Statistics and the NFL Point Spread

RSS Feed Homepage

Articles: 1, 2

New Blog Location
2008-05-09 17:00:00
This website has moved to: Statistics beats the NFL point spread
More About: General , Blog , Location

Thank you for a great NFL 2007 Season
2008-02-05 20:20:00
I want to thank all my readers for their visits, encouragement, and feedback in this 2007 NFL season. In only 3 years of analyzing NFL spread data, I have made great progress in improving a systematized strategy to detect games where there is an EDGE and an opportunity for gambling. This progress has been achieved by back-testing different strategies, reading scientific papers on the NFL spread market, using readers' feedback, and of course hard work. In the past 2 years, I have felt comfortable providing my statistical results (free of charge) to you. This year not only did we improve our ATS percentage to 61% and created better computerized strategies, but the reporting and stats have improved significantly. We were able to detect trends in momentum/yardage and head-to-head spread history. Next year I want to: 1) improve the stat reports/graphs even more to provide you with even more useful information, 2) improve predictions and detection of spread inefficiency even more (for ...
More About: General , Great , Season

Super Bowl 42 NFL Spread Pick
2008-01-29 00:35:00
The Super Bowl between the New York Giants and the perfect New England Patriots will be on Feb. 3, 2008 at Arizona. The spread stands at 11 on some websites and up to 12 on others favoring the Patriots. It is a hard bet to place because New England is unbeatable, but in the past 8 games they have only covered the spread once averaging 7 points under the spread. The Giants are 10-0 on the road and they are 7-1 ATS in the past 8 games averaging 6.6 points above the spread. So clearly, the spread has been underrating the Giants offense and overestimating the Patriots' strength. What does that tell us? Not much, yet. One could argue that, for this game, the spread has adjusted for these biases (leading to a Patriots pick) or that this trend will continue (leading to a Giants pick). Below is a graph of these spread trends for both teams: Giants and Patriots Spread history To avoid my emotions and trend biases from hurting my Super Bowl pick, I consulted my statistical model as I alway...
More About: Pick

Super Bowl Squares
2008-01-26 04:06:00
Super Bowl Squares is a well-known party game many of us play with friends and co-workers during the Big Game. Here's a quick explanation: A 10-by-10 matrix (=100 squares) is made with one team on the top and the other is on the side. Each square on the grid is sold. Put the sold tickets in a hat, randomly select a ticket and fill up the squares in order. At the end of the quarter, if the last two digits of the score corresponds to one of your squares you win (usually 1/5 of the pool for a quarter and 2/5 for the final score). A Variation of the Betting Squares Super Bowl Game My friends and I like to introduce a bit of skill into this game so that is not completely determined by randomness. We all get together to watch the game. Our names are written in a piece of paper and selected randomly from a hat. One by one the draft order is selected. We then pass around a piece of paper in the order the names were selected (like a fantasy football draft) with the 10x10 grid to pick a non...
More About: Super Bowl , General , Super

An Edge in NFL Odds
2008-01-22 05:44:00
What makes a gambler bet and go to casinos? Many bettors believe it's fun and worth losing the money, but most believe they have an edge over the house and believe they can profit with their predefined strategies. Gamblers that bet on the roulette for example, might study the roulette and figure out that some numbers are called more than others thinking the wheel is not leveled. Blackjack player who count cards can spot when the deck is in their favor and make larger bets at this point and smaller bets when it is against them. Whichever the game, a good gambler sticks to a proven strategy and does not let his/her emotions affect decision making. Human Emotions are Your Worst Enemy Before I started putting real money on NFL games, I spent two years gathering data, analyzing different systematized strategies, and making gambling decisions in real time, but not with real money. It was easy sticking with what the computer said since my emotions and biases were not playing a role. Aft...
More About: General , Edge , Odds

