Writers of Pro Football Prospectus 2008

13 Oct 2005

Dr. Z: The Odds are Against You

Paul Zimmerman writes this week about why sportswriters have such a lousy record of picking games. Why are these experts unable to do any better than 50-50?

"Because they're not experts," says Zimmerman's source within the betting industry. "They're just people pretending to be experts. Sportswriters are notorious for not being able to pick games. We, and by we I mean the professional handicappers, are in a different universe out here." Interesting stuff.

Posted by: Michael David Smith on 13 Oct 2005

50 comments, Last at 14 Oct 2005, 11:48pm by Clod

Comments

1
by Bettin Man (not verified) :: Thu, 10/13/2005 - 6:36pm

What was "Carl" last week? 11-2 or something like that?

If he tells us who to bet for this week I am going to do it.

2
by MAW (not verified) :: Thu, 10/13/2005 - 6:54pm

I like the anecdote about the bookie giving 25 bettors one team, and the other 25 the other team. The Sports Guy had it in his review of Two For the Money just yesterday.

Anyway, I can't imagine being one of the 25 given San Fran as a lock to beat Indy last week. You'd have to be really stupid, or really gullible.

3
by Teddy (not verified) :: Thu, 10/13/2005 - 7:00pm

They wouldn't have said SF would win; they'd have said they would cover. That makes it slightly more believeable.

4
by Starshatterer (not verified) :: Thu, 10/13/2005 - 7:02pm

There are a few dynamics that explain some of this.

1. According to a show on football handicapping from NFL Films (really!), many local "experts" feel obligated to pick the home team and against their rivals (Note: does not apply to Ron Borges.). That'll ruin your average when the local boys are having an off year.

2. Many of these picks are against the spread, which means it's not the odds of picking the winner, but the odds of picking the winner more often than does the "wisdom of crowds." Considering high-priced mutual fund performance in the stock market, experts being less wise than the crowd is not all that far-fetched.

3. Finally, I suspect Dr. Z is guilty of remembering the failures, like the touts who claim 67% success are remembering their successes.

5 to 4 he's selectively remembering. ;-)

5
by JMR (not verified) :: Thu, 10/13/2005 - 7:17pm

There's one huge fallacy in his discussion that assumes 50% is the mean or "blind" success rate. This would be true if NFL lines were set to predict the outcome of the games. But they're not; lines are designed to maximize the expected profit for the casinos, so it is reasonable to see these "experts" that presumably mirror the betting biases of the population as a whole to land under 50%. Click on my name for a link to a paper that discusses this in depth.

6
by princeton73 (not verified) :: Thu, 10/13/2005 - 7:36pm

I like the anecdote about the bookie giving 25 bettors one team, and the other 25 the other team. The Sports Guy had it in his review of Two For the Money just yesterday.

I've been hearing that story for AT LEAST 30 years--afraid it might be a UL

(but a good one)

(and by the way, Larry Merchant's book that Dr Z. mentioned is very good, if you can find a copy)

7
by Will Allen (not verified) :: Thu, 10/13/2005 - 7:54pm

I suspect that very few of the people who can consistently be right more than 60% of the time against the spread are known to very many people. I know Hank Stram had a formula which produced a phenomenal record on Super Bowls; did he ever apply it to regular season contests? I've long thought the Super Bowl is about the easiest game of the year to pick, given all the uneducated or homer money affecting the spread.

8
by stan (not verified) :: Thu, 10/13/2005 - 7:58pm

Let's face it, a lot of coaches and players don't know what the heck is really happening out there. Just listen to the ex-players and coaches serving as "analysts" for the networks. How many times do you have to listen to them use stats improperly or confuse causation ("teams that run X times in a game win a huge majority of the time, so this loss was due to a failure to call enough runs") before it sinks in?

These players and coaches can provide a lot of insight on topics such as technique, atmosphere in the locker room or huddle, etc. But it should be clear that they very often don't understand the real keys to the outcome of the game.

