March Madness Predictions!

Hello guys!

So recently, a lot of my friends have been asking me to predict the result of this year's March Madness. To be honest, I don't really follow college basketball ( you know, there are just so many teams and players to look at!) but I figure why not give it a shot.

Using the same method I did in December to predict this year's NBA playoff ( a technique originally introduced by The Basketball Distribution), I arrive at the following results.

( Note. if you want details on how to do these predictions yourself, feel free to view my NBA Playoffs Predictions blog from December where I explained my procedure step-by-step, or you could also check out original Basketball Distribution article I linked above)

Elite Eight

At the time of writing, two games in the Elite Eights have already taken place, with Texas Tech upsetting #1 seed UNC (75-69), and Virginia making a wild comeback against Purdue (80-75)! Thus, we would only need to predict the match-ups between Duke vs Michigan State and Auburn vs Kentucky

For simplicity, we would pick the team with the higher predicted score to be the winner!

Since Duke and Michigan State have very similar Offensive and Defensive Rating, their predicted scores are actually extremely close ( being 78-78 if one were to round up). Assuming both teams will perform at their usual levels, we can expect this match-up to be highly intense, with both teams trading buckets back and forth! This game might even go into overtime! However, since we are sticking to our simple picking rule, we would pick Duke to advance.

On the other hand, even though Auburn does have a slightly higher Offensive Efficiency relative to Kentucky, they severely trails behind in the defensive end. This is supported by their respective Defensive Efficiency, with Auburn allowing up to 96.9 opponent points per 100 possessions, while Kentucky only allowing 90.4. This gap in the teams' defensive capabilities would, thus, be the key factor that would spin things in Kentucky's favor!

Final Four

With 4 teams left in the mix, the NCAA bracket would dictate the following match-ups:

1. Duke vs Texas Tech

2. Kentucky vs Virginia

Using the same process we did earlier, we would arrive at the following results.

Again, our method predicts that Duke is going to have another rough game against Texas Tech. Although Duke relatively excels in the offensive side, Texas Tech are better defensively! However, I would predict that Duke will barely survive this match-up, assuming both teams play at their usual level.

On the hand, Virginia should expect to have a good win over Kentucky, since it has an edge both offensively and defensively.

Championship

I predict this year's NCAA Championship game to be between Duke and Virginia. Both are extremely talented teams, but in the end, there could only be one winner. Applying the same method, we would arrive at the following result.

Assuming both teams perform at their expected levels ( no choking, injuries, or any major surprise factor), our predictive technique indicates that Virginia will become this year NCAA Champion! Statistically, Virginia outshines Duke in both offensive and defensive ends. Thus on a pure mathematics standpoint, we should expect Virginia to be the winner.

However, studying the past 2 regular season match-ups between the two schools this, Duke has won both games! Thus, using this historical pattern as a predictor, one should expect Duke to win!

With these two conflicting predictions, which approach should we trust more? Current statistics or historical trend?

To be honest, I don't have a good answer. However, we do know for sure that this upcoming Championship game is going to be very exciting! As of now, I will stick to what the numbers says, and go with Virginia as this year NCAA's champions!