Monday, March 25, 2013

A Case for More NCAA Teams...

2013 has proved to be an exciting year for March Madness.  History has been made.  FGCU has accomplished what no other #15 seed has been able to do -- and they did it with ease.  In the West bracket, we almost (controversial call?) witnessed a Sweet 16 without a 1-4 seed which has never happened before.  Finally, I'll leave it to you if you are "shocked" by some of the results! 

The amazing thing is that La Salle, a play-in game participant, is still in the tournament with a good chance to advance to the Sweet 16.  Now another case can be made -- is the First Four enough?  It would appear not.  Can we open the tournament up to more teams?  Should we? 

Typically, the At-Large bids for the tournament do not go higher than the 12th seed.  Those teams grabbing the final spots are known as the bubble teams.  This year, Wildcat nation got the first of two major upsets when they were denied the chance to defend their national title by having their bubble burst.  The most popular "upset" to pick the first round for many - the 5v12 matchup.  From 2002-2012, the #5 seed has held the advantage with a record of 26-18.  This year, the #5 seed went 1-3 to close the gap at 27-21.  This means that almost half the time the bubble team defeats a top 20 caliber team.

It is very difficult for a #12 seed to reach the Sweet Sixteen (as it has happened only 19 times in tournament history).  It is even more difficult to win a Sweet Sixteen matchup with a record of 1-18.  However, isn't the first two rounds what makes March Madness exciting?  Don't we love it when Duke and Missouri fall in the first round in the same year?  I think nearly every basketball fan was either streaming the FGCU game last night or at least clicking refresh every two seconds to see the result.

So let the debate begin.  Should we add more play-in games?  Should the play-in games consist of better matchups?  Does anyone even watch the NIT unless your hometown team is in the tournament?  What's wrong with a little more basketball...     

Friday, March 22, 2013

The Year of the 14 Seed?

We all remember last year's tournament as being notorious for two #2 seeds (Duke and Mizz) being knocked out by #15 seeds.  Yesterday kicked off the second round of the NCAA tournament and it looked like the #3 seeds were going to join Georgetown on the infamous one and out list.

A Curry-less Davidson proved to be a strong matchup against Marquette.  They essentially led the entire game with a couple lead changes mid way (as seen in the Statsheet.com chart below).


As my model had chosen Marquette to be a Final Four contender, this game was of particular interest, if not stressful.  What astonished me was the end of this game.  Statistically, how could Marquette have pulled a victory out of this game?  Looking at the play-by-play still makes me scratch my head. 

With 1:10 left, Davidson held a strong lead at 54-48.  I know you're thinking -- 6 point lead isn't much with a minute left.  I would tend to agree except Davidson looked like they weren't going to miss free throws as they had hit 5 of their last 6.  

It was time for Wilson and Blue to take the game over.  At 1:02, Blue swished a three point shot from outside the elbow, bringing Marquette within three.  This prompted them to clamp down and play tight defense rather than resorting to fouls.  However, a miscommunication on defense led to an easy layup for Davidson while knocking 20 precious seconds off the clock.  

Struggling to get open, Marquette finally got a shot off at the 0:27 mark and nailed another three cutting the score to 56-54.  Marquette either had to force a 10 second violation or immediately foul -- Davidson reached the front-court and Marquette was forced to foul 20 seconds left.  Davidson continued their clutch play by draining both free throws giving them a four point lead.  

Wilson immediately sprinted up the court only to swish another three point shot with 10 seconds left.  Down by 1, Marquette needed a quick foul and some luck on the free throws.  Lady Luck visited them in a different way -- Davidson made a poor pass leading to a turnover with 6.2 seconds left.  

Watching the game live on tv, CBS gave the fans a close look at the Marquette huddle.  It appeared (by the drawings on the clipboard) that Marquette had a play in which they would take a shot at one of the two elbows.  Rather, Blue sprinted in for a beautiful layup to put Marquette up by 1 and ending the game.  

A great game, but more importantly, Marquette had to be perfect to win this game.  In the final 1:33, Marquette went 1-1 on 2pt field goals, 3-3 on 3pt field goals and 2-2 on free throws.  In addition, Davidson in the final two minutes went 1-1 on 2pt field goals and 7-8 on free throws while having a six point advantage.  All it took was one turnover...

In other games, congrats to Harvard for taking down the Lobos (who many predicted as a real contender) to be the lone #14 seed left in the tournament.       

Wednesday, March 20, 2013

The Perfect NCAA Bracket

A special post today:  Perduco Sports' 2013 NCAA Tournament Picks

Although we haven't spent as much time as we would like in setting up this year's tournament picks, we still had to post our bracket!  The entire bracket was created using statistical analysis techniques such as neural networks, factor analysis and a few others -- no subjective input. 

To create the bracket, we utilized the last 10 years worth of tournament results.  We developed models for every possible seed pairing and used season statistics to determine who would be the victor each matchup.  These models are similar to the ones we use for predicting NBA games and spreads. 

We anticipate a strong showing this year -- but next year we should be in full force with optimal NCAA algorithms to pick the perfect bracket! 

