How to Bet NFL Futures

How to Bet NFL Futures

by Dan Gordon

[From Blackjack Forum Vol. XXI #2, Summer 2001]
© 2001 Blackjack Forum

[Dan Gordon is the author of Beat the Sports Books. Portions of this article are excerpted from Danís book, but most of the nuts and bolts of his NFL futures handicapping system are available only in this issue of Blackjack Forum. ó Arnold Snyder]

As Labor Day approaches, one of the biggest businesses in the world prepares to restart. Itís a business that generates well into 11 figures over a five-month period and involves the participation of millions of Americans and citizens from other countriesóand those numbers grow every year. This business is wagering on National Football League games.

Pro football is the biggest betting sport in the United States, with total wagers on it now running more than $1 billion a week. According to USA Today over half of the American adult population had a financial stake of some sort in the outcome of last yearís Super Bowl. That meant over $5 billion wagered on that game alone.

Because of extensive media coverage of the point spread, increased access to betting through the Internet, and the fabulous returns promised by sports touts, millions of fans now believe they have the expertise to beat the sports books, as if the people who book NFL games were the biggest suckers in the world.

Nothing could be further from the truth. Bookies love the NFL season. It provides them with their greatest profits in sports. Many local bookies operate only during the NFL season because the profit from these few months enables them to live well the rest of the year. Bookmakers represent the vast majority of people who make a profit from wagering on the NFL.

The late Bob Martin, manager of Las Vegasís first casino sports book, once told me that the number of bettors who win betting pro football is so small that “Öit is virtually the same as if no one won.”

Year after year, NFL bettors go into the season filled with confidenceóand end up losing. How can you avoid being one of them?

This is the question I answer, in part, in this article. But first, for beginners to sportsbetting, letís go over a few basics.

How Pro Football Is Bet and How the Line Is Set

Before you can win betting pro football you have to know how the line is set.

Many years ago all sports, including football, were bet according to odds. If someone liked a favorite in a game, he could give, say, 2-to-1 odds on the game (meaning he would have to risk $2.00 for every $1.00 he hoped to win). On the same game, a bettor could take 8-to-5 odds if he liked the underdog (risking $5.00 for every $8.00 he hoped to winóor roughly 63¢ risked for every $1.00 hoped for), thus making the “true” oddsóor how the betting public perceived the gameó9-to-5 for the favorite. (The odds of 9-to-5 are the midway point between 8-to-5 and 2-to-1, which can also be expressed as 10-to-5.) The difference between the give on the favorite (2-to-1, or 10-to-5) and the true odds (9-to-5) was the edge for the bookie on the favorite; the difference between the take on the underdog (8-to-5) and the true odds (9-to-5) was the bookmakerís edge on the underdog.

A problem with this type of betting on football came in mismatches, where a bettor might be asked to give 5-to-1 or 6-to-1óor moreóon the favorite. Few bettors wanted to lay such high odds because they had to risk so much to win so little. Neither would many bettors bet the underdog when it was perceived as having so small a chance of winning. Getting high odds was no consolation if your team seemed certain to lose. Thus, mismatches drew very little betting action (and very little “juice”óor profitófor bookmakers.) This was a situation odds makers could never feel happy about.

This problem was solved with the advent of the point spread (also called the “spread”). As far back as the 1920ís, point spreads were used on a limited basis. By the late 1940ís, the point spread had become the accepted way to bet on individual games.

The point spread is a handicap given to the favorite in terms of points. In the last regular season game of 1996, for example, the New England Patriots closed as 9 1/2-point favorites over the New York Giants. In this game, the odds makers believed that the Patriotsó-in the eyes of the betting publicówere 9 1/2 points better than the Giants. This meant that the Patriots had to win the game by more than 9 1/2 points (ten points or more) for a Patriots bettor to collect. An outright Patriots loss or a Patriots win of less than 9 1/2 points meant a loss for a Patriots point spread bettor.

If you bet the underdog Giants, on the other hand, you received 9 1/2 points on the Giants on your bet. If the Giants won the game outright or lost by less than 9 1/2 points (nine points or less), you won your bet. If the game landed exactly on the point spread (which obviously canít happen with a line of 9 1/2), the game is considered a “push” (or tie) and neither the bettor nor the bookmaker makes a profit.

