Announcement

Collapse
No announcement yet.

NFL Goodies...a Re-Post

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • NFL Goodies...a Re-Post

    .
    .
    .
    Here's a re-post, in the proper forum, of what I posted a few weeks ago. Comp is at bottom.


    With the NFL Season right around the corner, I want to share with the forum some really cool stuff I have worked on during the off-season. Most of it is a hybrid of statistical handicapping and technical handicapping, although some of it is situational. This is all new material I developed by experimenting with my Excel Program, and the numbers are pretty amazing...so far.

    NOTE: The results represent what I consider to be the meat of the season; namely, Weeks 7-15 (nine weeks). The results are from the last 4 NFL seasons, 2005-2008, and are all Against the Spread.

    Why Weeks 7-15? I believe this is the heart of the season. Once 6 Weeks have been played, each team has at least 5 games under their belt. I think this is the point at which the data is sufficient, which makes the lines more accurate. Five games gives each team the opportunity to play at five levels:

    Their best. Above average. Average. Below average. Their worst. Once the first 6 weeks are past, then I believe technical handicapping aspects are less vulnerable to bad lines.

    As for the first six weeks, I think a statistical form of handicapping is at it's best and probably the best way to go as you can exploit bad lines early.

    Regarding Weeks 16 and 17, I have made it a habit to treat the last two weeks differently as teams rest players, jockey for playoffs and home field advantage, and play spoiler roles. The lines are often significantly adjusted for these factors, and some teams are looking forward to the off-season. This blunts both technical and statistical handicapping to some degree.

    Before I indulge the new stuff, here's two of my oldies and all time faves. They hardly ever come up, but it's worth mentioning in the event that they do appear this season.

    1.) Go AGAINST any NFL team that scored 30 or more points in each of their last two games while allowing 10 or less in each of those same two games...if they are now playing a team outside of their division.

    The rationale is simple. In the NFL, after a team "peaked" offensively and defensively back to back, they tend to let down when faced with an opponent of lesser significance.

    2.) Go with a team that is .500> if they are at home off 3 SU losses and played at home last week.

    This rarely comes up, but is effective. Following the train of thought goes like this: A winning team has lost 2 games and now they are at home in an environment where you would expect them to bounce back. But they don't. The following week they are at home again and, despite their 3 losses, are at least still .500 or better. This is a good spot for a rout, especially if they are favored.

    That's a couple oldies but goodies to look for this season. If we're lucky, we'll get 1 play from each of them...and two winners if we're doubly lucky...

    gl,
    dave

    .
    .
    .
    General observations, methods, and thoughts.

    POWER RATINGS. Although far from perfect, my method of figuring out, fairly accurately, who's who in the NFL is simplistic but gets the job done.

    I like to look at a team's RANK in 6 primary categories:

    Points For Rank (PFRK)
    Points Against Rank (PARK)
    Offensive Yards Rank (OYRK)
    Defensive Yards Rank (DYRK)
    Net Points Rank (NPRK)
    Net Yards Rank (NYRK)

    The Power Rating formula for each team is:

    The ABSOLUTE VALUE OF: ((PFRK + PARK + OYRK + DYRK + NPRK + NYRK)/6-100)*.6774-31.9368

    I won't bother going into the how's and why's, but the highest possible PR is 99 (if a team is #1 in all six categories) and the lowest is 78 (if a team is #32 in all six categories).

    The largest possible line on a neutral field is 21, but it's highly unlikely. In reality, pitting the highest PR against the lowest PR in a given season will put the line at around 17-18 on a neutral field.

    Power Rating Rank plays a key role in some of the following systems, so it is important to note that the systems are based on MY power ratings as formulated from the method above. Obviously, ATS Results would vary if using a different Power Rating Formula.

    So, in some of the following systems, you'll need a Power Rating Rank for each team.

    OFFENSIVE YARDS PER POINT. You'll also need a OYPP Rank for each team. OYPP is Offensive Yards Per Point. While Power Ratings determine strengths and weaknesses of teams, as well as pointspreads, OYPP is widely believed to be one the best statistics when considering ATS results. The reasons behind that are many, which I won't go into. In short, many believe that no other statistic encompasses the whole of team efficiency as does Offensive Yards Per Point.

    So I decided to marry these two (PR Rank and OYPP Rank) and run a full range of possibilities through my Excel Program. In addition, I factored in the technical aspect of capping (i.e., cause and effect--as opposed to situational), and the results were eye opening...for now of course.