Enjoy our picks!

** Click on image to zoom in **

Monday, March 18, 2013

Unsolved Mysteries: NBA Edition #5

An interesting question posed by Kevin Pelton in his 10 questions to be answered in the NBA.

Question 5:  How much effect do players have on their teammates’ statistics?

I am not sold on the idea that this is really a question that needs to be answered.  I think the more important question is how players affect the team’s statistics – not individual teammates.  Yes, it is true that high scorers, elite rebounders and elite passers will take away shots, rebounds and assists from other teammates, but this isn’t necessarily what is of interest (unless you are an agent trying to get your player more stats).  The true question is what effect the player has on his team while he is on or off the court. 

The most common ways to look at this problem is the +/- or different per minute statistics.  From this, we can see what overall effect the player had on the court.  Basketballvalue.com does a fantastic job at looking at the efficiency of players in different lineups.  They aggregate the entire season and see how each player did with every possible lineup throughout the season.

Currently, Perduco Sports is examining how a player contributes in different areas (shooting, rebounding, turnovers, etc.) throughout each game.  We want to determine at what point during a game a particular player is more efficient and has a positive effect on their team.  The graphs below show the difference in Houston stats (turnovers per minute and personal fouls per minute) when Patrick Patterson was on the court (blue) versus when he was off the court (red).  We analyzed each minute of the game as whole minutes.  Next, we looked at the opponents stats (points per minute and rebounds per minute) when Patterson was on the court versus off the court. 

 
 
 We can see that at the beginning of the second quarter and at the end of the game, the opponents score a lot more points per minute (bottom right) when Patterson is parked on the bench.  Rebounds per minute (bottom left) show the same trend; however, this is throughout the entire game.  Although Patterson may not be rebounding as much as one thinks, he is forcing the opponent to get fewer rebounds per minute with his presence on the court.  Houston also commits fewer turnovers (top left) and personal fouls (top right) per minute with Patterson running the floor. 

Looking at analysis like this allows analysts, coaches, and teams to see the true impact of a player during the game.  This is just the beginning phase in determining how a player contributes to their team’s success.  After all, basketball is a team sport! 

Friday, March 15, 2013

Unsolved Mysteries: NBA Edition #4


Now we start getting to the hefty statistical questions presented by ESPN Insider Kevin Pelton.

Question 4: How do players’ roles on offense affect their efficiency?

Since joining the sports domain and talking to analysts across all of the major sports, this question seems to hit the mark on one of the most important questions to answer.  Each player on the court plays a different role – whether they are volume scorers, rebounders or key defenders.  The important thing to keep in mind is that not all roles can be compared the same.  Therefore, one efficiency metric is not sufficient in this type of problem. 

Eli Witus’ article is well written and discusses usage versus efficiency between highly used players and low usage players.  Eli uses individual offensive rating and individual possessions as metrics.  Again, it all comes down to what metric is used to measure efficiency.  It is important to remember to classify players differently based on their roles. 

At Perduco, we are examining this problem even closer.  To truly measure efficiency versus usage, it is important to take into play fatigue and where in the game fatigue occurs.  It’s possible that a starter is efficient for three quarters and sees diminishing returns when reaching the 40 minute mark.  However, that same player could excel in 4th quarters.  This leads to the assumption that there is fatigue or poor efficiency somewhere else in the game. 

The graph below shows some very preliminary work being done to show per minute efficiencies of players.  We are examining how a player does at each minute across an entire season (i.e. is the player more effective at different points in the game?).  The graph shows the pts/min for Patrick Patterson when he played for the Houston Rockets in the shortened 2011-2012 season.  We can see a dip in performance at the beginning of second halves throughout the season.  Patterson saw lower usage at these game times and a lower efficiency (we are only showing pts/min) as well.  We will continue to post our results that will look at how players perform throughout the game rather than just games in general.  
 

Friday, March 8, 2013

Unsolved Mysteries: NBA Edition #3

We approach an interesting question proposed by Kevin Pelton in his 10 questions. 

Question 3:  What is the best way to develop young players?

No matter how statisticians approach this problem, this could be the most difficult topic to convince coaches/staff that statistics can answer the question on how to develop NBA players.  Every player is different and a combination of talent and coachability comes into play.  Some players thrive with a mentor present while others need to find their own path to greatness.  In the NFL, would Aaron Rodgers be the quarterback he is today if he started playing immediately in his career?

Too much variation exists among how players are developed – even from the young toddler stage.  Trying to use statistics to answer this question is almost impossible since there is so much missing data and backstory that all play a role in player development (in terms of how Pelton’s describes the problem).  It’s hard enough to convince coaches on how to execute other aspects of the game based on statistics – trying to influence how to develop/coach a player based on “unproven” statistics could be a bit far-fetched. 

Pelton mentions the role of minutes young players play.  I believe this could be a good research problem in terms of how players do over their career based on the minutes they played early on.  However, this does not answer any part of the question of how NBA players develop.  Nothing is taken into account about the type of coach, mentors, teammates, previous coaching, etc.  