Thus, when you bet a favorite, you are giving (or laying) points; when you bet an underdog, you are taking the points. In the above example a way of writing up a Patriots bet would be: “Patriots -9 1/2.” For someone betting the Giants, the bet would be: “Giants +9 1/2.”

The reason the point spread is considered better than odds is that, in theory, a handicap makes a game an equal proposition in the eyes of the betting public, with bettors just as likely to take one team as the other no matter how great the mismatch might be. In mid-season of 1996, the lowly 1-6 Bucs went to play the 6-1 Packers (the team that would eventually win the Super Bowl). In the old days, the Packers would have been prohibitive odds favorites and the game would have received few wagers. But with the point spread, the line eventually settled at Packers -17, making the game just another of the 14 games played that weekend that got its share of betting action.

The point spread method also makes it easy for sports books to adjust the “price” (or handicap) on a game as money comes in on one side or the other. Letís return to the Patriots-Giants game discussed above.

Las Vegas Sports Consultants (founded by Michael “Roxy” Roxborough, who is now retired) in Las Vegas, the largest odds-making office in the world and the one with the most clients (sports books that use their numbers), felt that making the Patriots an 8Ω-point favorite would make the game an equal proposition to the nationís bettors. How- ever, as the week went on, more money came in on the Patriots than on the Giants. Sports books in Nevada and bookmakers around the country were quick to raise the price on the Patriots, moving their handicap (the “number” or “line”) to -9 and then -9 1/2. When wagers continued to come in on the Patriots despite the higher price, some books were even forced to move the number to -10. Finally, Giants money came in and most bookmakers closed the game at -9 1/2.

As you can see, betting on pro football is truly a free market activity. Most bettors who wagered on this game considered the Patriots a bargain at -8 1/2 and -9. Their betting pushed the price on the Patriots up. When the line reached -10, bettors at last considered the Giants a good deal and their wagers pushed the line back down. The closing number of Patriots -9 1/2 was the result of pure supply-and- demand.

And this is where opportunity arises for the professional sports bettor.

The reason you can make money betting the NFL is precisely because the lines put out by the odds makers are made not to predict the actual outcomes of games, nor to educate the public about the relative strengths of the teams, but to try to split the betting public by making one team as attractive as the other. And the publicís view of a match-up is occasionally incorrect.

A professional bettor looks for lines that are inaccurate in terms of the real differences between the two teams. When he finds such lines he wagers on themóand that is the only time he wagers.

While good odds makers have to tie their lines to public sentiment, the professional maintains a distance from such sentiment in order to be able to recognize those games where public sentiment is incorrect.

Before going on to specifics about how you recognize good bets, I want to mention three points about how I view the NFL.

The league, like the rest of pro sports, has become one of big contracts and big money. Itís my impression that this causes players and even coaches to become, at times, mentally fat. If a game or two has been played well and won, details might not seem that important for the next game. Letdowns happen more now than in the days when players were less well-paid and when todayís job security and guaranteed money did not exist. This doesnít mean that the game was better in the good old days. It just means that it was different to handicap. Teams tended to play more true to form in those days.

What big money and big contracts also mean now is that when teams perform beyond their capabilities for a game or two, one is more likely to see a bounce-back than in years past. For a handicapper this means that it is important to track and be ready to pounce on situations where such bounce-backs are likely to occur.

Second, there is tremendous parity in the league. That means, despite public perception and constant media star-making, that there is not much difference between the leagueís best player at a position and the worst. When a team of slightly worse players is more motivated than a team of slightly better players an outright upset is possible. Most certainly, itís possible for the “inferior” team to cover the point spread.

Third, the point spread tends to nullify any obvious scrimmage edge (skill or power advantage) a team has over its opponent. In the 1995 and 1996 seasons, for example, there were 166 games in which the point spread was seven or more points. These were games where one team was perceived to have a big edge over its opponent at the line of scrimmage. While the underdog won just 39 of these 166 games outrightó23.4 percentóthe underdog covered the point spread in 87 of the games (while tying it in three): a success rate of just over 53 percent. In 1997, there were 69 games with a spread of seven points or more. The underdog won just 12 of these games outright (17.4 percent), but had a 36-31-2 record against the spread for a success rate of 53.7 percent.

