2016 NBA Finals series win probability

nba_champ_trophyFor the NBA playoffs, I developed rankings that use both data from games and the markets. These numbers give the following win probability for the finals.

Golden State has a 75.3 percent chance of winning the series.

This number has changed since the start of the playoffs.

Back on April 16th, the Warriors had just won a record 73 games during the regular season. My numbers said the Warriors were 4.5 points better than the Cavs on a neutral court, which implies a 83.4% series win probability.

Since then, Steph Curry got hurt, and perhaps hasn’t played up to his MVP form since his return. The Warriors struggled in a seven game series against Oklahoma City, getting outscored by 7 points during the series.

Meanwhile, Cleveland has played exceptional during the playoffs, as they have lost only two games (both in Toronto). Shooting 43.4% from three compared with 36.3% during the regular season has helped.

Now, the numbers imply Golden State is three points better on a neutral court than Cleveland, which gives the 75.3% win probability.

As of Tuesday morning (May 31st), the markets imply a 65.5% win probability for Golden State.

NBA series win probabilities for the 2016 conference finals

These numbers come from rankings that use data from games and the markets. To see my numbers for the entire playoffs, check out the interactive visual for NBA win probabilities.

Oklahoma City vs Golden State.
Golden State has a 78.5 percent chance of winning the series.

Toronto vs Cleveland.
Cleveland has a 75.1 percent chance of winning the series.

The numbers point to a Golden State versus Cleveland rematch of the NBA finals.

To see update numbers on these series probabilities, check out the predictions page.

How does Steph Curry’s injury impact the Warriors series win probability?

Steph Curry sprained his knee and will miss the next two weeks of the playoffs.

How will this impact the Warriors, the clear championship favorite with Curry? We can estimate his impact by looking at the closing spread against Houston with and without him.

For example, the markets made Golden State a 3.5 point favorite at Houston in Game 3 without Curry. This moved to 8.5 points in Game 4 with Curry. Let’s use this to estimate Curry means 5 points per game to the Warriors.

With Steph Curry

With Curry, my numbers give these series win probabilities against the Warriors next two opponents.

  • Los Angeles Clippers: Warriors have a 92.2 percent chance of winning the series.
  • Portland: Warriors have a 95.1 percent chance of winning the series.

Unless the sky falls in, the Warriors should win these series with Curry.

The road gets tougher in the conference finals, but the Warriors should win.

  • San Antonio: Warriors have a 69.5 percent chance of winning the series.
  • Oklahoma City: Warriors have a 80.0 percent chance of winning the series.

Without Steph Curry

Without Curry, the story dramatically changes.

  • Los Angeles Clippers: Warriors have a 69.6 percent chance of winning the series.
  • Portland: Warriors have a 77.7 percent chance of winning the series.

Now, Golden State is vulnerable against either the Clippers or Blazers, even though they should still win.

In the conference finals, the Warriors will no longer be the favorite.

  • San Antonio: Warriors have a 33.9 percent chance of winning the series.
  • Oklahoma City: Warriors have a 46.7 percent chance of winning the series.

NBA championship probabilities

What about the chance of winning an NBA title? With Curry, my numbers give the Warriors a 59.0% chance, which hasn’t changed much since the beginning of the playoffs. Without Curry, their championship probability drops to 16.2%.

Duh, the Warriors aren’t as good without Curry. However, the numbers and market data put in perspective the importance of the NBA’s reigning Most Valuable Player.

How to use analytics to build an NBA champion according to book Chasing Perfection

chasing_perfectionYou love basketball and need an insider look at the NBA.

The stories behind a trade are not enough. It’s 2016, and you know analytics and technology play a huge role in how teams improve their chances at a championship.

In his book Chasing Perfection: A Behind-the-Scenes Look at the High-Stakes Game of Creating an NBA Champion, former Sports Illustrated writer Andy Glockner takes an intimate look at the modern NBA. To the extent that teams will talk, he explores how numbers, health and good old common sense all play a role in the management of a team.

Let’s look at some of the most intriguing stories in the book.

How to build a champion

Golden State has become the marvel of the NBA. The Warriors won the 2015 NBA championship and won a record setting 73 wins the following regular season.

Steph Curry gets most of the headlines because of his ground breaking shooting, and deservedly so. However, Glockner points out that the Warriors “have up to a dozen players with wingspans of nearly seven feet to rotate among the shooting guard, small forward, and power forward positions.”

This length has made the Warriors a sneaky good defensive team. In terms of points allowed per possession, Golden State has ranked first and sixth in 2015 and 2016 respectively.

Chasing Perfection also has another interesting nugget about the Warriors. Assistant GM Kirk Lacob admits that they are “not at the top of the league in terms of either personnel or resources thrown at data analysis.”

