Betting the NFL, Part 1

[reno]

It is my contention that the only NFL angle that is a long-time winner is betting the the Monday night home dog. And that angle obviously yields very few plays.

I've checked out dozens upon dozens of NFL angles, and seen them all fail in the long run. Even ones that made a lot of sense ended up failing.

I had a whole list of angles from years ago that I recently checked out for the past several seasons, and all were losers. Here are just a few of them:

Refrigerator Bowl Angle (This angle hit 60%+ over a 6-year period in the 80's): Bet under in November and December on games that are played in the cold (dome-free) cities near the Great Lakes.

If a team scores over 30 points 3 games in a row, bet against them their next game. (There are many variations on this theme, all worthless in my estimation.)

Go against a team that won by more than 21 points the previous week.

Anyway, if anyone has found an NFL angle that is long-time, consistent winner, I'd be surprised.

[WhiteShoes]

i agree because I feel there's alot of
creedence put on "angles in a vacuum" without
looking at other factors. Still, there just
isn't that many NFL games in any year (in
any decade really) for anyone to be sure
of anything. people are gonna believe what
they want to!

I once back-tested several incredible angles
that were something like 80-10 over the 10 years, and seemed to make perfect sense. They broke even by themselves and did little to alter the current events in the league.

the word trend itself sounds like a passing
fancy rather than a prediction of the future.
How can numbers be trendy? geez, my post
is currently on a trend...to end

[GoalGoalGoal]

The ol MNF doggies still show a profit imo or
at least they do so going into '98 and
subsequently taking a bath.

Ah but such are trends.

Used to be a few years back and who knows
maybe even so today one could go to a
newsstand and buy a magazine with nothing but
trends.

Anything and everything was covered, for
example if Tom Landry- this is how long since
I've checked- suffered double-digit defeat he
was something like 72% likely to cover the
spread in the revenge game.

How could enough material come up to keep
such a yearly publication filled was always
something to ponder. Going with the Landry
angle, it was always in the qualification ie
this was tracked "since winning a super bowl"
or some other convenient yardstick.

There are websites out there for free or fee
which have daily trends ie "Team A is 2-5 the
game after scoring 7 or more runs" or
whathaveyou. Interesting but not really a
large enough sample to be conclusive. One
always beware though, for if Team A is 2-5 in
the previous 7 games they may be 5-5 overall
and thus no advantage.

Draw the line like so and the trend is potent, draw it somewhere else and it becomes
meaningless.

Well having said that an absolutely fabulous
trend does come to mind and that is one of
the Steelers never having covered a game in
the state of Florida, going something like
0-9. Enjoy!

[zippy]

I’ve always been highly skeptical of the relevance of the “Dome teams playing a Monday Night road game on artificial turf after three straight losses beat the spread 80% of the time” type of trends. Flip a coin a few dozen times and it will rarely come out exactly 50% heads and 50% tails, but obviously any “trend” towards heads or tails here is of zero value in predicting future outcomes. (If you allow yourself to pick and choose the starting or ending points of the run, as touts and losing gamblers routinely do, then variations from 50% are even more prevalent, and every bit as meaningless.)

My “angle” is to do hundreds of hours of research a year on the NFL teams. I read several of the preview magazines that come out in the summer, scouting reports on the rookies, periodicals like “Pro Football Weekly,” newspapers and various other sites on the Web, etc. Then I assign point values to every player on every team. (Quantifying player values like that is admittedly subjective and imprecise, but I’m convinced that it can be done in an informed enough way to have some merit.) I give different weight to the different player numbers based on the position they play, whether they’re a starter or reserve, etc. I add or subtract a little bit based on intangibles such as the quality of the teams’ coaching, which team is playing at home, etc. I try to stay on top of whose points need to be removed or reduced because they are out with an injury or playing injury-impaired. Then I total it all up to determine which team is better and by how much, and I compare that with the available betting lines.

So the only “trend” I see much merit in is, “The team with the better players wins more often than not.” Following this method over the years, I’ve done significantly better than chance, though I’m certainly not getting rich.

[WhiteShoes]

one measure of handicapping instinct I like
to compare is time spent handicapping and seeking information vs. time spent perusing games on TV and radio. It takes all kinds
though as watching games on TV can lead to insights one wouldn't have otherwise. There's too much to do in reality, so that one must pick and choose his road. (methods,
time management, articles to read, games to watch)

imo, team stat trends can be meaningful when looked at as, "recent performance" during
one season. Leaguewide stat trends are hard
to fathom in any one year, and can be looked
at as aberations, (ats) (teams with higher turnovers win ATS, teams with more yards per
carry win ATS) since lines can be adjusted.

Motivational, emotional trends can be harder to pick out (anti-number crunching) but are thus more useful to the avg. handicapper, imo. Knowing which apply better to specific teams is one thing that can also be faded or
added onto a line by the linesmakers.

as

[mls]

I read that NFL teaser trends have consistently won for several years. What teaser trends?!?

[reno]

mls, as I've stated before, I believe that 2-team and 3-team 7 point NFL teasers, at the odds offered by ABC and Milennium, offer sharp bettors excellent value and opportunity.

However, I'm not aware of any trends or angles regarding teasers. If you know of some, please fill us in.

[BBKing]

It seems to me that as long as a trend isn't TOO good, it doesn't seem to eat itself up with extra action. There are a lot of contrarian trends that have hit 55% or so over hundreds of trials (since 1983 or so). They seem to fit these necessities:

1) they capitalize on public bets on the favorite (as evidenced by long term 'dog bias)

2) They are not so great as to interest the average bettor, only the "investor".

