NFL Interceptions prop bets

Date: Thu, Sep 6, 2007

I found some year-long prop bets in the NFL that I think have edge. The props were on the total interceptions made by each team's defense. I bet the Under on 17 different teams (listed below). The limits are low, but they add up over 17 different bets that represent the same idea.

Taking the shotgun approach, I bet the under on any total of 16.5 or higher. Here's my rationale:

Over the last 4 years (2002-2006), the average number of INTs made was 15.85, with the median at 15.

For the 32 teams over the last 4 seasons (128 total teams), if the total was 16.5, the Under won 77 times and lost 51. For a winning percentage of 60.2% or about -150.

The initial question is: "So what? You need to take into account the good defenses as they are likely to have higher INT totals."

And I did that. I took teams with 18 or more INTs and compared how they did in the next year versus a total of less than 3.5 INTs, but with a minimum of 16.5. For example, if BAL had 20 INTs in 2003, then I would look how BAL did versus a total of 16.5 in 2004.

The total record against a total of 3.5 fewer INTs the following year, but no lower than 16.5, was 7-20.

Next, I ran correlation numbers. Using 2003-2005 INT numbers and correlating it to 2004-2006 INT numbers for the same teams. The correlation was a paltry 0.02. This along may be enough reason to bet Under on high totals.

To top it off, the 17 teams that I bet, which all had totals of 16.5 or greater, had an average number of INTs fewer last year than the total. The average line I bet into was 18.5. The teams that I wagered Under on averaged 17.6 interceptions last year.

Here are the wagers:

ATL U17 -120
BAL U21.5 -115
CAR U18.5 -115
CHI U22.5 -115
CIN U20 -135
DAL U17 -115
DEN U18.5 -120
GB U18 -120
IND U18 -135
JAX U17.5 -115
MIN U20 -115
NE U20 -120
NYJ U18 -120
PHI U17.5 -115
PIT U18 -115
SD U16.5 -115
STL U16.5 -120

Read Full Poker Blog Post

Sidebar from Marc Cooper's article

Date: Sun, Sep 2, 2007

When I first read Marc Cooper's article, I did not notice the additional sidebar.
Here is the sidebar with more quotes from me.

Marc wrote the article for the typical L.A. Times reader, not for professional (or degenerate) gamblers, so the advice is basic but useful for most people.

Read Full Poker Blog Post

Article in the L.A. Times

Date: Fri, Aug 31, 2007

An article in the L.A. Times by Marc Cooper mentions Weighing the Odds in Sports Betting, with a few quotes from me. Marc Cooper is normally a political columnist, but he enjoys Las Vegas and has written a few articles about gambling in Las Vegas.

Read Full Poker Blog Post

NFL Totals and skewness

Date: Sun, Aug 26, 2007

I was looking over NFL totals information today and found something interesting. In reviewing NFL regular season and playoff games from 1989 to 2006, this is what I found:

The average total was 40.23
The average game score was 41.1

Number of Overs: 2183
Number of Unders: 2254

So the average game score was higher than the average betting total, but there were more unders than overs. How is this explained?

The reason for this result is that games can go over the total by a lot more than they can go under. Let's say the total on a game is 40. The lowest possible scoring NFL game is 0-0 (a tie played through overtime). The highest possible scoring NFL game is boundless. So when a game goes over by a lot, it can have a bigger impact on the average than when it goes under a lot. A term for this is skewness. Wikipedia.org describes skewnewss as

• In probability theory and statistics, skewness is a measure of the asymmetry of the probability distribution of a real-valued random variable.

See the link for a much deeper mathematical explanation of skewness in a probability distribution.

The lesson here is that looking at averages for betting purposes can sometimes be misleading. Remember, for straight wagers, we don't care how much we win or lose by, we just care if we win or lose. Winning by a half point is just as profitable as winning by 25 points.

Read Full Poker Blog Post

Small changes to Weighing the Odds in Sports Betting

Date: Tue, Aug 21, 2007

Here is a list of small changes for the next printing of Weighing the Odds in Sports Betting. There aren't any major changes on the list.