Also, as I shall demonstrate shortly, the difference between teams in the middle of the league is not as great as most bettors think. The statistical difference, for example, between a 10-and-6 team and an 8-and-8 team is only three points on a neutral field.

In summary, the team that most bettors think is the better team may in fact be better. However, the point spread usually more than nullifies that edge. Thus, bettors who bet favorites are often betting poor value.

On the other hand, sometimes a team will perform very well or very poorly for just a game or two, and these short-term performances will give the public a false reading of the team. The odds makers mirror this public reading with their numbers. At this time anyone betting on an overrated team or against an underrated team is again betting poor value.

How can you tell if a team is overrated or underrated?

Power Ratings

In their strictest sense, power ratings are values given to each club that tell where that club stands in skill and strength in relation to every other club. The difference between one team and another in my power ratings tells me what should happen in a game between the two clubs on a neutral field with everyone healthy.

Of course, teams donít play on neutral fields in the NFL, but weíll deal with that later. In the meantime, when the rest of the betting public is blown astray by what happened in the last game, power ratings are my guide to the real strengths or weaknesses of a team.

Some handicappers rely mainly on objective factorsóscores and other statisticsóto set their power ratings. Other handicappers rely mainly on subjective factorsótheir own feeling about a team based on certain types of team-membersí behavior.

Through my experience in over twenty years of handicapping, I have come to use bothótwo sets of power ratings. I have devised these two ratings to reflect both my judgment about teamsí strengths, and their performance as measured by scores.

he first ratingóthe only one we will discuss in this articleóis a letter power rating in which each team is ranked someplace between A+ (the best possible rating) and E- (the worst)ósee Chart #1.

Chart #1: Power Ratings

These letter rankings correspond to how I feel a team is playing presently, and are meant to reflect how a teamís record would look after a 16-game season playing at this level, assuming that their competition was average. For example, if a team is playing like an 8-8 team, I give them a ranking of Córight in the middle of my range and in the middle of the league. At the very top (A+) of my ratings would be both 15-and-1 and 16-and-0 kinds of teams, while at the very bottom (E-) would be any 1-and-15 or 0-and- 16 teams. Only tthe 1985 Chicago Bears ever made it to A+ in my ratings. No team has ever sunk to E-.

Because they reflect a teamís current performance, letter power ratings are updated continuously throughout the season. And, in my letter power ratings, teams donít always end up ranked according to their actual record. This is because their actual wins and losses may involve an unusually easy or tough schedule, or unusually good or bad luck. Instead, I look closely at the games teams have played to put a “team picture” together.

The letter power rating is to some extent a subjective one, and will only be as good as the experience and judgment of the rater. One way to check your power ratings, as you acquire experience and judgment, is to compare your ratings to those of other respected handicappers. (The Gold Sheet and Power Sweep are among the reputable publications with power ratings in them. You may obtain the current issue of either of these publications from GBC in Las Vegas, or you can subscribe to them directly. The Gold Sheet also has a website at www. goldsheet.com, where you can download a sample back issue of their publication.)

However, if power ratings could be reduced to a simple formula, all good handicappers would be betting the same games. Theyíre not. Different winning handicappersólike different winning card countersócan be looking at different things. Ultimately your ability to find value in a line (and make winning bets!) will depend on your developing your own skills.

Once I have assigned letter power ratings, I use a simple chart to assign corresponding points by which to separate various teams. Iíve arrived at these point assignments through twenty yearsí study of how NFL teams do in pointwise ratio (that is, points scored versus points given up.) The “Average Edge” in Chart #2, below, is the average number of points by which each rank will outscore opponents or be outscored by opponents.

Chart #2: Power Rating Edge

My study shows that when a team goes 8-and-8, it gives up and scores the same number of points on average over a season. When a team goes 9-and-7 it outscores teams by 1 1/2 points per game on average. When a team is 7-and-9, it is outscored by 1Ω points per game. Thus the difference between a C- (7-and-9) kind of team and a C (8-and-8) kind of team is 1Ω points. The jump from C to C+ (a 9-7 team) is also 1Ω points. The jump from C- to C+ (two steps) is three points. This means that if a C+ team plays a C- team, a C+ team will be three points superior on average on a neutral field.

