## Saturday, April 25, 2015

### The Improbability of the Warriors' Comeback

Hindsight has a way of making the improbable seem inevitable. Of course the Warriors erased a 20 point deficit in the fourth quarter (despite being only the third playoff team to do so). Of course Steph Curry hits that game tying three from the corner to force overtime (despite having missed from the wing just three seconds prior).

But a 20 point comeback is anything but inevitable, and we tend to forget the games in which a blowout stays a blowout because, well, those games are forgettable. So what do the numbers say?

My own win probability model put the Warriors chances as low as 0.2%. That low point occurred after a miss by the Warriors' Shaun Livingston with 6:24 left in the game and Golden State down by seventeen. Livingston would rebound his own miss for the put back slam dunk, tripling his team's chances to 0.6%.

## Sunday, April 19, 2015

### Live Win Probabilities for the NBA Playoffs

Barring technical difficulties, I will be sharing live versions of my win probability graphs for the remainder of the playoffs. The live charts at Advanced Football Analytics have long been a fixture of my NFL viewing experience on Sundays, so I'm glad to be able to offer something similar for the NBA.

Here is the link: Live Win Probability Graphs for the NBA. The graphs currently show yesterday's games, but will update once today's games get underway.

The charts do not update automatically, so you'll need to hit the "Refresh!" button for now to get the latest probabilities. There is also a bit of lag in the data.

In addition to win probabilities, player statistics are also tracked real time, including the current MVP and LVP (Least Valuable Player) according to win probability added.

The Excitement Index and Comeback Factor are tracked real time as well. The Excitement Index just tracks the cumulative movement in win probability over the course of the game. For complete games, the Comeback Factor represents the victorious team's odds of winning at their lowest point. To calculate this real time, I have modified the definition. Comeback factor at any point in the game represents the winning team's odds at their lowest point minus their current odds of winning. For example, if a team has a current win probability of 75% (odds = 1/3), and their lowest point in the game was 25% (odds=3/1), then the comeback factor is: 3/1 - 1/3 = 2.7. This definition reduces to the complete game version since a win probability of 1 is equal to 0 odds.

This is still a somewhat experimental feature, so any feedback is appreciated, here in the comments, or on Twitter.

## Friday, April 17, 2015

### The Game that Broke My Win Probability Model

On the eve of the playoffs, I take a look at the most improbable game of the 2014-15 NBA regular season.

Earlier this year, I rolled out an improved version of my NBA win probability model. The new model resulted in better predictive accuracy in out of sample testing. It also reduced the number of "impossible" comebacks. Prior to the upgrade, there were several games in which a team with a win probability of zero ultimately came back to win the game. This would indicate an overconfident model.

After the new model was rolled out, most of those "impossible" comebacks were downgraded to merely "improbable". Except for one.

## Wednesday, April 8, 2015

### Playoff Seed Motion Charts return for the NBA

2015-04-15: Note, it appears that I had one of the tiebreaker scenarios incorrect (for the top 4 seeds, a non-division winner always loses the tiebreaker to a division winner. These charts overestimated the Clippers' chances at a two seed).

My NBA betting market rankings are published daily, and with each version I use those rankings to simulate the remainder of the regular season and summarize playoff seed probabilities. As I did last year, I have loaded those daily probabilities into an interactive motion chart that summarizes the evolution of each team's season:
• In mid-December, the Atlanta Hawks are a solid 3/4/5 seed. Just a month later, they have a chokehold on the Eastern Conference top seed.
• The Indiana Pacers have spent the entire season straddling that Playoffs-Lottery line. We'll see if the return of Paul George can tip the scales in their favor.
• Masochistically-inclined fans of the Phoenix Suns can relive the month of February - the month their playoff hopes slowly blinked out of existence.

## Sunday, April 5, 2015

### Detailed Per Possession Statistics now Available

What do last year's 76ers and this year's Warriors have in common? Certainly not offensive efficiency, where the Sixers ranked dead last in points per possession, compared to the Warriors' current #4 rank. The answer is pace. Over the past four NBA seasons, The Sixers and the Warriors are the only two teams to average less than 14 seconds per possession on offense.

Measures of pace and per possession efficiency, long a staple of advanced metrics, have historically been limited to what can be gleaned from the daily box score. In recent posts, I have attempted to extend the understanding of pace and efficiency by using detailed play by play data, rather than box score aggregations. For example, the Golden State Warriors have one of the league's slowest paces on defense, despite leading the NBA in overall tempo. And the Brooklyn Nets have a criminally low offensive efficiency for possessions that begin off of turnovers.