Page 17, end of second paragraph.
add: "...and Jeff Brueggman.”

Page 18, end of third paragraph or insert new paragraph after third paragraph
add: “Thanks to Al Rogers of Pi Yee Press for making things run smoothly.”

Page 32, end of third paragraph.
Change: “My research shows:”
To: “Let’s assume the following distribution:”

Page 33, in “Talking Cents”, 4th sentence
There is duplication in the sentence. It should read:
“Sportsbooks understand that a cent is worth less as the line gets farther from +100, so they offer wider spreads as a result.”

Page 37, add to the second paragraph:
“Some people use ROE (return on equity) or ROR (return on risk) to mean the same as ROI. See Chapter 5 for differences between ROI and ROR when applying to scalps and middles.” (new section in Chapter 5 added for page 77)

Page 42, 7th paragraph
Change: “…get lost because the items you want to be are correlated…”
To: “…get lost because the items you want to bet are correlated…”

Page 50, 1st paragraph, 5th sentence
Insert “a field goal” in the middle of this sentence:
“On the other hand, his database may show that a touchdown is the first score more often than a field goal in a typical NFL regular season game.”

Page 57, last line on the page
Delete “like this”
The last sentence should read: “But there are other markets (delete “like this”) where individual entities…”

Page 62, 4th line
Delete “not” in 4th line
Change: “It is the few lines that are not inefficient…”
To: “It is the few lines that are inefficient…”

Page 77, after “Time Spread” and before “Conclusion” add:
Return on Risk
When analyzing the profitability of short-term scalps and middles, it is better to use ROR (return on risk) rather than ROI (return on investment). The risk of a well-executed scalp or middle is small even though a large investment is often necessary. ROI underestimates the profitability of scalps and middles in relation to other type of bets because it is not possible to lose the entire investment.

Page 96, 9th line
Insert “a” before “small window”
It should read: “There is a small window of time…”

Page 108, Table 4
Change “Fair Line” heading to “Rule of Thumb”
In the Key, change: “Fair line” reflects the fair vigorish given the numbers in the Actual column.

To: “Rule of Thumb” reflects the rule of thumb adjustment given the numbers in the Actual column.

Page 109, at the end of the first paragraph and before “My RSW Picks for 2006”, insert this paragraph:
“It is important to keep in mind that the rule of thumb is a shortcut. The rule of thumb becomes less accurate as the money line gets farther away from even money. “

Page 109, in “My RSW Picks for 2006”
Third bullet, it should read: “The expected wins for each team is less than (delete: its) what its expected wins would have been had it played a neutral schedule.”

Page 115, starting with the second paragraph, change the teams in the example from ATL to BAL and NYJ to CHI. ATL and NYJ are used in the previous example and it may confuse some people. So the section should read:

“Another issue to consider is when the accompanying money line on the point spread is not -110. Here is an example.

BAL -3 -120 CHI
CHI +3 +100 BAL

The line is shaded towards BAL -3, and the fair market line is BAL -3 -110. In that case, if you see BAL -2.5 on the parlay card……On the other hand, if the parlay card has CHI +3.5, then ……”

Page 124 in “3-point lead at the half”, second paragraph
Change: “Table 9 shows that the home team by 3 is worth 12 cents plus or minus 2 cents”
To: “Table 9 shows that the value of a half-point when the home team is -3 in the first half is worth 12 cents plus or minus 2 cents”

Page 124 in “Other valuable numbers”, add at the end:
“The push percentages for the 4 and 7 are both significantly higher in the first half than they are for the game (see Table 6 on page 117 to compare).”

Page 126, add to “Caution Using Table 7”
When a team has a playoff spot and position locked up going into the last week of the regular season, they may not play to win as they normally would. Sometimes the first-string players will play only a portion of the game while the back-up players play most of the game. These teams do not have winning as a primary goal, so the relationship between the first half and the game will not be like a typical NFL regular season. The tables in this chapter are not applicable for those games.