Any of those in the know, why would these trends extinguish themselves? If they have proven themselves statistically over long periods of time and show no signs of slowing (due to publication or rule changes), what would make them stop paying?

I'm not talking about short term stuff, but trends that go over 100 trials, and usually over 200.

[reno]

BBKing, I gotta tell you that the thrill is gone for me regarding betting trends in the NFL. Seriously, my take is that once you get down to 55% or below as a win percentage, you better have considerably more than 100 plays of data.

I checked a lot of angles that were 54%+ for over a 100 plays. And in the much longer haul, they failed. Going against line movement and getting value the other way is not a trend or angle.

[ajax50]

IN MY HUMBLE OPINION THE ONLY WAY TO BET FOOTBALL,OR ANY OTHER SPORT, IS TO GET THE BEST POSSIBLE NUMBER AVAILABLE. IF A GAME OPENS 4 AND DRIFTS UP DURING THE WEEK TO 6 OR 6.5,YOU CAN ONLY BE ON THE FAVORITE IF YOU LAID 4.IF YOU LIKE THE FAVORITE AND YOU MISSED 4 ,I BELIEVE YOU HAVE TO PASS ON THE GAME. THE SAME THING GOES FOR BEING ON THE DOG.IF YOU DONT GET THE HIGHEST NUMBER AVAILABLE ,YOU CANT BE ON THE DOG.LINE HISTORY IS VERY IMPORTANT IN BETTING FOOTS. THAT IS MY "HANDICAPPING". P.S.PRAY THAT YOU STAY ALIVE BETTING FOOTBALL SO YOU CAN FIRE DURING BASEBALL SEASON.FOOTBALL IS NO BARGAIN EVEN WITH MANY OUTS ,LAYING 05,ETC.

[chestrockwells]

Before I say anything, let me say that I am extremely sceptical of trends/systems etc,etc.. If Reno or someone cares to follow up on this it may prove educational. Last year in the NFL if you took every weekly high total and went over as well as every weeks low total and went under that number your record would have been 31-14-2. Now, before everyone suspects my ability to add simple numbers, here's how I came up with it. If there are two 49's like this week you'd play both. Ditto for the unders. And this is based on the closing line as per Gold Sheet's stats. So that was 47 plays over the regular season last year. Any comments, or can anyone go back 4-5 yrs. on this?

[BBKing]

chestrockwells-

your numbers seem to go against the "regression to the mean" idea that comes up every so in my handicapping. I'd like to check those numbers back a few years but don't have access to my data (I am at college)

ajax-

It still has to come down to value. If there is value in laying a higher number (due to injury, for example), it may make sense. I sure hate to lay those extra points, though. I feel like I missed a sale at the mall or something.

[Count Zero]

I have checked nearly 500 such trends against my database since the early eighties, and have only about 20 that I even pay attention to. I use these modify the number my sets of power ratings produce, but in most cases I do not give them much weight. To decide if the angle's winning percentage is statistically reliable, I divide (covers minus did not covers) by the square root of the number of trials (covers plus did not covers). If the difference is 2 or more, I assume no more than a 5% difference either way from the given winning percentage. This is a trick (?) I picked up as a blackjack player years ago. Don't know if it makes sense, but it seems to work.

The MNF angle I use is: Bet the home team if they are dogs or favs by less than 6, except don't bet divisional favorites. This is better than 60% since 1983, but has been on a downtrend lately. I think it's still usable; even last year, there were plenty of posters with 2 or 3 years of experience in BH slamming the whole idea, which usually means expect it to pick up again soon. There are a huge number of unsophisticated players on Monday night.

I notice no one has bothered to put up a winning angle, so I will give you mine, for what it's worth: Find posters here or at other internet forums who post their picks in all-caps with lots of exclamation points. Going against these players is pure gold. Example:

nyg + 5.5
oak + 9
atl/min under 49

This guy might know what he's doing. No play. But

====>>>>>>>> GOOFY'S PIX!!!!!! <<<<<<<<===

1,000,000 * DENVER MINUS ANYTHING!!! WILL kill 'em!!!!! I LOVE THIS GAME !!!!!!!!

100,000 * GREEN BAY MINYUS WHATEVER !!! THE PACK IS BACK!!!! THIS WILL BE OVER BY HAFTIME

**************** MINN/ATL OVER!!!!! FINAL SCORE MIN 75 ATL 60!!!!!!!!!!!!!

GOOD LUCK!!!!!! ------ I AM DA MAN!!!!!!!!!!

-------GoofyGoofyGoofyGoofyGoofy------------

This is the guy to fade. No one with a brain who has played NFL sides for more than 10 minutes feels like this. This guy will make every heart bet, fall into every trap, etc. Admittedly, there is some subjectivity involved -- not everyone will derive the same picks -- but last year I caught 16 winners and only 7 losers. Not much to go on, but the rationale is faultless.

I would also like to mention that there are more sophisticated ways to study angles than just running down the W-L. The game changes, especially re points scored, and many angles can be dramatically improved by using a formulation that recognizes this. There is at least one well-known totals angle that has fallen into decrepitude over the last 5-10 years, but which is doing better than ever when recast to acknowledge the changing realities in the NFL. If you have a database, look into it.

[Jeff]

another classic

[j]

Count Zero,

How good to have you back. Hope you'll post a bit during the season...