Computing win probabilties for NFL teams


My fantasy team uses a playoff system that Bill Simmons suggested a few years ago. In order to study it, I'd like to run Monte Carlo simulations of last year's post-season. To do so, I need to be able to compute the win probability for a pair of teams given the site of the game. Note this is not a prediction against the spread, just the orders of one team winning. This format also means that I can't use Vegas's point spreads, money lines, etc. because I need to be able to make predictions for pairs of teams that didn't meet in real life.

I thought that it would be easy to find a suitable model online, but I haven't been able to. Here's what I've tried:

AdvancedNFLStats: In principle they supply all the needed information but I've been unable to reproduce their results. You can see a long post from me at the end of the week 7 team efficiency page that has gotten no replies. I would love to be able to get this model working. NFL Predictions uses stats from AdvancedNFLStats...
0 0

Here I generated a model which computes the probability of an NFL team winning a game based on the state of said game (i.e. field position, score, time remaining, etc.). Using NFL play by play data from 2002-2011 (only free data I could find), I generated a model using an artifical neural network. It is important to note that this model assumes the home and away team equal, i.e. there is no information about individual teams or players on the team. Therefore the probability of the 2008 Detroit Lions (0-16 year) beating the New England Patriots in Detroit would be the same as the probability of New England winning that same game in Foxboro (as a Lions fan I can tell you that is far from realistic).

The neural network used the following 7 input variables to create the model:

Who has the ball (Home or Away)Quarter the game is inTime remaining in quarterScore differentialCurrent downYards to go for a 1st downCurrent field position

By filling out the form below, you...

0 0
Football Commentary: A Computer Program For Playoff Tiebreakers

A model-based approach to football strategy.


In this article we present a computer program that applies the NFL tiebreaking rules to determine the playoff teams and seedings in each conference, given actual or hypothetical outcomes for all the games in the regular season.1 This determination, often tedious when done by hand, takes a computer only a fraction of a second. We exploit the program by embedding it in a simulation. The simulation produces an estimate of each team's probability of making the playoffs or of winning the Super Bowl, given probabilities for the outcomes of the remaining regular-season games.

One possible application of the simulation is to help bettors estimate the proper odds for each team to win the Super Bowl, at least late in the season when the tiebreaking rules can have an important effect on teams' prospects.


0 0

Desperate for football, I'm watching the Hall of Fame Game tonight. There's four minutes left in the fourth quarter and the Redskins are up by 7 over the Colts. Can Jared Lorenzen lead his 4th-string squad to a comeback? I must be the only person in the country who cares. And I only care because I'm investigating Win Probability (WP) in NFL football.

WP is simply an in-game estimate of who's going to win based on the current score and other game variables. This post will examine the potential application of WP and will illustrate a first cut at actual WP for various scores and time remaining.

Two of my recent posts discussed measures of utility in football. I looked at first down probability and at point expectancy. First down probability analyzes how likely an offense is to convert their present down and distance situation to a first down. The success of a play can be judged based on how it changes the probability of a first down. Point expectancy measures how many...

0 0

These NFL football computer predictions are based on a variety of factors that are completely determined based on previous games’ data. These predictions do not account for injuries, free agent signings, weather or any uncontrollable factors that could cause future game outcomes to be swayed in one direction or another. See my predictions disclaimer page for more information about these predictions: Predictions Disclaimer

Also note, that the Predictor should be used for entertainment purposes only. However, if you do wish to learn more about recommended sportsbooks, then I highly suggest reading: The Top 5 Online Sportsbooks for U.S. Players.

All upcoming NFL football predictions are listed below. There you will find information on probabilities to win, predicted score, total score and more.

NFL Computer Predictions – 2017/2018 Season

NFL Game Predictions: Week 1

Bet NFL Football at Bovada – Our top-rated Sportsbook: Join Bovada

0 0

If you have been watching the NFL, you have probably heard some reference to win probability (or WP). It most likely came up when something really crazy happened, like when the Jets won the opener on an improbable penalty, or at the end of the Ravens-Vikings game when the teams combined for five touchdowns in the last 2:05. Maybe it was in reference to the Dallas Cowboys, who seem to make the improbable happen all the time.

How accurate is win probability? To test, I looked at a specific point in time for every game in the NFL this year, and logged who was favored to win and by what win probability according to pro football reference. The Pro Football Reference win probability uses current margin, remaining time, and expected points on the possession at the point in game being measured, and also uses the point spread. Thus, a heavy favorite is seen as more likely to make a comeback, or win a close game, than a heavy underdog would be.

The point in time I used...

0 0

Before scrolling down and peeking at the table, take a guess: Which team has the highest projected end-of-season win total according to FiveThirtyEight’s latest NFL Elo ratings and playoff odds?

While you’re thinking, let me briefly explain what these numbers are all about. A team’s Elo rating represents its current strength — or at least an estimate thereof using a simple method that incorporates margin of victory, home-field advantage, strength of schedule and team quality in previous seasons. (For the really gory details of how the ratings work, click here.) To give you a sense of the scale, the average NFL team always has an Elo rating of 1500, and the ratings usually range from 1700 on the high side to 1300 on the low end of the spectrum.