Page 135, first paragraph, delete the last two sentences (extraneous).
Delete: “Unforunately, most Super Bowl props are thinly bet. One wager can move the line.”

Page 154, in “Zigzag Theory”, third paragraph
Include: “in the previous game” into the first sentence.
It should read: “A refinement to the zigzag theory is to add in the qualification that the losing team in the previous game did not cover the point spread.”

Page 155, first paragraph, last sentence,
Add “in future games” to the end of the last sentence.
It should read: “Thus blindly betting the zigzag theory may be negative EV in future games.”

Page 157, first paragraph, third sentence.
Change to: “The money lines for the individual games are connected to both as well.” (delete: “the series money line and the exact series line”)

Page 171, first paragraph, last sentence.
13% should be 13.5%.

Page 174, second paragraph, second sentence
Add: “implied”
It should read: “Kentucky Derby winners always become fan favorites and always see their odds decrease (meaning an implied greater chance of winning) in the other two legs of the Triple Crown.”

Page 178, “Lower Odds in the Belmont”, 5th line
“legend” should be changed to “legendary”

Page 180 “Putting the Odds Together”, 4th paragraph
Changed “estimated expected winning percentage”
To: “adjusted expected winning percentage”

Page 181 “An Example From 2002”
Add sentence in ()
It should read: “I thought the “No” at -1050 looked like a good bet and made a wager on it (this was before I did the analysis shown in this chapter).”

Page 212, first line
Insert “not” in the second sentence. It should read:
“But that does not mean you should expect ….”

Page 227, “Embedded futures bet”
Delete: “The legality of the Jordan’s Furniture promotion is being debated as this book is being printed.”
(I don’t know of any debates)

Page 229, “Season Tickets”, last line of page
Change “those” to “regular-season”
“…There is strong demand for tickets to regular-season games, and so…”

Page 230, “Fantasy Sports”, second sentence.
Change: “same as are” to “similar to that”
“…Their evaluation techniques are similar to that used by sports bettors who bet on player proposition bets.”

Page 232.
Change “FezziksPlace.com” to “LasVegasAdvisor.com”

Page 232, after “Donbest.com”
Add: “GJUpdate.com: updates current lines and line movements.”

Page 235, near the bottom of the page, insert “he”
“Worthy has admitted he made a bad defensive play by jumping out and leaving his man uncovered.”

Glossary, add:
Efficient line: a line that accurately portrays the true odds in the sporting event

Read Full Poker Blog Post

Chapter 10 not applicable to preseason NFL

Date: Sat, Aug 18, 2007

I've received a couple of emails asking about using the information in Chapter 10: NFL First Halves for preseason games. In my opinion, the data from the regular season is not applicable to the preseason.

The reason the data is useful for the regular season is that teams really do want to win regular season games. Both teams are motivated, whether they are favorites or underdogs. But in the preseason, the motivation factor can be very different from team to team, and it can depend on the quarter and half. A team could be very motivated to perform well in the 1st half as they are tuning up the starters for the regular season. Or they could be trotting the starters for just one quality series and planning on sitting them down afterwards.

In short: do not use the data in chapter 10 for the preseason. It is only useful for the regular season. In fact, a general rule that should be applied is: preseason NFL is not the same sport as regular season NFL. Attack the preseason as a completely different sport.

Read Full Poker Blog Post

KC Season Win Total

Date: Thu, Aug 16, 2007

In a post I made on August 9, I posted that I took KC Over 7.5 wins at even money. I don't like the bet much anymore. That's because Fezzik, a handicapper I respect, announced through Jon's conference call that he liked the Under. He didn't state why he liked the under, but obviously one has to take his opinion into account. If I had known he liked Under 7.5 before I made the bet, I would have re-evaluated and possibly not bet the Over (although I would probably not take the Under as it would take too big of an adjustment for me to like the Under even given Fez's opinion). That's how strongly I value his opinion.


Interestingly, last year Fezzik and I also had different opinions on KC. He took KC Under 9 and I took Over 9. I got lucky as KC won their last two games for a total of 9 and we both pushed on the bet. There won't be any pushes this year.