In my letter power ratings, I list the majority of NFL teams someplace between B- and D+. At the end of the 1996 season, for example, just eight of the 30 teams fell outside of this range.

When the ratings move above B- and below D+, note that the equivalent point values get bigger. Teams above the B- class outscore their opponents by more and in larger jumps per class than those that are rated between D+ and B-. Teams below D+ tend to be outscored by more and in larger jumps per class than the teams in the D+ to B- range.

Thus, from Chart #2 you can see that a B+ team will outscore an average (8-and-8) team by eight points a game (which is the same as saying it will outscore all opponents, on average, by eight points a game). Teams that are rated A- will have a 10Ω-point-per-game edge, while “A” teams will have, on average, a 13-point-per-game edge. Meanwhile a D- team will be outscored by eight points per game; an E+ team will be outscored by 10Ω points per game; and an E team will be outscored by 13 points per game.

And, at least in the world of power ratings, if an A+ team plays an E- team, the A+ team should be ranked 34 points better on a neutral field.

However, since most NFL teams are ranked between D+ and B-, with a difference between these two letter rankings of only six points, the teams in this area of my letter power ratings are very close to even. It is in this area, incidentally, that I concentrate my handicapping, because it is here that emotional edges mean more.

Now, let us go on to how you can convert power ratings into winning bets.

Uncovering Profitable Season Over/Under Bets

I have mentioned very briefly how power ratings can be used to find value in a gameís line.

But straight bets on games are only one of a number of types of wagers an NFL bettor can make. In addition to straight bets on one team or the other against the point spread, you can wager on over/unders, also known as “totals,” in which you bet against a number that represents the total points scored in a game, betting on whether more or fewer total points will be scored.

There are also a number of “exotic” bets, including parlays, teasers, and reverses ómost of which are sucker bets most of the time, and all of which are beyond the scope of this article. In this article, we are going to concentrate on season over/under future bets, which are some of the first profit opportunities that will arise this football-betting season.

“Future” wagers in the NFL are basically bets you make on how a team will perform in the upcoming season, including the playoffs and Super Bowl. These wagers can be bet at a number of reliable sports books in Las Vegas, on the Internet, and offshore. In regular season over/under future bets, you bet on how many wins a team will garner. In other future bets, you can wager on particular teams to win their conferences or the Super Bowl.

But how do you make money predicting how a team is going to do over a whole seasonóparticularly when you have to overcome a healthy house edge? This again is where your power ratings come into play.

You begin by assigning each team a letter power rating at the start of the season. You do this either by starting with your power ratings from last year, modified for changes in team personnel, or, if you are a beginner, by consulting the power ratings of a more experienced, reputable source. Chart #3, below, contains my power ratings for the start of the 2000 season. Note that there were no teams on the A+ through B+ levels and none between the E+ and E- levels at the start of the 2000 season.

Chart #3: Preseason NFL Power Ratings, 2000

The next step is to convert your letter power ratings into projected point spreads for each game, using a schedule for the upcoming season, and the Power Rating Point Differences in Chart #2. [Note: You can download a complete 2001 NFL schedule at the NFLís official website: www.nfl.com] In addition, because over the long run in the NFL the home field advantage has been shown to be worth about 2 Ω points, every home club should be given an additional 2 Ω points in your seasonís projected point spreads, unless there is a very compelling reason not to.

After projecting point spreads for each game, I convert the point spreads into decimal numbers (see Chart #4) that represent fractions of games won, according to the percentage of times a team should win a game at a given projected point spread. (This fraction, like the other statistics in this article, is derived from my own research.) These fractions (decimal numbers) can then be added to project each clubís season number of wins.

Chart #4: Spreads as Decimals

After you project win records for each club, you are ready to check the over/under totals the sports books have put up. You need a two-game or more difference between a sports bookís total and your projected number of wins to make a wager.

Letís take an actual wager I made in the 2000 preseason as an example.