Conference Championship
2008-01-19 01:01:00
Perfect Record in the 2007 playoffs so far... Looking back at my posts I had: Wild Card: SEA (-3) over WAS Divisional: GB (-8) over SEA and SD (+10.5) over IND THIS WEEK'S SPREAD PICK: Conference : 1 pick this week New York Giants @ Green Bay Packers This Sunday, the favorite (by -7) young guys at Green Bay face a hopeful Giants team for the NFC Champ ionship . The Vegas line posted on Monday has remained steady at -7, although some sites have raised the odds from 110 to up to 130. Many bettors (about 65% according to covers.com) are favoring the Packers and some sites predict that the line will move half a point before Sunday. What will the Giants have to do to beat the Packers? Stop Ryan Grant, pressure Favre all day, and get a great game from all its key players including: Manning, Jacobs, and Burress. The weather will be close to 5 degrees fahrenheit which might favor the Giants to cover, still the odds of keeping the Packers withing 7 points are small. In cold it is easier to ...
More About: General

San Diego @ Indianapolis Divisional Playoff Pick
2008-01-13 15:49:00
The spread has shifted 2.5 points for the Colts since I wrote the predictions and picks for this week on my previous post. Bodog has the Colts winning by -10.5. I ran my model again to see if the increase in the point spread would increase the percentage of wins significantly. It did increase the percentage from 56% to 58%, that is, when the home team is favorite by more than 9 points and my prediction is lower than the spread by 4-6 points is have gotten a 58% ATS. Now there is a catch, injuries are not taken into account. Since I only take into account team statistics, if a player is injured, there is currently no way to account for that. I argue that if a bench player or a player not selected for the Pro Bowl is injured, the pick is still valid. On the other hand, if a player like Antonio Gates (leader in receiving yards for San Diego ) is injured the model should be ignored. So, although I like San Diego covering a spread of +10.5, if Gates does not play which is still a game-t...
More About: General , Indianapolis , Pick

NFL Divisional Playoffs Picks
2008-01-11 18:15:00
Quick Picks: Green Bay -8 This week the statistical model is identifying one game with a great opportunity to bet on. Although the prediction favors Green Bay by 9 (only one point more of the Vegas spread), in 37 games dating back to 2000, this prediction favors the home team 70% of the time. That is, when the spread is between 6 and 8 points for the home team and the prediction favors the home team slightly (less than 2 points), betting for the home team has produced a 70% success rate. In this case, we are identifying games where the Vegas spread is under-rating the home team when they should be more heavily favorite. Favre's last chance at a Super Bowl ring, at a noisy Lambeu Field, no key injuries, I like this bet. The recent shaky performances by the Packers are keeping the spread at 8 and gamblers are split 50-50. One other possible opportunity I have highlighted in yellow is the Indianapolis/San Diego game. Both team have playoff experience and a spread of 8 seems rather h...
More About: General , Playoffs

NFL Wild Card Spread Picks
2008-01-04 16:27:00
Quick NFL Picks: Seattle -3.5 Last week the statistical predictions only included one game and one winner! KC came back in the 4th quarter to tie it and lose the game but to cover the spread of 6 against the Jets. This week, of the 4 NFL wild card games, only 1 seems like a good bet and I personally like it because the spread has moved 1.5 points in the other direction. The game is Washington @ Seattle which bettors favor Seattle by 3.5 but originally it was a 5 point spread. The model has a solid 61% success rate of games with the spread between 2-4 for the home team and the estimate being within 2 points of the Vegas spread. Seattle if predicted to win by 5. Journalists are favoring Washington because they have "momentum", something that has been scientifically proven to be a fallacy. It is just random that a basketball player makes 50 free throws in a row or a team wins 4 games in a row. The relevant statistic is that the player is a 90% free throw shooter or a team has a 60%...
More About: Card , Wild , Wild Card , Spread

Fooled by Randomness
2007-12-26 20:28:00
Quick Pick: Kansas City +6 - No comment Currently, I am reading a very interesting book called Fooled by Randomness written by Nassim Taleb. The book is about "how we perceive and deal with luck in life and business". It contains a story about two people, one who got lucky in the stock market and 'explained' why his move was a 'smart' one after the fact. The other character who recognized randomness in the market, played it safe, and although never made huge amounts of money, never lost it all as the first character did once randomness kicked in. One got lucky period and the other had some knowledge, played it safe, and was well in the long run. The same can be said with NFL gambling. Say I create a methodology that is tested for 10 years to be 65% successful (which I have). I can then create a story behind the methodology explaining why it is a good strategy (something like the spread is overreacting to recent games or giving too many points to home favorites) to back up thi...
More About: General