Read Full Poker Blog Post

SportsConferenceCall.com

Date: Wed, Aug 15, 2007

I'll be on Jon Spevack's conference call tonight. The call starts at 11:00pm Eastern (8pm Pacific), I'm booked to show up at 11:30pm. You can get more information from Jon's site at SportsConferenceCall.com

Sorry for the late update. I knew about the scheduling for a while, but forgot to post it on the blog. In order to listen live, you'll have to email Jon to get a password. Shows are archived and there is a link on the site.

Read Full Poker Blog Post

Good explanation of baseball's waiver process

Date: Tue, Aug 14, 2007

The baseball trading deadline may be over, but teams can still get trades done. What? That doesn't make much sense. It is all due to the waivers process, which is often confusing to fans. Here is a good explanation of baseball's waiver process, written by Dayn Perry.

Read Full Poker Blog Post

Preseason Football

Date: Sun, Aug 12, 2007

Here is my article in the Two Plus Two Internet Magazine on Preseason Football. The article covers basic stuff and it is only the tip of the iceberg.

Here is a post by blogguy on the LVA Sports Forum that I think is interesting regarding preseason week 1 performances: blogguy's post. Blogguy often has interesting and insightful posts. I also acknowledged him in my book for commenting on a couple of my chapters when they were in review. He's a guy that posts on internet forums and I always make sure to read his posts carefully.

edited: unfortunately it looks like blogguy's thread has been deleted from the LVA Sports forum. It could have been deleted due to a mistake or poor behavior by other posters in that thread. That's too bad. Keep an eye out for blogguy's other posts if you like to read message boards.

Read Full Poker Blog Post

Other Regular Season Win bets

Date: Thu, Aug 9, 2007

Here are other Regular Season Win bets I made in Las Vegas. Rather than go into them in detail, I'll just list them all with my expected fair value numbers.

ARI Under 7.5 -110. FV -134
BAL Under 9 +120, +130. FV +109
CAR Under 9 -120. FV -152
CHI Under 10 -115. FV -141
DAL Under 9 +125, +120. FV +100
GB Under 7.5 -150. FV -173
KC Over 7.5 EVEN. FV -125
MIA Over 7 -130, Over 7.5 +130. FV -135, +118
PHI Under 9 +125, +120. FV -101
SF Under 7.5 +135, Under 8 -125, -130. FV +104, -156

Add those to the previously listed: OAK Over 5 EV, IND Under 10.5 +110 & NO Under 9.5 -120.

That's a total of 3 Overs with totals below 8. Of the 10 Unders, three of them are under 8 wins.

A couple of notes.
In the NFC, I think the best teams are: CHI, PHI, DAL and NO (with the last three very close). But I took the under in each of those teams. That's due to the large difference between AFC teams versus NFC teams. The expected line in the Super Bowl is AFC -6/-7. I run my season total estimations along with my playoff and Super Bowl simulations, so they are using the same data. I made sure the Super Bowl simulations calibrated so that the AFC is expected to win the Super Bowl about 67% of the time. Naturally, that means taking down the NFC teams a bit.

Several of my bets I made in Las Vegas have better lines in the offshore books. But frankly, I'm leary of leaving my money out there until the end of December. There are too many risks for my taste. Even if I trusted the offshore books, there can be exogenous events that they have no control over. Many of those exogenous events can't and won't happen to Las Vegas sportsbooks.

I'll revisit these 13 bets at the end of the year, and maybe in the middle. Remember, going 9-4 or 4-9 doesn't mean much either way. Just like going 6-0 in the bets I posted last year (see my book or archives in the blog), did not mean anything. The sample size is just too small.

Read Full Poker Blog Post

Indianapolis Colts Season Win Total

Date: Thu, Aug 9, 2007

Last week, I made a few regular season win total bets in Las Vegas. One of them was taking the Under on the Indianapolis Colts. I made three bets that I thought had value.