Although an interesting and important issue, this author believes this problem is more of a philosophy question rather than one to be answered by statistical analysis.   

Wednesday, March 6, 2013

Unsolved Mysteries: NBA Edition #2

We continue our examination into Pelton’s 10 questions to be answered.

Question 2: How do basketball players age?

Pelton references a WSJ article that states that 25 is the age at which basketball players peak in performance.  Pelton performed his own analysis and found a peaking age of 27.1, which is in line with MLB players.  Both of these analyses focus on wins per season or improvements with respects to wins. 

An interesting problem indeed – yet, approached in a limited fashion.  First, as the authors have admitted, not all players age the same.  But to choose essentially one metric to classify how a player ages is too simplistic.  Why not look at how players at different positions age differently?  Is it appropriate to put Shaq and Stockton in the same group to measure age – I think not. 

Pelton discusses how rebounding age peaks earlier, hence why his age is 27 and the WSJ found an age of 25.  Not every basketball player is the same; therefore, let’s not classify them all the same.  In addition, every study is always looking at Lebron, Kobe, Duncan, etc.  If we are asking how elite players age, that is a different question than how a basketball player ages.  Elite players are on their own level.  Let’s reach back to all the other players in the league that seem to be around for a long time, yet may not have peaked such as Reddick, Delfino, and Novak.     

I’m not sold on the idea of wins per season as the metric of a player peaking in age.  It should be based on their position and expected contribution over time (given numerous stats and the position’s expectations).  Every player contributes differently and simply because a player does not produce wins per season doesn’t necessarily mean a player has peaked as an individual basketball player. 

Tuesday, March 5, 2013

Unsolved Mysteries: NBA Edition #1


ESPN Insider Kevin Pelton, inspired by the MIT Sloan Sports Analytics Conference, wrote an excerpt about 10 questions to be answered in the NBA, which are similar to David Hilbert’s 23 problems in mathematics and Keith Woolner’s 23 problems in Baseball.  Over the next 10 days, we will examine each question that Pelton poses and give our insight into the approach of the problem.

Question 1: Which better describes player value – individual stats or team impact?

Pelton nails it with his answer of “both.”  He also describes using SportVu’s optical tracking to better understand how screens and shot defense contribute to a player's value.

The first question to answer is what is meant by the word “value.”  If we are talking about monetary value, many could argue that individual stats are all that matter.  Take the case of Dwight Howard.  He puts up individual stats that warrant big bucks (with the exception of his Shaq-like free throws) but doesn’t add any real value to his teammates, as evidenced in Orlando and now in L.A.  If we shift the focus to team value, then we argue about the team impact of the player.  Players like Rajon Rondo and James Harden create opportunities for their teammates to shine – although the case could be made that they are now shifting to the type of players that are projecting All-Star individual stats. 

I think of player value as simply their team impact.  However, this can mean both how they do individually (their individual stats) and how their teammates perform while on the court.  Some measure of the +/- statistic (whether raw, adjusted, weighted, etc.) goes a long way to determine a player’s value.  Not only that, but Mark Cuban is on to something by having Don Kalkstein sit behind the bench at every game.  There is something to be said about the psychology of the sport.  There are reasons that NBA teams can lose 20 point leads that go beyond the physical realm.

In the future, Perduco Sports will examine the role that psychology plays into the game – and be able to quantify that metric.  There are other approaches that we will take to explain player value, but I will reserve those until we reach some of the other NBA questions to be solved.

Monday, March 4, 2013

SportVu: An NBA Analyst's 'Holy Grail'

Brian Kopp was interviewed at the MIT Sloan Sports Conference regarding STATS SportVu data.


This data is truly the future of all sports analytics.  Kopp discussed every possible way that coaches and owners could use this data to improve their team.  There are numerous pros and cons when dealing with this data and we are going to indulge in a couple. 

First, the possibilities are endless with this data.  Knowing where every player and the ball are at every second of the game allows us to purely quantify the movement of the game.  We can regain the countless hours spent watching video and allow computers to determine what types of plays were run during different phases of the game.  A player’s efficiency can be measured with different lineups and formations on the court.  If you have ever wondered if spacing the floor is more efficient than crowding the paint – we can now answer these questions with ease. 
One downside to this data is the amount of data per game and its possible complexity.  Watching the video and reading articles on SportVu gives analysts a sense of the amount of work required to simply parse the data.  Rules and conditions need to be established to determine what truly constitutes a pick and roll or when the floor was officially spaced.  Kopp has mentioned that it takes professionals to sift through the data and determine meaningful input for analysis.

One more major downside is the availability of this data.  Not every team possesses this data as it comes with a hefty price tag.  Therefore, for us NBA enthusiasts not working for one of the 15 NBA teams, we have to dream about what we could do with the data.  We at Perduco Sports have grand ideas on how to use this data and hope we are able to secure some in the future to showcase its potential.  Until then, we will continue to salivate and wonder how those lucky few are utilizing this massive set of data.