    MAKING LINES. This is not my forte', but there are a number of ways to guage a line. One, of course, is to simply use Power Ratings. Another way is to make a projected score for each team. One of the easiest ways to do this is as follows:

    Let's say Arizona is playing at Detroit. To get Arizona's projected score, you would divide their OYPP into Detroit's Defensive Yards Allowed Per Game. You would then subtract 1.5 for home field. Do the same with the Lions, only ADD 1.5 to their projected points scored, and now you have a projected final score.

    TIDBITS

    TEAMS OFF A BYE are 32-12 ATS* as favorites.

    DOUBLE DIGIT FAVORITES are 15-30 ATS* if they played last week and covered.


    *Results are from Weeks 7-15 (nine weeks) over the last 4 NFL Seasons, 2005-2008. This will be the case for all remaining systems.


    gl,
    dave

    .
    .
    .
    Terminology and breakdowns.

    SITUATIONAL HANDICAPPING. Teams do certain things in certain situations. Not individual teams, but all teams as a whole.

    Example: Double Digit Faves off a Cover are 15-30 ATS. No real specific cause and effect, they just don't do good in that "situation" after week six.

    STATISTICAL HANDICAPPING. It's all about the numbers. Forget cause and effect. Forget situations. Forget what they did or did not do last week. It's all about the numbers.

    Example: Team X is has a Top 12 PR Rank and a Top 12 OYYP Rank. This is all numbers. They are solid bets at home if not playing a team within the same criteria. It's all about the numbers. But be careful. You don't want to lay double digits based SOLELY on the stats.

    Take this Top 12/12 Team and mix in a Top 6 Rank Defense (yards allowed) playing against a team in the Bottom 6 in OYPP, and you've got a complete MISMATCH, a Blowout waiting to happen. This, again, is all about the numbers and does not factor in cause and effect or situation.

    TECHNICAL HANDICAPPING. This is cause and effect. The history of technical capping is that, by and large, it throws out the numbers. It often uses contrary ideology, let down and bounce back theory. Used by itself, it can be blind to the obvious.

    EXAMPLE: Go against a team that scored back to back 30-10 wins, or something along those lines.

    Techs can hurt you early when the lines are soft and hurt you late when teams give up, but it's a necessary piece of the puzzle.

    BREAKDOWNS

    Some of the breakdowns I'll look at are:

    Top 6/Bot 6 (both PR and OYPP)
    Top 12/Bot 12 (both PR and OYPP)

    Why 6? Why 12?

    I have broken down the 32 teams into Five categories. The Best, Above average, Average, Below average, The worst.

    So teams that are in the Top 12 (the best, and above average), might show different results than the Bottom 12. Not necessarily better results, just different.

    As you'll see, there is a place where even bad teams do good and good teams do bad.

    Other breakdowns include the obvious and the usual:

    Home, Road, Fave, etc.; Off Win, Off Loss, Off ATS Win, and so forth.

    Added dissections include teams with 350> offense, <300 Offense, and the same with defense, along with 350/300, and 300/350. The rest of the teams that don't fit in go into the "Tweeners" category.

    VALUE DIFFERENTIAL. The most intriguing category that I added this year is the Value Differential. What is that?

    I got to thinking. If the LINE is based on the POWER RATING and OYPP is often indicative of ATS results, then a team would be getting value in the line if their OYPP was better than their PR.

    Think about it. I noticed the Browns from a couple years ago. They were #5 in OYPP (good team bet on). If that weren't good enough, they were "under the radar" with a Power Rating Rank of #22 in the NFL. This means the line you got with the Browns was based on them being ranked #22. That's value.

    Running the numbers, it did not surprise me that BOTTOM 6 teams in POWER RATING (the worst teams in the league) were nevertheless 37-20 ATS...

    ...IF...

    ...they had a positive value differential of 10 or more, V-DD+, which stands for (VALUE: DOUBLE DIGIT +).

    In other words, if a team in bottom six in power ratings has a OYPP Rank which is 10 or more spots higher than their PR Rank, there's value...even with the worst teams in the league. Keep that in mind when I post where and when to go against the best defenses in the league (18-43 ATS). Like Solomon said, there's a time for everything under the sun...

    37-20 ATS

    gl,
    dave
    __________________

    .
    .
    .
    What if?

    What if a team's PR RANK dropped by 3 or more based on last week's performance? Increased by 3 or more???

    What about their OYPP Rank? How would they respond if the effects of last week's game was that their OYPP dropped by 3 or more? 2 or more?

    What if a team's PR Rank AND OYYP Rank dropped?

    What if one dropped and the other increased? Not likely, but possible. Vice versa? How would they respond?

    What are the supposed psychological effects, if any, of outplaying your opponent to the extent that your PR increases by 3, yet you lose the game due to lack of efficiency, turnovers, special teams, etc. (OYPP drops)?