NFL Point Spread Picks Week 16
2007-12-22 01:25:00
Point Spread Quick Picks: Tampa Bay -6 - San Francisco is weak and inexperienced Philadelphia +3.5 - Going against ESPN who chose NO to cover Denver +9 - I wouldn't bet on this one, but it is what came out Indianapolis -7 - Indianapolis knows what to do before entering the playoff, gain momentum. Last week the model did not perform well going 2-3 ATS on the 5 yellow picks posted. Overall, the record is now at 58% for yellow picks and 60% for green picks since I started coloring favorite picks on week 9. This week I have built a more robust model I will use to verify the current model's picks. I have tested it and it does as well as the current pick and when both pick the same team, we get a 2% improvement when predicting games from 2000-2007, going to 55% for all games. The purpose of statistically trying to predict games should be to find opportunities and not to try to predict ALL games successfully. I have found situations in which 30 and 40 games have been predicted correc...
More About: Week , Point

NFL Point Spread Picks Week 16
2007-12-20 01:45:00
Quick picks: St. Louis +8 Philadelphia +3.5 Baltimore +10 We had a bad week last week, going for 2-3 ATS, unacceptable for such a smart computer like mine. Now, I am 7-5 with the yellow picks and still 5-3 for green picks. It was partly my fault by making games yellow that shouldn't have been. This week, I am sticking with my coloring rules: above 60% confident GREEN and between 55-60% yellow. The purpose of this blog is to spot opportunities in NFL gambling and not to try to find a model or procedure that does well for all games. Therefore I am adding a new rule this week, I will color only at most 4 games (if there are 4 games that are worth coloring). This week there are 3 NFL games which the computer has seen as good opportunities for betting. Two visiting teams are predicted to cover against home favorites and a home underdog is expected to cover the spread. I could see St. Louis preventing the Steelers from winning by more than a TD and Philadelphia even beating New Orleans...
More About: Week , Point , Spread

NFL Computer Picks Week 15
2007-12-12 01:55:00
Quick picks: Buffalo +6 Swimming against the current, it's not easy Houston +1 Home underdog, same record, I like it. NY Giants -4.5 My computer doesn't even know that Jason Campbell is not playing! Indianapolis -11 I don't like hot streaks, but boy did they look good on Sunday! Jacksonville +4 Tough pick, but I think JAC wins the game. Last week the TOP 7 picks were 5-2, check out the previous blog! I started categorizing games into "green" and "yellow" since week 9. The purpose was to illustrate the success rate of the model for certain circumstances in which the game at hand fell into. Green implying a significantly good success rate, currently defined above 60% , and yellow an OK success rate currently defined at 55%-60%. Since then, the yellow picks have had a 8-4 ATS record and the green picks a 5-3 ATS. Overall, that is 13-7, 65%, huge right? Well, I am not convinced since it is only 5 weeks and anyone can get lucky. This week, there wasn't a g...
More About: Computer , Week

NFL Automized Picks Week 14
2007-12-05 01:46:00
Quick NFL picks: Cleveland -3.5 Arizona +7 Philadelphia -3 Washington -3 Another week with huge and unexpected SUCCESS. Carolina and Pittsburgh were both picked last week with 'yellow' confidence. These "yellow picks" have covered the spread 4-0 in the past 2 weeks. Also the highest green pick, last week was San Diego, also covered for the 4th week in a row! Still, the model is not and probably will never be perfect and disappointed me with Cleveland's loss against Arizona. This week I have some good picks for you as well. For those new readers, here' a quick explanation of the table below. The Estimate column is an estimate produced via a regression model of team rankings, off/def yards, and home field advantage. The team of record is the home team, which means negative favors the home team. The success of these picks is cross-tabbed by 5 categories of the Vegas spread and 5 categories of the difference between the prediction and the Vegas spread. The Total Games column rep...
More About: Week

NFL Point Spread Standings
2007-11-29 15:29:00
Here are the standings of NFL teams against the point spread up to week 13. Standings NFL Point Spread