Indianapolis expected wins: 10.28
Fair value: Under 10.5 -117
Wagers I made in Las Vegas: Under 10.5 EVEN, +105, +110, +115

Note 1: this seems to be a trendy pick. Pinny currently (August 9 morning) has Under 10.5 at -106.
Note 2: the line in the first game against the Saints at Pinny is Colts -6. Since I like bet both the Saints Under and the Colts Under, it means I have little rooting interest in this game (assuming I bet the same amount). It is only 1 out of 15 games, but I like the fact that the two bets hedge each other slightly. Individual game expectations (opponent, IND win probability, IND point spread)

NO 69% -6
at TEN 64% -4
at HOU 65.5% -4.5
DEN 64.5% -4.5
TB 81% -10.5
BYE
at JAX 50.5% Pick
at CAR 55.5% -2.5
NE 51.5% -1
at SD 40% +3
KC 74% -7.5
at ATL 62.5% -3.5
JAX 66.5% -5
at BAL 49% +1
at OAK 72.5% -7
HOU 81.5% -10.5
TEN 80% -10

Game 1 against the Saints is easy to estimate since there are already lines on that game. The games against SD and NE are interesting since it is a clash of the three best teams in the NFL. I have the Colts just slightly below the Chargers (I would have the Chargers higher if not for the question marks about a completely new coaching staff), thus a 3 point dog in SD. I have NE better than both the Chargers and the Colts, thus NE is only a 1 point dog in IND. Playing those two games in back-to-back weeks will be interesting.

The last three games are also interesting. I have the Colts as significant favorites at Oakland, and against Houston and Tennessee. Being that far away, the line could be very different by the time Week 15 comes around. Aside from that uncertainty, there is also the uncertainty of how much the Colts will need those games. If they are well ahead in the division, say they are 10-4 before the Houston game, it is possible the last two games mean nothing to them. For example, say JAX is only 7-7 and the Colts have the division wrapped up, and say both NE and SD are 12-2 and have tiebreaker advantage over IND, so the Colts cannot get a bye in the playoffs. While their talent may indicate they should be 10 point favorites against HOU and TEN, their situation may dictate a lower line if they do not need to win either game. Therefore, possibly this gives the Under 10.5 bet a slight advantage. Also, it may mean that the normal approximate of 50 cents for a half-win in regular season totals is not as applicable to the Colts case. A Under 11 bet seems more valuable than a typical half-win. I have read other people picking Under 11, unfortunately I did not see any sportsbooks using that line in Las Vegas.

Other notes not in the Colts' favor:
The Colts won the Super Bowl last year. Although they played well on defense in the playoffs, there are still a lot of question marks due to their defensive performance in the regular season. In addition, they lost some quality players on defense in the offseason, both to free agency (especially LB June) and injuries (DT McFarland). On offense, they are expected to be solid especially with Peyton Manning at QB. However, they did lose their left tackle (Tarik Glenn) due to retirement. Offensive linemen are often underappreciated. We never see them on a stat sheet and rarely see their good plays on replays. Most often, we only see them if they missed a block or committed a penalty. It is easier to pick out the bad ones if we constantly see the same player play poorly, but when an offensive lineman is playing well, fans rarely notice. The replacement may do a good job, but likely, he will not do as good a job as Glenn has done over the years.

The Colts have been at the top of the NFL for a few years. This means they have had high number draft picks (drafting late in each round) and thus they probably have less depth than other teams to replace the missing holes due to free agency, retirement and injury.

Being a Super Bowl champion and a Division winner, the Colts do not have an easy schedule. The two other division winners they face are New England and San Diego. Two teams many would rate higher than the Colts at this point.

The Colts are still a large favorite to win their division (I estimate about 66%, which is the third highest). Taking them to win Under 10.5 games does not mean I think they are a bad team, simply it means I think the number is slightly too high. Taking Under in teams with high win totals is usually the better way to go, and IND is a prime example.