As you can see from Chart #3, before the 2000 season began, I gave the Washington Redskins a B- rating, a letter power rating that translates to ten winsóagainst an average schedule. This is what I wrote about them as the pre-season was closing: “Team is under enormous pressure to win from owner Daniel Snyder. However, the main defensive additionsóBruce Smith, Mark Carrier, and Deion Sandersóare well over 30. The offensive line also has two old starters. Last year the Redskins were just 1-5 against other playoff teams. The loss of Cory Raymer (center) will hurt badly. The defense will be better but the Redskins will be overrated early in the season. Look to bet against early in the year.”

However, teams seldom have a totally normal or neutral schedule. To develop a projected win record strong enough to be the basis of a wager, I had to see how the Redskinsí actual season schedule would play out mathematically.

In their first game, the Redskins (B-) hosted Carolina, a D+ team. Since there is a six point difference between B- and D+, I rated the Redskins an 8 Ω-point favorite (remember, they got an additional 2 Ω points for being the home team). The decimal number equivalent for 8 Ω points is .762. I rounded off that number and gave the Redskins .76 for this game.

In their second game, the Redskins would play at the Lions, a C-rated club. That meant the Redskins were a half-point favorite (the three-point difference for the letter power ratings minus the Lionsí 2 Ω-point home edge). The decimal equivalent gave the Redskins .51 for this game.

In their third game, the Redskins would play the Cowboysóa C+ teamó at home. I projected them a four-point favorite and used the decimal .63. (For the complete Redskins schedule, and projected wins for the 2000 season, see Chart #5 below.)

Chart #5: 2000 Redskinsí Season Projection

*Home games. At home, the Redskins get an additional 2 Ω pt. home advantage. On the road, they get a -2 Ω pt. road disadvantage.

In the end, adding up the decimal numbers for all sixteen games of the regular season, I projected 9.48 wins for the Redskins.

Once I had projected 9.48 wins, I was ready to check the sports booksí numbers. Since it happened that several sports books had set their regular season over/under total on the Redskins at 11Ω wins, I had the two-game difference I needed to wager on the under. Moreover, in addition to the two-game edge, I was getting 1.20-1 odds on my bet. When I later saw an over/under total of twelve at another book, I went even further in my bet even though here I had to give 2.20-1 odds, since I now had a 2Ω-game edge on my under wager.

There was one other team that I almost wagered on. Using the above system, I projected the New England Patriots to win 6.35 games. One place in Las Vegas had the over/under on the Patriots at 8 1/2. I had my magic two-game edge. However, to bet the under on the Patriots, a bettor had to give hefty 3.20-1 odds. That, I knew from my research, was too great a price to give. Iíll spare you the calculations here, but as a general rule of thumb, the odds you pay should be no higher than the number of games you have as an edge (donít go over 2-1 for a two-game edge, 3-1 for a three-game edge, etc.). Too bad I passed. The Patriots ended up at 5-11 for the season. Still, you have to stand by your math.

While the Redskins gave me some anxiety early in the season when they started at 6-2, they slumped in the second half of the year. Their owner, Daniel Snyder, put undue pressure on the team as I knew he would if they had any problems. The Redskins lost six of seven and ended the year at 8-8: over a game worse than I had predicted they would finish.

The Odds You Must Overcome in Sports Betting

Now letís talk about the odds you must overcome to be successful in sports betting.

When you bet against the point spread, you are not only giving or taking points. You are also almost always giving 11-to-10 odds to a bookmaker or sports book, meaning that if you want to win $50 on a game, you must risk $55. When a bettor is giving 11-to-10 odds, he is bucking a 4.55 percent edge against him.

To make this clearer, take two bettors who choose the opposite sides of the same game. If each one wants to win $100, each must actually risk $110 (the extra $10 being a fee or commission for the sports book, called the “vigorish” or “vig.”) After the game, one bettor will lose the $110 he risked while the other gets it back plus his $100 win. Thus of $220 total risked $10 goes to the bookie. Ten dollars of $220 is 1/22 or 4.55 percent.