And if these folks don't know, why should a sportswriter? Especially because a sportswriter has had his head filled with years and years of "coach speak" which is very often deliberately inaccurate. That makes it even harder to separate wheat and chaff.

9
by Independent George (not verified) :: Thu, 10/13/2005 - 8:33pm

I think the thing to remember is that anybody good enough to win over 60% of the time is not going to advertise the fact. It's like picking stocks - if you really do have inside information, the very last thing you want to do is let anybody know about it. Your advantage disappears the moment you do. Anybody that good is going to try and remain under the radar.

10
by ycl (not verified) :: Thu, 10/13/2005 - 8:35pm

I read this somewhere once, and I believe it's true. The spread given for a particular game is NOT a reflection of what the odds-maker thinks will be the likely actual outcome. Rather, it's simply the number at which, in the odds-maker's opinion, there will be an equal number of bets placed on either side. The house shoots for this balance, of course, because the house will always make money off the vig.

The spread, in other words, is epiphenomenal: It's not a prediction of a real event (i.e., the final score of the game), but a prediction of what the betting masses believe about the outcome of the game.

The comment above about the Super Bowl being the easiest game to bet on thus rings somewhat true. (Though I tend to think the 1st round of the playoffs is the best time to wager). When oddsmakers know that there are more dupes or amateurs wagering, they may skew the spread to accomodate the additional ignorance. And more knowledgeable bettors, aware of this "ignorance" enhancement, as well as the true meaning of the spread number, may be better off.

11
by NFC Central Freak (not verified) :: Thu, 10/13/2005 - 8:41pm

As an individual who spends a lot of time at the racetrack and wins, I will certainly agree with the sentiment that the vast majority of folks betting on games have no clue.

Why?

First, they aren't serious about it. And by serious I mean taking on this task like it's a job. You can't expect to win consistently doing this "part-time".

Second, you had better enjoy research. Collecting data, knowing how to properly organize, and then properly analyze it. And most available data is garbage. There are no short-cuts.

Finally, you better have a facility with numbers. And beyond the technical stuff it also helps to be creative. You can win being a plodder (proficient) but that means getting by. You want to win big you have to have the "ah-ha" gene.

Folks are likely to read this and roll their eyes. Fine with me. Good betting is work involving higher math skills and some degree of perception. Most people who fit that description work on Wall Street or in Silicon Valley.

I like the horses. I travel to interesting places, I get to practice my language skills, the folks in the sport are diverse/fascinating (owners, trainers, jockeys, stable guys, etc.), and I am good at it. Must be the old farm boy in me coupled with that Q.A. degree. Nothing like Markov mixed with smell of mint juleps to get the blood pumping. :)

12
by Joey (not verified) :: Thu, 10/13/2005 - 9:26pm

"The spread given for a particular game is NOT a reflection of what the odds-maker thinks will be the likely actual outcome. Rather, it’s simply the number at which, in the odds-maker’s opinion, there will be an equal number of bets placed on either side."

That's exactly how they do it. I believe it was SI who did a really interesting story a couple years ago. There's this guy in Vegas who has been doing it for like 30 years who is the standard all the others measure themselves against.

13
by Trogdor (not verified) :: Thu, 10/13/2005 - 10:16pm

I refuse to believe that someone writing for a major paper, or a leading football website, could be so bad at picking games in his area of expertise. Russell? Vinny? You guys with me on this one?

Oh nevermind, I forgot your records are even worse than Michigan's, which is pretty hard to do.

14
by JonL (not verified) :: Thu, 10/13/2005 - 11:37pm

I think Mark Maske in the Wash. Post is somewhere between 60 and 65%. And he only picks winners, not against the line. (And for the record, he picked the Chiefs this week.)

15
by Dennis (not verified) :: Fri, 10/14/2005 - 12:01am

Re #5: The reason 50% is considered the "success rate" is because over 50% you make money and under 50% you lose money. If the public at large has a 45% success rate and you hit 48% to beat them, so what? You're still losing money.