NFL Week 13 Picks
2007-11-27 23:08:00
Quick Picks: Cleveland +1 San Diego -5 Pittsburgh -8.5 Last week was a HUGE SUCCESS! Overall, the statistical model had a 67% success rate and although my purpose is to select games with higher chance of success, it is still good to know that the model performed well overall. Specifically, the games that the model selected to have high confidence also performed well. Both yellow picks (mid-high confidence), last week's were Green Bay and Indianapolis, covered as predicted. Last week's highest confidence pick (New Orleans) covered by a wide margin. The only disappointment was the OAK @ KC game which (lesson learned) should have not been predicted because of injuries. The technique I use weighs recent games heavier, but games dating back to last season are still considered. During that time, KC ran with Larry Johnson and Oakland did not use Culpepper. Since there is currently no way for me to account for injuries, from now on when a team has a key player missing, I will not inclu...
More About: Week

NFL Point Spread Computer Picks Week 12
2007-11-21 18:50:00
Quick Picks: New Orleans -3 Kansas City -6 Indianapolis -11.5 Motivated by one of our readers, I have decided to run the current model and show you two GREEN picks and two YELLOW picks. At the top, with 71% confidence we have New Orleans covering a spread of 3 at Carolina. Personally I don't like this pick, but I have to disregard my thoughts and feelings since they might be biased by reading other people's opinions and blogs. Secondly, we got Kansas City covering a spread of 6 at home against Oakland. I do like this one. NFL Week 12 Picks The computer also shows some other attractive games which are worth mentioning. The model is predicting Indianapolis over Atlanta by 18 points which would comfortably cover a 11.5 spread. The line has shifted to 12.5 in some sites which is still a bargain since it is below two TD's. One curious game that I have not highlighted in yellow because it goes against most people's opinion (which might actually be good) is Houston defeating Clevel...
More About: Computer , Point , Spread

No picks this week, working on a new model
2007-11-17 22:05:00
Currently I am working on different kids of models and statistical techniques in order to obtain one that is satisfactory. So far, the best model gives 52% winning rate on ALL games from 2000-2007. This model would make a bettor break-even, unsatisfactory. We worked on finding circumstances where the model provides more accuracy and we had found that indeed there are situations where the model predicts with more than 65% accuracy, but these situations were hardly the norm. I have come to the conclusion that a simple linear regression model will not work with the data I have. Other similar techniques I have tried include robust linear regression, where outliers are down-weighted weighted in order to obtain better estimates and logistic regression which allows the responses to be binary in order to predict the probability (odds) that a team would cover the spread. A few of the problems I see are that 1) the data is not linear, 2) the observations are not independent, and 3) there a...
More About: General , Model , Week , Working , Workin

NFL Point Spread Computer Picks Week 10
2007-11-07 04:11:00
Quick NFL Picks: Dallas -1.5 Buffalo -3 My computer admits last week was brutal, she told me. But she did tell me that her most confident game was right on the nose. Houston indeed covered the spread. Brutality came with the other 2 medium confidence games. Although the Cleveland/Seattle game came down to the wire, there is no excuse. The question is, should we just pay attention to the 'green' games? Well, I wouldn't go that far, just yet. The good news is that in this NFL week 10, the model produced 2 high ('green') confident games. Is it Minnesota/Green Bay, Indianapolis/San Diego, or Chicago/Oakland. If I were to pick with my pure gut feeling, these are the games I would choose and would pick all visiting teams in those 3 games. The green games of the week are: Dallas @ NY Giants and Buffalo @ Miami. We will look at more detailed stats for these two games in the next post, but for this one, we'll stick to the picks and predictions. Below you will find the table with the...
More About: Computer , Week , Point , Spread

NFL Picks and their Measure of Confidence
2007-11-06 18:09:00
Lately, I have been highlighting games in green that have 'high' degree of confidence (las, but what is this measure of confidence? It is simple. Statisticians use a wide variate of deviance measures to understand the accuracy of models and procedures. For regression models (which is the current model being used in this blog) some examples include R-Squared (R^2) and Mean Square Error (MSE). The former measures the relative proportion of the variance explained by the model and the latter the expected value of the error (how far apart could I be from actual the point spread). These measures are used to select the model to use, i.e. the one with highest R^2 and/or lowest MSE. In the NFL, it is not so important to be accurate on predicting the point spread (Vegas does a pretty good job), it is most important to select winning picks. Therefore, this problem turns into a classification procedure that decides which team is more likely to cover the spread. The decision 'confidence' ca...
More About: General , Confidence

