2017 NCAA Tournament Round of 64 Upset Predictions

The Cabri Group / CAN Machine Learning Lower Seed Win Prediction tool has made its first round forecast! Without further ado:

East Tennessee St. (13) over Florida (4)
Xavier (11) over Maryland (6)
Vermont (13) over Purdue (4)
Florida Gulf Coast (14) over Florida St. (3)
Nevada (12) over Iowa St. (5)
Rhode Island (11) over Creighton (6)
Wichita St. (10) over Dayton (7)

 
* If the last play in games add another predicted upset, we’ll update that prior to the game starting.

Update: USC (11) over SMU (6)

One of the obvious observations on the predictions is: “Wait, no 8/9 upsets????” Remember these games show the most similar characteristics of the largest historic collection of upsets. This doesn’t mean that there will be no upsets as 8/9 nor that all of the predictions above will hit (remember we are going for 47% upsets) nor that all games not listed will have the favorites win. The games on the list are there because they share the most characteristics with historic times when the lower seed won.
Also, one of the key team members on this project, Matt, is a big Creighton fan (and grad). He was not happy to see Creighton on the list. I’ll speak to that one specifically. In the technical notes, I indicated that one of the many criteria that is being used is was Defensive Efficiency (DE). Machine Learning algorithm (Evolutionary Analysis) doesn’t like it when the lower seed has a large gap of DE between the lower seed and the higher seed. Creighton actually has a lower Defensive Efficiency than Rhode Island. Sorry Matt. Again, it doesn’t mean Creighton won’t win, it only means that the Rhode Island v. Creighton game shares more criteria with a the largest collection of historic upsets than the other games in the tournament.
As we indicated, we will use the odds as well as a count of upsets to determine how well we do as the tournament goes on. We’ll have a new set of predictions on Saturday for the next round of the tournament and a recap coming on Monday.
For more information about how we created the Machine Learning algorithm and how we are keeping score, you may read the Machine Learning article here:
http://can2013.wpengine.com/machine-learning-basketball-methodology

machine learning prediction

March Machine Learning Mayhem

Machine Learning and the NCAA Men’s Basketball Tournament Methodology

 <<This article is meant to be the technical document following the above article. Please read the following article before continuing.>>

“The past may not be the best predictor of the future, but it is really the only tool we have”

 
Before we delve into the “how” of the methodology, it is important to understand “what” we were going for: A set of characteristics that would indicate that a lower seed would win. We use machine learning to look through a large collection of characteristics and it finds a result set of characteristics that maximizes the number of lower seed wins while simultaneously minimizing lower seed losses. We then apply the result set as a filter to new games. The new games that make it through the filter are predicted as more likely to have the lower seed win. What we have achieved is a set of criteria that are most predictive of a lower seed winning.
 
This result set is fundamentally different than an approach trying to determine the results of all new games whereby an attempt is made to find result set that would apply to all new games. There is a level of complexity and ambiguity with a universal model which is another discussion entirely. By focusing in on one result set (lower seed win) we can get a result that is more predictive than attempting to predict all games.
 
This type of predictive result set has great applications in business. What is the combination of characteristics that best predict a repeat customer? What is the combination of characteristics that best predict a more profitable customer? What is the combination of characteristics that best predict an on time delivery? This is different from just trying to forecast a demand by using a demand signal combined with additional data to help forecast. Think of it as the difference between a stock picker that picks stocks most likely to rise vs. forecasting how far up or down a specific stock will go. The former is key for choosing stocks the later for rating stocks you already own.
 
One of the reasons we chose “lower seed wins” is that there is an opportunity in almost all games played in the NCAA tournament for there to be a data point. There are several games where identical seeds play. Most notably, the first four games do involve identical seeds and the final four can possibly have identical seeds. However, that still gives us roughly 60 or so games a year. The more data we have, the better predictions we get.
 
The second needed item is more characteristics. For our lower seed win we had >200 different characteristics for years 2012-2015. We used the difference between the characteristics of the two teams as the selection. We could have used the absolute characteristics for both teams as well. As the analysis is executed, if a characteristic is un-needed it is ignored. What the ML creates is a combination of characteristics. We call our tool, “Evolutionary Analysis”. It works by adjusting the combinations in an ever improving manner to get result. There is a little more in the logic that allows for other aspects of optimization, but the core of Evolutionary Analysis is finding a result set.
The result set was then used as a filter on 2016 to confirm that the result is predictive. It is possible that the result set from 2012-2015 doesn’t actually predict 2016 results. Our current result set as a filter on 2016 data had 47% underdog wins vs. the overall population. The historic average is 26% lower seed wins and randomly, the 47% underdog win result could happen about 3.4% of the time. Our current result is therefore highly probable as a predictive filter.
 
The last step in the process is to look at those filter criteria that have been chosen and to check to see if they are believable. For example, one of the criteria that was Defensive Efficiency Rank. Evolutionary Analysis chose a lower limit of … well it set a lower limit, let’s just say that. This makes sense, if a lower seed has a defense that is ranked so far inferior to the higher seed, it is unlikely to prevail. A counter example is that the number of blocks per game was not a criteria that was chosen. In fact, most of the >200 criteria were not used, but that handful of around ten criteria set the filter that chooses a population of games that is more likely to contain a lower seed winning.
 
And that is one of the powerful aspects of this type of analysis, you don’t get the one key driver, or even two metrics that have a correlation. You get a whole set of filters that points to a collection of results that deviates from the “normal”.
 
Please join us as we test our result set this year. We’ll see if we get around 47%. Should be interesting!
 
If you have questions on this type of analysis or machine learning in general, please don’t hesitate to contact Gordon Summers of Cabri Group (Gordon.Summers@CabriGroup.com) or Nate Watson at CAN (nate@canworksmart.com).
**Disclaimer: Any handicapping sports odds information contained herein is for entertainment purposes only. Neither CAN nor Cabri Group condone using this information to contravene any law or statute; it’s up to you to determine whether gambling is legal in your jurisdiction. This information is not associated with nor is it endorsed by any professional or collegiate league, association or team. Machine Learning can be done by anyone, but is done best with professional guidance.
 
 
 

Featured Posts – Click the Brain
Archives
CAN Jewels