16
by Loophole (not verified) :: Fri, 10/14/2005 - 12:46am

Does anyone know of a good source to find historical betting line information (i.e., the lines from games played over the last few seasons)?

17
by NFC Central Freak (not verified) :: Fri, 10/14/2005 - 12:57am

Actually, the "success rate" is another in a long line of B.S. to fool the public.

Because your winnings are a factor of how much is wagered.

One good bet on decent odds can overcome a series of minor wagers gone bad.

But the public, being mathematically illiterate, can't understand that in a 10 second clip touting a betting service. So they tell you they pick X percentage correctly.

Easiest example of this is the Breeder's Cup a few years back when I was getting hammered through the first seven races but stuck to my guns on The Classic and Pleasantly Perfect did me proud. I had the horse picked at the beginning of the day and even though I was ready to go home in the figurative barrel I stayed the course and walked away with serious cash as PP went off at 14-1. About twice the odds I expected on a quality horse even accounting for the competition.

Anyway, success rate is a bogus stat.

And folks who win money don't try and help others. Nobody is truly altruistic.

I help family and friends on a casual basis when we are spending a fun day at Keeneland or Saratoga. Beyond that I keep to myself. Do I have a system? Only if you call reading, interviewing folks around the horses, observing early morning workouts, and then figuring odds vs. various variables a system. I wouldn't.

Which is why football is a sucker's game. Too many unknown variables. In horseracing much of what is perceived as unknown is in fact known. But when it comes to football all those numbers folks discuss are mostly statistically insignificant. Plus the people factor. People are notoriously unreliable.

Folks tempted to wager on pro football would be better served to take their wagering money, put it in a can, and bury it in the backyard. Or if they are bound and determined to lose it then start a fire. At least you will be warm for a short while.

Again, I understand that these scribblings will be dismissed because folks just "know" they can figure it out.

Sorry to be the splash of cold water.

'Cause you can't and you won't.

18
by Jim A (not verified) :: Fri, 10/14/2005 - 2:21am

There’s this guy in Vegas who has been doing it for like 30 years who is the standard all the others measure themselves against.

I think you're thinking of Roxy Roxborough, founder of Las Vegas Sports Consultants, the company that provides the opening lines for most sports books, though I think he's passed most of his work onto other employees now. I do recall the SI article about them and it was fascinating. Lots of serious number crunching going on there.

There was also a lengthy interview with Roxborough in Chance Magazine a few years ago. One of the things he said was that the gambling lines were very efficient, in practice. That is to say they end up being about as good of a predictor of outcomes as you can get. For example, even if a bad line split the money 50/50, the high stakes pro gamblers would come along and take advantage, which would quickly correct the line. For this reason, it's important for the oddsmakers to get the initial line as accurate as possible or the books would lose money. The oddsmaker doesn't have to be clairvoyant, but he probably has to know as much about predicting outcomes as anyone.

19
by Kibbles (not verified) :: Fri, 10/14/2005 - 6:19am

Re #5: There’s one huge fallacy in his discussion that assumes 50% is the mean or “blind� success rate. This would be true if NFL lines were set to predict the outcome of the games. But they’re not; lines are designed to maximize the expected profit for the casinos, so it is reasonable to see these “experts� that presumably mirror the betting biases of the population as a whole to land under 50%. Click on my name for a link to a paper that discusses this in depth.

Ummm... but 50% *is* the "blind" success rate. If you make all wagers based on the flip of a coin, you're going to be right 50% of the time, and you're going to be wrong 50% of the time. Actually, you'll probably be right 45%, wrong 45%, and push 10%, or something to that effect.

Regardless, assuming there was no vig, if you bet based on a coinflip (or some equally random occurance, such as a die roll), then long run probabilities state that you'll wind up breaking even.

