City in 6
Honestly, 6 games feels like one too many.

2. San Antonio v. 3. Denver- San Antonio in 7
This will be a very fun series to watch with two incredibly potent offenses. It kills me to see Denver's run end here because I am a believer, but the Spurs are disciplined defensively and won't get rattled by Denver's frantic pace.

East Finals
1. Miami v. 3. Indiana- Miami in 6
The model gives Miami a 32% chance of not losing a game until the NBA Finals. Again, I think that's too optimistic and I think the Pacers are bound to take at least one game. Alas, Miami was far and away the best team in the East and will look to defend its title.

West Finals
1. Oklahoma City v. 2. San Antonio- Oklahoma City in 6
They figured the Spurs out last year and were ranked as the best team in league based on my numbers. Should be another entertaining series, but ultimately frustrating for the Spurs.

NBA Finals
1. Oklahoma City v. 1. Miami- Oklahoma City in 6
Ok, so the purpose of the model isn't to be overly clever and these were the two best teams this year. A rematch may not be exciting based on variety, but I think we will see some really good basketball between the league's two best players.

Potential Upsets
As a bonus I'll leave with two potential upset candidates for series. The first is Boston over New York in the first round. The model saw the Celtics as severely underrated and if the 3's aren't falling this could end up being another disappointing season for the Knicks. The other potential upset is Denver over San Antonio in the West Semifinals. The model says San Antonio is favored to win 64% of the time, one of the lower figures on the board.

No comments:

Post a Comment