NFL Week 9 Computer Generated Picks
2007-11-03 22:20:00
Quick picks: Seattle +1 Houston +3 Baltimore +9.5 I ran my model for this week, compared its success to this year and previous seasons, and automatically created picks accordingly. The system is based on which situations is the model predicting correctly. For example, when the Vegas spread is favoring the home team by 3 to 6 points but my model predicts the visiting team to win by more than 6, the visiting team ends up covering the spread 75% of the time. This week we have one such situation, Houston at Oakland. If you have been reading my blog, you will notice that I added a new dimension to measuring the effectiveness of the model. Before, I was defining 'situations' as intervals of points of the difference between my spread and the Vegas line. Now, I consider not just this difference, but what was the value of the Vegas line at such difference. You can visualize it as a 3-way cross-tab. I found some situations with 90% from 2003-2007, although sample sizes were as big as 10....
More About: General , Computer , Week

Clash of the Titans: New England vs. Indianapolis
2007-11-03 02:47:00
What a game! Is it Sunday yet? Indianapolis is undefeated and so is New England . Peyton Manning was last year's MVP and Tom Brady is on route to break the record for the most TD in one season. Although the Colts are playing at home, Vegas is giving a 5-point (Bodog has the spread at -6) advantage to the visiting team. In this post we'll explore some stats and predictions to see if this spread is justifiable. First, we start with the history when these two teams have faced each other. The graph below shows all the games since 2001 when these teams have battled it out. At the far right we see last year's AFC conference championship. The positive green bar shows that Indianapolis was favorite by 3 games and as we all know they went to beat the Patriots by a score of 38-34 and covering the spread by 1 point (hence the small positive red bar). In 2006, New England was favorite by four points but again the Colts covered and as you can see from the red bar, by 10 points. It wasn't un...
More About: Titans , Clash of the Titans , Clash

NFL Week 8 Computer Generated Picks
2007-10-26 01:40:00
Quick picks: Tennessee -7.5 Buffalo +3 Green Bay +3 Jacksonville +4 Chicago -5 I am running a regression model to try to predict the point spread for each game. Since the model only takes into account each team's performance for this year and the past two, I understand there are many other factors that can affect the prediction. Therefore, I run the model for every game, calculate the difference between the prediction and the Vegas spread, and then compute the percentage of times the model correctly picks games. When the prediction-Vegas column is negative, it means the model picks the home team and when it is positive the model picks the visiting team. In the table below I show you how I figure out which games are best to pick. For example, when the difference between the prediction and the Vegas spread is between -5 and -3 points (home team preferred), picking the home team has been correct 9 of 12 times (75%). One could argue that when the difference is between -1 and 0, picki...
More About: Computer , Week

Only 1 Undefeated Team as of Week 7
2007-10-25 00:33:00
As of week 7, there is only one team that has covered the spread all 7 times. Yes of course Tom Brady and the gang. New England faces a tougher challenge this week when it faces Washington (win record 4-2) while still preserving the 16 point spread it had last week against Miami's winless team. It is interesting to note that against the spread, Washington has a 50% record with two ties while Miami has a 1-4. I am not predicting New England to lose the spread, all I am posting is the winning records against the point spread as of week 7. Overrated teams: Denver, New Orleans, Baltimore, and St Louis Underrated teams: New England, Green Bay, Tennessee, Arizona, Dallas, and NY Giants Stay tuned for my statistical regression model of the NFL point spread. I will be posting how it has performed as of week 7, where opportunities lie, and which games are its best bet. Last week, it captured the covers of Giants, Dallas, and Atlanta while giving one erroneous pick for Tampa Bay (that's...
More About: Team , Week , Undefeated

NFL Week 7 Computer Generated Picks
2007-10-18 23:32:00
In green you will see games where the prediction confidence is at its highest, 65%-70%. In orange, the confidence is at the second highest level 55%-65%. Notice that my Vegas Line and point spread predictions take the home team as reference. That is, when the Vegas line is negative, then the home team is favorite. For example, the Arizona @ Washington game has Washington favorite by 8 points and the computer generated point spread prediction is 2.1 points more than the Vegas spread which falls into this year's low confidence category (see previous posts) and hence the row is kept white. Other games where the difference is too high, confidence is low since other factors not included in the model may be affecting the estimate. This week we have 2 games with high confidence and 2 games in 'medium' confidence. The San Francisco @ NY Giants, the estimate is favoring the home team by more than 4 points. The Tampa Bay @ Detroit game, the prediction is giving the visiting team a slight...
More About: General , Computer , Week