Anyway, for what it's worth, I would think the second round of the playoffs would be where the money is at. If a team gets hot in the first round and absolutely blows out its opponent, then the masses are going to be loving that team in the second round when, historically, the home team wins the overwhelming majority in the playoffs. I keep thinking of Indy being favored over New England after blowing out Denver last year.

Of course, that's certainly not foolproof, either, because I also remember Indy winning a shootout against KC after blowing out Denver the year before.

For a long time my favorite trick was pick against Denver in the cold, on prime time, and after week 6, because they tend to drastically underperform in all 3 situations. Especially the cold. Everyone assumes Denver has an advantage in the cold, being a cold weather team, but they historically do abysmally against the spread in cold games.

I know this because I'm a big Denver fan- which is also my curse. Every year, I tell myself that I'm finally going to put my system into effect and bet against Denver in every night game and every game under 32 degrees at kickoff that comes after week 6, but every year I can't bring myself to bet against my own team. So I don't bet, and I just track my success, and every season when Denver misses the playoffs or gets blown out in the first round I kick myself for not taking advantage and laying some money against them during the season.

20
by Moe (not verified) :: Fri, 10/14/2005 - 7:20am

Actually you need to be be better than 50% to make money since the house pays 100 for bet of 110.

Imgaine two bets from last week

Bet 110 on GB -3 vs. NO (win)
Bet 110 on Ravens +1 vs. DET (lose)
Total Bet = 220

You won 50% of your bets, but you walk away with only the 110+100=210 from Green Bay. Net result = loss of 10.

Someone better with percantages than me can tell us what winning percentage is break even.

21
by Loophole (not verified) :: Fri, 10/14/2005 - 8:26am

To break even in Vegas you need to win just over 52.38% of your bets (11/21).

22
by VarlosZ (not verified) :: Fri, 10/14/2005 - 9:23am

"The spread given for a particular game is NOT a reflection of what the odds-maker thinks will be the likely actual outcome. Rather, it’s simply the number at which, in the odds-maker’s opinion, there will be an equal number of bets placed on either side. The house shoots for this balance, of course, because the house will always make money off the vig."

That's not necessarily true. The bookies are out to maximize their expected value, not to minimize their variance. While they don't mind balanced betting, they also don't mind unbalanced betting if they're confident that it will be unbalanced in the wrong direction. Most games, in fact, get action that is unbalanced past the point where the vig will make up for the "wrong" team winning.

23
by Jets Fan (not verified) :: Fri, 10/14/2005 - 9:28am

did I actually read "seperate wheat and chaff" in post #8?, what the heck is going on here?

24
by Al (not verified) :: Fri, 10/14/2005 - 10:40am

My favorite NFL picks feature in a newspaper is in the NY Post every Friday. A bunch of local car dealers get together and buy a two page ad and make their NFL picks for the week. Not only is it ridiculous that anyone would care who Joe and Nick from Manfredi Auto in Staten Island like this week, but they also fill the ad with inside jokes about the car dealers that no one would get.

This week, they have a graphic labeled "The Lords of the No Rings" with pictures of Don Mattingly, A-Rod, Mussina and Giambi, which is kind of funny. But then they photoshoped the faces of two random people, who I'm guessing are car dealers, onto the faces of Frodo and Gollum. WTF?

25
by Ray (not verified) :: Fri, 10/14/2005 - 11:25am

I think that the "success rate" is of value because, theoretically, the odds of winning a bet are 50/50. So then it follows that if you have the same odds on all of your bets, you should distribute the money evenly as well. If you're distributing the money evenly, then you don't make more money based on one bet winning, you make money based on how many of your bets were correct. 60% success rate means you make money.

That said, if there was a solid "system" that you could use, then it wouldn't be gambling, and the house would change the game.

26
by NFC Central Freak (not verified) :: Fri, 10/14/2005 - 11:43am

It is an axiom in gambling that to win big you have to bet big.