Read Full Poker Blog Post

New Orleans Saints Regular Season Win Total

Date: Tue, Aug 7, 2007

I went to Las Vegas this weekend and made some NFL futures bets. Most of my bets were on regular season win totals (RSW for short). Because the sportsbooks make tighter markets in those wagers (30 cent lines) than other futures, it makes sense that positive EV in RSW totals are easier to find. Over the next few days, I’ll post a few bets I made with explanations. I’ll also post these at SharpSportsBetting.com. Please visit the thread at this link to comment or read comments.

I made six different Under bets on the New Orleans Saints. I took them Under 9 and Under 9.5 (see prices below). Included are the individual game expectations that I used to arrive at the fair value prices and expected wins.


New Orleans expected wins: 8.77
Fair value: Under 9 –124, Under 9.5 –181

Wagers I made in Las Vegas:
Under 9 +115, +115, +110, +125
Under 9.5 –120, -135

Note 1: sadly, Pinny has Under 9 +131, but I can’t bet there. Arg. Well, I guess it doesn’t matter that much. In the current atmosphere, I wouldn’t play any long-term futures at any offshore book. FWIW, I've always felt Pinny was the best at individual game markets since they had an active tight market. But I often found value on relative-value plays and futures there. So I am not concerned that their line seems off, like I would be if it was an individual game where they get lots of two-way volume.
Note 2: the line in the first game against the Colts at Pinny is Colts -6

Individual game expectations (opponent, NO win probability, NO pointspread)
at IND 30.5% +6
at TB 53.5% -2.5
TEN 68.5% -6.5
BYE
CAR 60% -3
at SEA 43% +3
ATL 67% -5.5
at SF 50.5% Pick’em
JAX 55% -2.5
STL 67% -5.5
at HOU 54% -2.5
at CAR 44% +2.5
TB 69.5% -6.5
at ATL 51% -1.5
ARI 69% -6.5
PHI 56.5% -3
at CHI 38% +3.5


I think the market overestimates New Orleans. In particular, I think bettors will be thrilled to bet the Over, and the sportsbooks are aware of that so they are happy to take my Under bets. The main reason I think the market (made up of mostly squares) overestimates New Orleans is that they have a flashy offense. They have two big names at the skill position: Brees and Bush. Bush is a really exciting player. They throw the ball a lot, and fans love that. The Saints offense is flashy, thrilling, exciting and fun football to watch. But their rushing offense and defense concerns me: the Saints rushed for 3.7 yards per carry, only 25th in the NFL. The Saints defense had a tough time stopping the run, allowing 4.9 yards per carry, tied for 2nd worst in the NFL.

Another issue against the Saints is that they had a relatively easy schedule in 2006. They only won 3 games in 2005 and were able to play the designated 4th place teams from other divisions, including: GB (won 34-27) and SF (won 34-10). GB and SF were both improved in 2006 as well, but they were still only middle-of-the-pack teams and not division leaders. So the Saints did receive a little help in the schedule. They won the division in 2006, and now they get the brunt of the schedule in 2007. They have to play division winners from 2006 in PHI and CHI. The Saints may get some luck in that they play at CHI in the last game of the year. The Bears are a big favorite to win their division, so it is slightly more likely that the last game of the year won’t mean anything to them, and thus they may be playing the second-string players, allowing NO to have an easier game. But a lot of things can happen. It could turn out CHI needs that game for a Bye in the first round of the playoffs. Or maybe the Saints are 9-6 going into the game and have the division clinched, and maybe they are the ones playing the second-string.

In summary: Betting the Under in RSW bets when the total is over 8 is better than betting the Over. Flashy offenses are usually more overrated teams, and defenses are less appreciated (good or bad). The Saints have a flashy offense and a mediocre defense. The projections of the lines during the season lead to Under 9 and Under 9.5 looking like they have value.

Read Full Poker Blog Post

August article on Two Plus Two

Date: Mon, Aug 6, 2007

I've been traveling for the past week (still traveling right now), and my laptop's hard drive crashed on me in the middle of my trip. So that's why my post for the link to my August article for the Two Plus Two Internet Magazine is so late. Well, here it is...it is on betting the NFL preseason

Read Full Poker Blog Post