And maybe that’s ok. To win a title, it helps to have Steph Curry develop from above average guard to Most Valuable Player. It also doesn’t hurt to have the greatest talent evaluator in NBA history, Jerry West, as an advisor.

How not to build a team

If the Warriors are drinking champagne in the NBA penthouse, the Sixers are crawling through the sewage pipe below the building.

Sam Hinkie brought analytics and new ideas when he became the general manager of the Sixers. As documented by Pablo Torre of ESPN, one of those ideas was to draft players with a long wingspan.

You can teach a player to shoot a basketball, but you can’t teach length. Just like the Warriors, the Sixers collected a stable of long wing players. Then they hoped that these players would make Kawhi Leonard type improvements in shooting.

It didn’t work. The Sixers have been the worst team in the NBA since Hinkie’s arrival. While the defense has been serviceable, the offense has been atrocious.

And it gets worse.

Glockner writes the Sixers took most of their shots from three or near the rim. Since these shots have the highest efficiency, this seems like a good strategy. Not for the Sixers, who have finished a distant last in points scored per possession in each of three seasons of the Hinkie tenure. From Chasing Perfection:

While on a normal team, you might criticize the coach for consistently creating shots his players couldn’t convert, the idea was backwards with the 76ers. (Coach) Brown wanted to create these decent shots, and then have his players (or new ones) learn to make them.

This is utter stupidity. The Sixers set up their players for failure. It’s the opposite of the Spurs strategy, which seeks to maximize the ability of each player.

In April of 2016, Sam Hinkie was fired.

The technology revolution in the NBA

The most interesting part of the book details the technological revolution in player health.

For example, the comany P3 (Peak Performance Project) has an apparatus to measure the force in each leg and make a movie as a player jumps. If you find that one leg produces 20% more force than the other, you predict an injury could be looming for the player.

P3 took their technology to the 2014 NBA pre-draft combine. After testing all the players, they ranked the top 60 players by their likelihood to get hurt. They also predicted the location of the injury.

While Chasing Perfection doesn’t provide any details, P3 said the list “ended up being very predictive.”

P3 also helped Hawks guard Kyle Korver. In a chapter devoted to the sharp shooter, Glockner tells the story of how Korver went from a broke down player who thought about quitting to the healthy, valuable player for the Hawks.

Health analytics played a role in this recovery. P3 showed Korver how one of his knees bowed every time he took a shot. Horrified, Korver got a program to strengthen his body and fix his problem.

This all seems so logical. Players benefit from balance and symmetry in their muscles. With technology, P3 can identify potential problems and get players on a program to fix this muscle imbalance.

The details of P3 also shed light on a cryptic quote from the book Soccernomics
by Simon Kuper and Stefan Szymanski.

AC Milan’s in-house medical outfit found that just by studying a player’s jump, it could predict with 70 percent accuracy whether he would get injured. It then collected millions of data on each of the team’s players on computers, and in the process stumbled upon the secret of eternal youth. (It’s still a secret: no other club has a Milan lab, and the lab won’t divulge its findings.)

There’s no doubt that the Milan lab had similar technology to P3.

Best analytics nugget

Jon Nichols, who now works with the Cleveland Cavaliers, did an interesting study on which college basketball statistics best predict NBA performance. He found that block rate translates best, even more than rebounding and assist rate. The correlation is surprisingly strong.

Best quote

Tom Penn spent 11 years as an NBA executive before moving to ESPN in 2010. He said the following about the adoption of analytics:

Every team over the last fifteen years – doesn’t matter whether they believe in it or not – they do this in order to cover their tail and to demonstrate that they are sophisticated.

Teams do analytics just for show. I wish Penn said things that hilarious and insightful on ESPN.

Most unbelievable story from the book

Buzz Williams, the coach at Virginia Tech, gives his players a weekly talk to make them comfortable with data. The topics range from personal finance to brain science, but they happen each week during the off season.

Holy shit, that’s a huge commitments to the well being of your players. Williams, a long time believer in basketball analytics, may have made a strange move from Marquette to Virginia Tech. However, he deserves recognition as one of the more innovative minds in basketball.


For the hard core hoops junkie, Andy Glockner’s Chasing Perfection gives an inside account of how NBA influencers incorporate data and analytics into their decisions. Of course, it doesn’t tell all, since teams didn’t spill everything to Glockner.

However, the book contains many fascinating anecdotes that reveal the inner workings of basketball. I could have written another thousand words telling the stories like how Ben Alamar’s analytics convinced Oklahoma City to draft Russell Westbrook.

However, it’s best to let Glockner tell those stories in Chasing Perfection.