To the general public the opposite information is conveyed because good gamblers WANT the suckers in the mix. It helps to have a stupid crowd wagering for obvious reasons. Because you aren't really betting against the house. You are wagering against the other bettors.

Spreading money around is exactly what the house wants to do as you believe you are minimizing your losses while they rake in a take on each bet. From a psychological standpoint the play is defeatist in nature. Think about it. The bettor is saying, "I am going to lose so to offset those defeats I need to bet elsewhere." So you have LOST before you have even wagered.

It ain't gambling if you know what you are doing.

27
by pawnking (not verified) :: Fri, 10/14/2005 - 12:10pm

“Because they’re not experts,� says Zimmerman’s source within the betting industry. “They’re just people pretending to be experts. "

In other words, sports reporters are exactly the same as every other reporter.

Also, is it suprising that the many would be smarter than the one? Not according to this. A bunch of uneducated dolts who bet on sports will put the line where a more educated than usual man cannot guess which way to go, if you believe this theory.

I happen to believe that sports reporters know more about football than the average better, but not more than the sum intelligence of all betters, as represented by the betting line. Thus, they are unlikely to do better than .500. Factor in a desire to please the home team readership and the desire to cause a stir by giong with daring, bold picks (for which they will be remembered). These represent ineffeciencies in picking. Given this information, you can see that newspaper experts should be expected to have a losing record in picking games.

28
by Jeremy (not verified) :: Fri, 10/14/2005 - 12:13pm

I'm not sure how fair it is to judge bettors when they pick all the games (as most do in newspaper columns). I doubt that many professional gamblers are playing that way.

29
by pawnking (not verified) :: Fri, 10/14/2005 - 12:19pm

Jeremy, if you assume that a handicapper must pick all games, you have to assume that he'll go .500 in games that he doesn't have any good feel for, and above or below .500 in games he has a strong opinion about. The fact that they don't know about a lot of the games indicates a stronger liklihood they will regress to .500 in total.

The fact that so many have an average below .500 indicates that the games they have a strong opinion about they are picking very poorly. Remove the stabalizing influence of the games they know nothing about, and their record would actually look worse, not better.

30
by MDS (not verified) :: Fri, 10/14/2005 - 12:44pm

The pros don't bet every game; they choose the ones they're most confident about and bet those. But those would be covered by the "best bets" that a lot of newspaper prognosticators list, and they're often worse on the best bets than they are overall.

31
by jay (not verified) :: Fri, 10/14/2005 - 12:55pm

I'm aware that the lines are not set as a reflection of the average outcome, but as the spread that should result in the maximum profit for the bookkeeper.

Given that this is a statistical website, what I'm curious about is how much of a difference this makes. Just about any habitual gambler is aware of this fact, and uses to explain why *they* can beat the spread. After all, said habitual gambler is not just anyone, but they know that the bookkeeper is setting the spread for morons, not a goliath of gambling such as themself. I think this fact is what allows bookies to make so much money- it's almost a willing suspension of disbelief.

What would the lines look like if they were set as probable outcomes for games? Maybe in a few years Aaron et al can answer that question.

32
by Will Allen (not verified) :: Fri, 10/14/2005 - 1:11pm

I suspect #22 is likely correct, in that I firmly believe that if I were to openly stroll up to the window at the Las Vegas Hilton every week, and place my wagers at a consistent 65% success rate, it wouldn't be too long before I was informed that my wagers were no longer welcome.

33
by pawnking (not verified) :: Fri, 10/14/2005 - 1:17pm

Bottom line on spreads. If you believe that the information market for football games is an effecient one, then you have to believe that it is basically impossible to beat the spread long term.

Read this book. I think we will agree that the money at stake on Wall Street is at least as much as the money gambled on football. Does the fact that the "experts" in picking stocks don't do statistically better than a monkey throwing darts at an index page tell you anything about how an effecient market works?