Anyway, once computed, they can be used to derive win probabilities for each game and even point spreads. That’s how we’re able to simulate the remainder of the NFL schedule thousands of times and track each team’s chances of...

0 0

People love sports for being unpredictable, but that doesn’t mean sports are actually unpredictable. It just means they feel that way. And no one knows this better than Bobby Skoff because Bobby Skoff, the Co-Founder of Swish Analytics, has an automated machine learning platform. His machine sucks in statistics, runs them through a set of algorithms, and pumps out player level predictions for every major U.S. sport. Does he know what’s going to happen before it happens? Of course not. He’s just a lot better at guessing than the play-by-play guy.

Among all the statistics his platform can produce, the one that stands out for Skoff is win probability, the percentage chance that a given team can eke out a victory at any given point in a contest. He’s an evangelist for this information because it allows fans to better understand what is happening in any game, but particularly in football games, which are governed by arcane rules that experts often struggle to explain (just ask a...

0 0

Below are win probability charts that help tell the story of every game in the 2016 NFL postseason. The most recent charts will be added to the top of this post shortly after games conclude. Check back for a look at the biggest moments of the playoffs.

Win probability measures the chance that a team will win a game, given a particular combination of circumstances, including score, time remaining, field position and down and distance. Win probability is based on a model built on actual outcomes of NFL games from recent seasons that featured similar circumstances.

Patriots 34, Falcons 28 (OT)

How improbable was the Patriots victory?

The Patriots' chance of winning bottomed out at 0.3 percent (after Julian Edelman's incomplete pass). There were 20 different points in the game in which the Falcons' win probability was 99 percent or greater.

What plays had the biggest impact by win probability? The pass interference that moved the Patriots from the...

0 0

Win probability is a statistical tool which suggests a sports team's chances of winning at any given point in a game, based on the performance of historical teams in the same situation.[1] The art of estimating win probability involves choosing which pieces of context matter. Baseball win probability estimates often include whether a team is home or away, inning, number of outs, which bases are occupied, and the score difference. Because baseball proceeds batter by batter, each new batter introduces a discrete state. There are a limited number of possible states, and so baseball win probability tools usually have enough data to make an informed estimate.

American football win probability estimates often include whether a team is home or away, the down and distance, score difference, time remaining, and field position. American football has many more possible states than baseball with far fewer games, so football estimates have a greater margin of error. The first win...

0 0

If you wanted to figure out the probability that your favorite football team will win their next game, how would you do it? My colleague Eduardo Santiago and I recently looked at this question, and in this post we'll share how we approached the solution. Let’s start by breaking down this problem:

There are only two possible outcomes: your favorite team wins, or they lose. Ties are a possibility, but they're very rare. So, to simplify things a bit, we’ll assume they are so unlikely that could be disregarded from this analysis. There are numerous factors to consider. What will the playing conditions be? Are key players injured? Do they match up well with their opponent? Do they have home-field advantage? And the list goes on...

First, since we assumed the outcome is binary, we can put together a Binary Logistic Regression model to predict the probability of a win occurring. Next, we need to find which predictors would be best to include. After a little research, we found...

0 0

In this post, I will describe my attempts to model the probability of a goal being scored in soccer. After correcting for team imbalances, I find that a trailing team has a


probability of scoring in most situations. This result has potential implications for strategy and whether teams should be adopting a more aggressive style of play.

The Model

Using the same dataset I used for my

win probability model

(~3,000 matches from five of the top European Leagues), I employed


smoothing to build a model that predicts the probability of a goal being scored within the next minute of game time. The model is a function of the following:

game time goal difference team strength

I derive the team strength from the pre-match betting odds, and convert it into an expected goals scored per game. Including team strength as a parameter is crucial for this type of analysis, because the model is also a function of goal differential. There...

0 0
Win Probabilities{"topic":{"topicId":4032873,"title":"Win Probabilities","photoPath":"","markLike":0,"articles":1},"errors":{},"location":{"country":"Russia","state":"Moscow","city":"Moscow","latitude":55.75222,"longitude":37.61556,"photoPath":"","locationId":11094,"label":"Russia, Moscow,...
0 0

What Real-Time Gambling Data Reveals About Sports: Introducing Gambletron 2000

What are the most exciting matchups in sports? What’s the most exciting sport? What if a computer could tell us which games are hot right now—like an NFL RedZone channel, not just for NFL football, but for basketball, soccer, hockey, and baseball?

Introducing, a tool that uses live in-game gambling data to quantify excitement in sports, write automated game recaps, finally settle the debate about whether the first half of NBA games is even worth watching—and much, much more. It might even make you rich.

It’s 12:52am on January 2nd, 2007. Boise State has just shocked Oklahoma to win the Fiesta Bowl. Meanwhile, I’m staring at the game’s TradeSports page, watching the gambling odds fluctuate almost as wildly as the game itself. The last few minutes of the TradeSports graph looked like an EKG gone haywire: one minute Boise State had a 90% chance to win, then...