Can anyone beat North Carolina? A 2016 Final Four preview

FinalFour_2016_winprobWhich team will win the Final Four? A talented North Carolina is the clear favorite, as my numbers give them a 47% chance to win.

However, three other hungry teams will attempt to knock off the Tar Heels in Houston. In a year of college basketball parity and complete March insanity, can one of these teams do it?

Let’s break down the four teams.

North Carolina

Before the tournament, I wrote this about the Tar Heels.

However, there’s something not quite right about this team. They don’t feel like the top team in the nation when you watch them play.


North Carolina has looked every bit the championship contender. There were plenty of pre-tourney signs that said they would arrive at the Final Four as the front runner.

  • North Carolina was first in my college basketball rankings entering the tournament (still are, check out rankings here)
  • North Carolina began the season first in the preseason AP poll, a surprising predictor of post season performance.
  • Roy Williams elevates his team the most in the tournament according a one of my studies.

Of course, North Carolina has had their share of good fortune as well. They didn’t have to play Kentucky, West Virginia or Xavier, the other top teams in the East Region by my numbers.

Prior to the start of the tourney, North Carolina didn’t have the best odds to win partially due to this tough region. The numbers like Kansas out of a weaker Midwest region.

Now, North Carolina is a clear Final Four favorite, and my numbers give them a 47% win probability.


Before the tourney, I warned people about picking Nova as champion. The Wildcats took 44% of their shots from three, and these high volume three point shooting teams tend to not win the tournament.

However, Villanova has shifted their offensive strategy over the last three games. Against Iowa, Miami and Kansas, the Wildcats have taken 33% of their shots from three. They no longer live by the three, die by the three.

Over these three games, the Wildcats have shared the ball. No one player has taken more than 11 shots except for Jason Hart (18) against Kansas. With their typical stellar defense, Villanova made the Final Four.

Villanova has also had their good fortune. They shot a blistering 59% (20-34) from three against Iowa and Miami in comfortable double digit wins, much better than their 34% average for the season.

Against Kansas, Villanova didn’t shoot that well. However, Kansas had a miserable shooting game from three (4-18). It got painful watching Wayne Selden miss shot after shot from behind the arc. The pain reminded me of watching Mike repeatedly call a girl in Swingers.

Villanova is 2nd in my team rankings with a rating a hair below top ranked North Carolina. These rankings would predict a toss up between these two teams in the title game.

However, my team rankings probably overestimate Villanova. My algorithm tends to weight games against good teams more, and Nova had some good fortune from behind the arc against their last 3 quality opponents.

I also calculate market rankings based on the closing point spreads in the markets, a predictor not as affected by the noise of game results. These numbers rate Villanova about 1.5 points worse than North Carolina, and I expect that to be the line should these two teams meet in the final.


The Sooners are on a roll, as they blew threw the West region to secure a Final Four berth. Buddy Hield earned almost mythical status with 37 points against top seeded Oregon in the regional final.

Oklahoma came through on their 38.5% chance to make the Final Four by my numbers, highest of any team in the West Region. However, I still see flaws with this team.

Oklahoma takes a high rate of three point shots (41% of their field goal attempts). In addition, they have hit on 43% of these three pointers, a rate destined to regress against better defenses in the Final Four.

The markets have Oklahoma as a two point underdog against Villanova, and all of my calculations agree with this point spread.

But wait. Didn’t Oklahoma beat Villanova by 23 points this season? Yes, but the Sooners got blessed with good fortune from three point range. They shot a blistering 14 of 26 from three, while Villanova struggled in going 4 of 32.

My calculations do consider this game, but it’s one of 30 some games that contribute to my team rankings. Looking over a more complete body of work, Villanova is the better team.

Oklahoma can certainly beat Villanova, as a just a bit of jump shot luck can easily make up 2 points. However, Villanova defense ranks 5th in the nation by my points per possession adjusted for strength of schedule. Good looks from three will not come easy.

The odds should favor Villanova to beat Oklahoma.


Many pundits thought Syracuse shouldn’t have made the tournament at all.

The Orange lost to a woeful St. John’s team this season and couldn’t even win their first game in the ACC tournament. And perhaps this wasn’t surprising, as Syracuse didn’t get a single vote in the preseason AP poll.

However, the committee did select Syracuse. And in a year of college basketball parity, Syracuse made the Final Four as a 10 seed.

Syracuse wins with defense. Jim Boeheim recruits tall and long players to execute his zone defense, and this defense ranks 17th in adjusted efficiency numbers.

To make the Final Four, Syracuse beat Virginia when they Malachi Richardson hit a series of contest three point shots. Can this continue against North Carolina? Probably not, as my numbers give them only a 15% chance to pull the upset.