It does not matter if the group picking the stocks are individually not as smart as the expert. In total, they are much much smarter. Thus the expert (not just someone pretending to be an expert, mind you, but an expert) cannot hope to outperform the market. If he could, then the market would figure out how he's doing it, and adjust the stock price before the expert could evaluate the data.

So it is with football. If an expert can truly pick winners, the result would be like Robert DeNiro in "Casino." That is, the line would move based on his opinion. Only real life would take it one step further. The bookies would co-opt DeNiro or at least his methods, and adjust the line before he could make the pick. Thus DeNiro would not be able to outguess the line long term.

Of course, there are those who do outperform the average for a short term, maybe even for years. However, the laws of statistics will always catch up to them. Think of it like this:

supposing that instead of picking the winners against the spread, millions in America was picking the results of 16 coin flips every week. Thousands of experts would give their picks, and based on the results, we'd find that some of them picked correctly at least 11 or 12 of these picks right. Maybe someone would actually pick all 16 correctly. The next week, some of those winners would again have winning percentages, and so on the next week.

After a few weeks, the winning pickers would have gained huge recognition for their eerie ability to consistently pick winners. They would be invited on radio talk shows, have their own 900 numbers, and etc etc. simply because they have shown they are able to correctly pick winners. But the sad fact is that they are simply a result of a large statistical sample inevitably giving some a winning percentage over a short time. After a few more weeks, they would not have a better record than average.

34
by Will Allen (not verified) :: Fri, 10/14/2005 - 1:53pm

pawnking, the stock market IS very efficient. However, it is not perfectly efficient, and since none of us are immortal, we don't have to pay excessive attention to what will happen in the very long run. Thus, Warren Buffett's methodology has served him exceedingly well, even if it serves him far less well now than it did 30-plus years ago.

35
by Dan Babbitt (not verified) :: Fri, 10/14/2005 - 2:10pm

"There was also a lengthy interview with Roxborough in Chance Magazine a few years ago."

http://www.amstat.org/pressroom/roxborough.pdf

36
by pawnking (not verified) :: Fri, 10/14/2005 - 2:34pm

Will, WB certainly has done well for himself, but
1) he got lucky,
2) he had a buy and hold philosophy, which compounded his luck, and
3) the magic of compunding interest caused his luck to bloom to amazing levels.

Not to be too dismissive of his accomplishments, but I think his success is mostly attributable to those three things. Can you prove me wrong? If he really has a crystal ball, why isn't it eqully effective every year, especially lately?

37
by Will Allen (not verified) :: Fri, 10/14/2005 - 2:51pm

Pawnking, I think you misinterpret my characterization of Buffett's success. I never claimed he had a crystal ball. Buffett's methodology, particualrly in the early years, involved identifying companies which the public undervalued, due to it undevaluing, or not fully appreciating, the collection of assets owned by the companies. This worked phenomenally well after the late sixties boom ended, when pessimism began to take hold regarding stocks. He hasn't been as effective lately simply because his methodology is much more widely employed, so he has a much more difficult time finding firms with an undervalued collection of assets. This gibes perfectly well with your theory above.

However, since even Warren will eventually leave this Vale of Tears, he doesn't have to worry about the very long run; a couple decades of above-average success more than suffices. As Keynes once remarked, "In the long run, we're all dead." In between the short run and the long run, however, some above average investing can be done, but it sure as hell ain't easy, or even easy to identify who will do so.

38
by pawnking (not verified) :: Fri, 10/14/2005 - 3:31pm

Will, we find ourselves in agreement, which is unsuprising given we both frequent this site.

I think that when the history of business from the 60s on is written, the largest innovation which will be discussed will not be the computer, but the use of statistical analysis in business as having the biggest impact on the economy. Football is just another example of this trend.