The odds on futures bets are differentóhidden in the price. Iíve already mentioned that on my bet on the Redskins to win under 11Ω games I got 6-5 (equivalent to 1.20-1) odds. Since I couldnít bet enough at this price (some sports books have ridiculously low limits) I went further with this bet against the Redskins betting a small amount on them winning under 12 games. On this part of the bet I gave 11-5 (equivalent to 2.20-1) odds. To compute these odds (or “money line”) equivalents for yourself, simply divide: 6 ˜ 5 = 1.20; 11 ˜ 5 = 2.20.

What were the odds or vig for an “over” bettor? At the book where the over/under was 11Ω wins for the Redskins, a bettor would have had to give 7-5 (1.40-1) odds if he felt that going over 11Ω wins was a good investment. Since the difference between what I got on the Redskins (1.20-1) and what a Redskins bettor would have had to give (1.40-1) is .20, this line is often called a 20-cent line.

The odds makers who set this line were saying through their numbers that the real line on the Redskins winning over 11Ω wins was 1.30-1 in favor of it happening. Letís say these odds were true (they might or might not have been, remember, due to how lines are set). I was getting 1.20-1 odds against something that had a 1.30-1 chance of happening.

Letís do the math on this. Letís say I put up $10 on this bet and play it out 23 times (once for every possible outcome: 13 losses for every 10 wins). If the odds were the true line on this event, I would win 10 of the 23 times. I would lose the wager 13 of the 23 times. All told, I would be risking $230 and my return would be $220 (I would get back $22 each time I won: the $10 I risked and a $12 profitóremember I was getting 1.20-1 odds). Thus, the sports bookís profit would be $10. The vig in their favor would be a little over 4.3% (which you figure by dividing the $10 profit by the $230 invested).

If a bettor wagered on the Redskins winning over 11Ω wins he would risk $14 for every $10 he wanted to win. Walking through this 23 times, our Redskins backer would win 13 of the 23 times and get back a total of $312ó13 x $24 (the $14 risked plus $10 profit) on a total risk of $322 ($14 x 23). Here the sports bookís edge or vig is just 3.1% (their $10 profit divided by the $322 taken in).

Thus, any bettor who wagered on the Redskins to win over 11Ω games had over one percent less vig to overcome than us anti-Redskins bettors.

On the part of my bet where I bet the Redskins under 12 wins, the vig worked a bit differently. That is because bettors who took the Redskins to win over 12 games got just 9-5 (or 1.80-1) odds. Remember that those of us betting under 12 wins were giving 11-5 (or 2.20-1) odds. Since the difference between 1.80 and 2.20 is .40, this is often called the 40-cent line.

To calculate the vig against a bettor on this bet, you do it the same way as in the above example. The mid-way point between 1.80-1 and 2.20-1 is 2-1. Assuming thatís the real line, you play out the wager three times (once for every possible outcome), with the Redskins going under 12 wins twice. On a $10 bet, I would be putting up $66 on three bets (2.20 x $10 = $66)óor $22 a wager.

I would win two of the three bets getting back a total of $64 on my total $66 investment. The sports bookís profit would be just over three percent ($2 divided by $66). For bettors going with the Redskins over 12 wins and taking 1.80-1 odds, the sports booksí vig would be greater. These bettors would put up $10 three times but win only once, getting a return of $28. The bookís profit would be $2 on $30 risked, or 6.7%.

If the lines in sports betting represented the real odds, there would be no point in betting. In other words, if the actual odds of the Redskins winning over 11Ω games last year had been 13-10 in favor and the actual odds of them winning over 12 games had been 2-1 against, I would have been going up against insurmountable odds. Though small odds, they would still be insurmountable and in the long run, I would go broke.

However, since the odds set are not the actual odds but the public perception of them as seen by the odds maker, NFL season over/ unders and other sports bets can be highly profitable.

To give you an idea, a solid NFL handicapper with a 55% win rate has a 5.5% edge on every bet. An expert handicapper with a long-term 57% win rate has a 9.7% edge on every bet. Top handicappers, with a long-term 60% win record, have a 16% edge on every bet. You should not bet until youíre confident that your win rate is at least 54%.

It is the job of the NFL handicapper to find the weak spots in the line. As with card-counting, in sports betting discipline and patience are key. ♠