NFL Week 6 Regression Model Picks
2007-10-14 22:35:00
I am a bit late with these picks, Sundays are hard to wake up early and crunch numbers. It is currently 12:30pm and some games have started, so some of these picks are already too late to use. I have not looked at the scores although my picks these picks are automatic and consider only stats from the past few years up to last week. In the previous post I talked about a regression model that estimates the point spread, compares it to the Vegas line, and makes a pick accordingly. I took the difference between the automatic pick and the Vegas line and noticed that depending on this difference, the model behaves in an opportunistic way. Sometimes it is lagging behind (weighing previous years data too heavily) and doing the opposite is opportunistic, other times it is predicting well (especially in favor of underestimated visiting teams) and other times (i.e. when the difference between Vegas line and prediction are close) it predicts correctly close to 50% of the time not allowing for...
More About: General , Model , Week , Regression

NFL Computer Picks Generator v1.2
2007-10-13 19:14:00
In this post, I will present the difficulties with NFL data on tabulating how a computer model would have made picks in past games, I will describe the first simple computer model for this year, show the results of how it performed in the past 5 years/current year, AND in the next post what it picks for the coming NFL 2007 week 6. Introduction This website was originally intended to test statistical models that automatically select NFL picks, but have yet provided their results this year. Finally, the first one is here with 'out of sample' results on how it performed over the last 5 years. Out of sample means (in the context of NFL data) that the estimated model for a week does not use data of that week nor future ones, i.e. the way it works in real life NFL gambling. Researchers build computer models guaranteeing certain percentages of success, but the games they are 'predicting' are already included in the sample data. Data mining techniques use this out of sample idea to tr...
More About: General , Computer , Generator

NFL Spread Computer Forecasting Models
2007-10-08 14:11:00
In the past weeks I have been posting informative graphs and NFL statistics of matchups. Then, I started to make personal predictions of who would cover the spread but the only information I was using was the exploratory analysis of the historical data of the NFL point spread that I have collected. As you probably have seen, I have not been quite successful. There are way too many 'experts' and sites that continually claim winnings against the spread. That was not the original purpose of this blog. As a statistician I know I should not be doing this and instead I should be looking for patterns in the data and running models/algorithms to detect them. For example, when a visiting team's offense/defense strength is at a certain level and the opposing team's at another level, the point spread should be X. If that X is significantly different from the vegas line, then an opportunity can be explored. Another example is to give each team a time varying ranking so that the difference ...
More About: Models , General , Computer , Forecasting , Spread

NFL Spread Cover Percentages as of Week 4
2007-10-07 08:01:00
Here are the winning percentages for each team against the spread. We have 4 NFL teams which have covered the spread in all games as of week 4. There are 6 teams that have yet to cover the spread this year including my struggling Bears who face the spread threatening Green Bay Packers. Hope this info is helpful. NFL Point Spread Winning Percentages
More About: General , Cover , Week , Ages

Seattle @ Pittsburgh NFL Week 5
2007-10-06 02:53:00
Interesting game with a point spread of Steelers at -5.5. Let's dig into 2 graphs I think are important and understand why the line is where it is. First, the offensive strength. NFL Week 5 Pittsburg h One stat really sticks out and that is Pittsburgh's rushing yards in the past 5 games has been on average 60 yards better than their opponents. Also notice how although Seattle does not have anything too high, it has remained positive in yardage against its opponents. So where does this lead us? Into who are their opponents. For Pittsburgh it's Arizona, San Francisco, Buffalo, and Cleveland. If I had to build an index on schedule strength I would have to say it is weak. Seattle has faced San Francisco, Cincinnati, Arizona, and Tampa Bay not such a strong schedule either so I conclude strength schedule is not a factor. Now, let's look at their previous NFL results against the spread:NFL Point Spread Week 5 Pittsburgh covered the spread by more than 10 points in the first 3 games ...