39
by Astro Boy (not verified) :: Fri, 10/14/2005 - 3:37pm

RE: 35 Thanks for the link – it was an interesting article. I particularly liked Roxy’s comments about teaser bets. Basically, as other posts have made clear, the goal of a line is to split the money wagered, not to predict the outcome. However, the lines often do a such a good job of predicting the outcome within a few points that the extra six or so points that a teaser bet provides can be critical. Last year, the success I had with against the spread picks just fell off a cliff, but I still had a positive season moneywise because I had a lot of teaser bets pull through. Generally, I like to play a lot of three-team teaser bets, with the third game on Sunday or Monday Night. If the two early plays pull through, I’ll hedge by betting on the other side in the night game and try for a middle. I caught a big one two weeks ago in the Carolina/Green Bay game, when I had teased bets with Carolina laying only one point, but an against the spread bet with Green Bay getting seven and a half points. Carolina won by 3, so I won all bets.

I was hoping to respond to post 17 by noting that I’ve won money each of the past two years and that I’m ahead this year as a result of that middle, but the amount of money I’ve won is nothing to brag about. Basically, given all the effort that went into it and all the risks I assumed, I’d be hard pressed to say that I wouldn’t have been better off burying the money in a can in the backyard and digging it up a few years later. One reason for my low win totals is simply that I don’t wager that much. NFC CF is correct in saying that to win big, you have to bet big. I could increase my stakes, but I’m not so sure how much more money I want to wire down to those Caribbean sports books. I’d love to hear other people’s experiences betting and actually collecting their winnings from these houses, but I realize that the goal of this website isn’t to create a gambling forum.

Just the same, this topic won’t go away. It’s one thing to write on Tuesday and explain everything that happened the previous Sunday. It’s another thing to write on Friday and try to predict what will happen this coming Sunday. Dr Z’s piece did a good job describing what’s really an open secret: on the whole, sports writers are absolutely atrocious at picking games. In fact, so they’re mind-bogglingly bad at it that the most logical “system� for gambling in the NFL is to find a documented loser in a paper that publishes its “expert� picks and go opposite all his picks for the rest of the season. This may work for a season or two, but over time, other variables, such as your ability to manage your bankroll, come into play, and I imagine you’re not likely to see long-term success using this technique.

40
by Björn (not verified) :: Fri, 10/14/2005 - 3:53pm

The dudes at espn.com pick games straight up, and they are a combined 335-252. 57.07%. If you take out Theismann's picks (36-33), the guys are 299-219, 57.72%.

41
by Clod (not verified) :: Fri, 10/14/2005 - 4:00pm

Sorry for the double thread post, but this does have to do with odds in that I am wondering the average winning % for teams coming off of the bye.

Does anyone know the long term winning percentages of teams coming off of a bye? I seem to remember reading somewhere that it was considerably higher than Home team winning percentages and other “advantages".

Second question, do the schedule raters use this to rate the toughest schedules at the beginning of the year or just simply the teams cumulative winning percentages from the prev year?

My point is this…Look at this teams 6 team stretch and tell me they don’t have a tougher schedule than NE.

@ NE
Pitt off a bye
@ OAK off a bye
@ PHI off a bye
KC
@ NY Jets off a bye

That seems like the roughest shedule stretch in the NFL this season. Now i’m not going to throw the word unfare around, but I’d be curious what other team has this many teams (let alone quality ones) coming off of there bye week in there schedules.

BTW, its the Chargers

42
by Troy (not verified) :: Fri, 10/14/2005 - 5:03pm

Odds-wise, teasers are a horrible way to bet. The payout does not square with the reduced odds ... parlays are the exact same way.

43
by MDS (not verified) :: Fri, 10/14/2005 - 5:06pm

That's interesting, Clod. I know I've seen records after bye weeks but I can't recall where. I really don't see why there's so much talk about the Patriots having such a tough schedule. There's a formula that determines every team's opponents. The only thing that's determined by the league is when you play, and I've seen zero evidence that the Patriots have it any harder than anyone else as far as short weeks and rested opponents.