0 0

During games, the process gets slightly more complicated. First, we need to modify Winston’s formula to account for the diminishing amount of time remaining in the game. To quote Winston again:

“If we assume that the changes in margins during different parts of the game are independent and follow the same distribution (the technical term is identically distributed), then the standard deviation of the margin during n minutes of [a] game is:

(game standard deviation of margin) / sqrt(fraction of game that n minutes is)”

Using the 13.45 standard deviation we derived earlier, that formula is as follows for NFL games:

STDEV = (13.45 / SQRT((60 / minutes_remaining)))

So after 1 quarter, the expected standard deviation of scoring margin goes from 13.45 at pregame to 11.65, etc.

In addition to modifying the standard deviation about the mean, we also need to adjust the mean (the Vegas line) itself to account for the reduced amount of time remaining...

0 0

Using the handy spread-to-moneyline converter available at SBR Forum, we came up with tables for favorite and underdog win percentages based on the point spread across the NFL, college football, NBA and college basketball.

Note: Apologies to those who had this post bookmarked in the past. That post has been deleted, but here it is again. Enjoy. And bookmark!

Note II: You can also use these to calculate projected wins for a season based on projected — or actual — point spreads. For example, the Seahawks are favored by 5 over the Packers in the Thursday Night opener, so that’d count as 0.681 wins for Seattle and 0.319 wins for Green Bay. Do that a bunch of times and you get a rough estimate of the “Vegas projection” (or, more accurately, the CG Technologies projection) of wins for each NFL team.





0 0

WARNING: Math post.

PFR user Brad emailed over the weekend with an interesting question:

"Wondering if you've ever tracked or how it would be possible to find records vs. records statistics....for instance a 3-4 team vs. a 5-2 team...which record wins how often? but for every record matchup in every week."

That's a cool concept, and one that I could answer historically with a query when I get the time. But in the meantime, here's what I believe is a valid way to estimate that probability...

Add eleven games of .500 ball to the team's current record (at any point in the season). So if a team is 3-4, their "true" wpct talent is (3 + 5.5) / (7 + 11) = .472. If their opponent is 5-2, it would be (5 + 5.5) / (7 + 11) = .583. Use the following equation to estimate the probability of Team A beating Team B at a neutral site:

p(Team A Win) = Team A true_win% *(1 - Team B true_win%)/(Team A true_win% * (1 - Team B true_win%) + (1 - Team A true_win%) * Team B...

0 0

How do you handicap an NFL team's full season in one bet?

Team by team 2017 NFL season win total odds futures are posted above.

Current NFL Season Win Totals 2017 -- the NFL football season win total futures line can be viewed as the Las Vegas proposition to the betting public of how many wins a given team might achieve during the regular season. Wagers can be placed on whether a team will win more than "over" or less than "under" the posted total of wins for the 16 game regular season.

Some bettors view the Las Vegas regular season NFL win total odds over/under line for a team as the number of wins that attracts even betting action on the "over" and the "under" for a given team, and not necessarily how many wins Las Vegas bookmakers believe a team will actually achieve.

Betting on Las Vegas NFL football season win total future odds is an interesting option for those who like to look at the NFL season as a big picture event.


0 0

(I originally posted this at the S-R Blog, but I thought it would be very appropriate here as well.)

WARNING: Math post.

PFR user Brad emailed over the weekend with an interesting question:

“Wondering if you’ve ever tracked or how it would be possible to find records vs. records statistics….for instance a 3-4 team vs. a 5-2 team…which record wins how often? but for every record matchup in every week.”

That’s a cool concept, and one that I could answer historically with a query when I get the time. But in the meantime, here’s what I believe is a valid way to estimate that probability…

Add eleven games of .500 ball to the team’s current record (at any point in the season). So if a team is 3-4, their “true” wpct talent is (3 + 5.5) / (7 + 11) = .472. If their opponent is 5-2, it would be (5 + 5.5) / (7 + 11) = .583. Use the following equation to estimate the probability of Team A beating Team B at a neutral site:

p(Team A Win) = Team A true_win%...

0 0

We all know that Dallas and New England are currently the favorites to meet in Super Bowl LI. But being a Super Bowl favorite isn't just about winning games. It's also about being ahead of your competition.

New England is the Super Bowl favorite right now, even though the Patriots have the same record as Oakland and are one game behind Dallas. That's because both Oakland and Dallas face more competition than the Patriots do.

For Oakland, the competition is for a division title. The Raiders have only a one-game lead on the Chiefs, and they have to play in Kansas City in two weeks. So while the Raiders may be tied for the best record in the AFC, they also have only a 59 percent chance of winning their own division. Making it to the Super Bowl will be a lot tougher as a wild card. Oakland, Kansas City and Denver all have much worse Super Bowl odds than New England does because only one of them can win the AFC West.

For Dallas, the competition is for the conference...

0 0