44
by Parker (not verified) :: Fri, 10/14/2005 - 5:16pm

I think one of the other issues with the newspapers experts having a less than stellar win percentage against the spread is that they pick every game. In the real world that doesn't seem like a good idea. I'm not a big gambler and am not suggesting that they would do much better if they only picked a few games each week (clearly their 'best bets' don't do too well either) but being compelled to pick a winner in a game you would rather not is a great way to lose money.

For those that gamble: Does it make any sense to bet on every game each week?

45
by Astro Boy (not verified) :: Fri, 10/14/2005 - 5:21pm

Troy-

You are correct about parlays. For example, with no vigorish, a three-team parlay should pay 8-1, and even if you include vigorish, the payout should be about 7-1, but most houses offer payouts in the range of 6-1. Teasers, however are a different story. It's true that they come with an ugly risk attached: you can go .500 with teaser picks and get wiped out, whereas going .500 against the spread means that you only lose the vig. The issue of increased payout vs. reduced odds is more complex with teasers, because it depends on how many games there are in which the extra points make a difference. That number changes week to week and year to year. The article linked in #35 noted that when this number was higher than expected one year, the house had to reduce the payout on teaser bets because some people were making more money than the house's model predicted. Some internet betting houses offer better payouts on teasers than I was able to get in Vegas a few years ago, so I'm happy to try to take advantage of them. And, as noted, if you hedge and go for middles, you can significantly increase your possible winnings while significantly reducing your risk.

46
by Jim A (not verified) :: Fri, 10/14/2005 - 6:43pm

I'm somewhat fascinated by individuals who claim to be able to beat the sports books on a consistent basis over the long term. We can probably assume that most of these guys are frauds who fabricate or massage their betting records to promote their 900 number service, or subscription web site, or newsletters/books which sell their "secrets". But I've often thought about what I would do if I genuinely had such ability? Wouldn't I simply earn my billions and retire to a Caribbean island or at least a more challenging hobby? Or perhaps my skill were real, but I didn't have enough initial capital to spend that would grow at a fast enough rate to make a luxurious living for myself. So maybe I would incorporate myself like any other entrepreneur with a moneymaking idea and allow others to invest in my abilities by selling shares publicly. Now I would have funding to make some comfortable profits, while my investors would share my successes via dividends and capital gains for no work at all.
Now perhaps the books would catch on and adjust their lines to take away my advantage. But I might be able to avoid detection by spreading my bets around the world (recall the MIT blackjack ring did this for several years, I believe, before getting caught). But I suspect the fact that nobody has done any of this, at least not publicly, but rather we are flooded by touts, strongly suggests any such long term success is unlikely to be sustained to this extent. Or maybe someone out there has done or is doing this, but simply hasn't been noticed.

47
by clod hopper (not verified) :: Fri, 10/14/2005 - 7:43pm

The Patriots have had amoung the top five or so hardest schedules every year since the 2002 season and have had an unbelievable record of like 21-1 against teams with a winning record since 2003. The Colts are second best with just over 50% against winning teams in the same time span, and have played half as many decent teams.
The Pats have had tough schedules, including against good playoff teams for a while. The Pats defense stinks this year and its going to be all on Brady to win or lose shootouts so you can throw those great statistics out the window.
Maybe the Chargers have a tougher schedule, let me know if they win three out of the next four superbowls and set a nfl record winning streak!

48
by Clod (not verified) :: Fri, 10/14/2005 - 9:11pm

HOMER ALERT!!

What that has to do with my post, I have no idea. I'm not sure what part of my post was patronizing to the Patriots to elicit such a response.

49
by Björn (not verified) :: Fri, 10/14/2005 - 9:42pm

According to DVOA, SD has the hardest sked in the NFL, and Seattle the easiest.

50
by Clod (not verified) :: Fri, 10/14/2005 - 11:48pm

That is interesting actually, but I wasn't looking for validation for there schedule as much as records of teams coming off of the bye.