    What if a team's net yardage last week was 100 or more yards BELOW their season average? Above?

    What if they they got blown out by 18 or more points in that same game? What would happen the following week? What if they had a Positive Value Differential the following week in addition to last week's embarrassment?

    I've spent the off-season constructing this excel program, inputting data from the last 4 seasons, and then asking...

    ...WHAT IF?

    These are not generic what ifs, but creative what ifs. How do teams respond if a given performance affect their season numbers by x-amount?

    Here's a couple of softballs followed by a trailer of what's to come.

    Teams with a Top 12 PR Rank (the best teams, above average teams) are 56-27 ATS if their performance last week dropped their net yardage average by 10 or more yards.

    They are 29-12 ATS Home, 18-6 as a Dog, and 19-5 if they have a double digit negative value differential (V-DD-). That said, the recommended play is the category of ALL, which is 56-27 ATS.

    What if a team's net yardage average INCREASES by 10 or more yards? Remember, to move the AVERAGE by 10 yards or more in one game takes a significant effort. Apparently the DEFENSE plays a key role in that effort as evident by the following week's results.

    Teams whose defense is allowing <300 yards per game fall flat the next week, going only 18-43 ATS. As Faves they are a mere 10-33 ATS. Recommended play is the category of: AGAINST AS FAVES.

    Thus far, that's 272 plays, or 68 per year, or about 7.5 picks PER WEEK on average over the nine weeks spanning from Week 7 to Week 15. Obviously there will be a handful of picks that either coincide or contradict. But that's a LOT of action, and there's more to come.

    But if you like less plays, you'll love what's coming. The best of the best is 36-3 ATS, including a 15-0 ATS record on Game of the Year Status Picks. These GOY picks average COVERING by nearly 18 points per pick.

    If you like my DOUBLE 30-10 PEAK system, which I post every year online, you'll love these games. I'll go on record to say it's the best I've developed in 29 years of handicapping. Plus, the system puts out more than 1 pick per year (as is the case with dbl 30-10). It puts out 1 pick per WEEK...on average. It's put out 39 picks in 36 weeks with a 36-3 record. Hopefully the results continue to do well.

    Here's the Game of the Year Status Plays, which are 15-0 ATS with an average cover of nearly 18 points per game:


    2008

    Car -3.....30-7
    Ne -3.....41-7
    Hou -5'.....28-21
    Bal -1.....36-7
    Chi -7.....27-3
    Ten -11.....47-10
    Phi -3.....48-20


    2007

    Grb -5'.....34-0
    Min -4'.....29-22
    Car -3.....31-14
    Cle -4'.....8-0


    2006

    Pit -9.....20-3
    Atl -1'.....24-14
    Ne-12'.....40-7


    2005

    Nyg -1.....36-0


    gl,

    dave

    .
    .
    .
    With a myriad of queries, perhaps the one I was most interested to know about was how teams did after dropping/increasing x-number of spots in OYPP RANK. The numbers were intriguing and solid ATS, but they took back seat when I began mixing in PR Ranks in conjunction with OYPP Ranks. Here's what I found:

    Teams that dropped 3> spots in both their PR Rank and their OYPP Rank responded 28-12 ATS the following week, while teams that did the opposite responded by going 15-33 ATS in their next game. This brings the running count to 360 plays, a whopping 10 plays per week before cancel outs and duplicates. The W/L units record is thus far 249-111 (+126.9) before cancel outs, or a conservative projection of over 32 units per season on average, so far.

    gl,
    dave

    .
    .
    .
    Teams that win by 18 or more points and their net yardage is 100 or more yards greater than their season average are 28-40 ATS. Teams that do exactly the opposite are 44-25 ATS.

    Regarding the former, there are no significant breakdowns, and at 59% when going against these teams, and a sampling of 68, I'm gonna pass. However, it is a nice compliment to:

    Teams that increase 3> spots in PR and OYPP. When combining them, teams are only 2-16 ATS. Road dogs get blown out, going 0-5 ATS while missing the spread by an average or 17.70 pts per game.

    Regarding the latter, which is a solid 44-25 ATS, there are a lot of breakdowns which catch the eye. So I'll have to come back with more cross references to insolate what works the best.

    Some of the more notable initial breakdowns include:

    30-11 ATS at home
    33-10 ATS if you discard teams that are in the Bottom 6 in both PR and OYYP.
    23-3 ATS if the team is +/- 10> in Value Differential.

    I'll have to break this down further in a reply of its own.


    gl,
    dave

    .
    .
    .
    HYBRIDS. Teams who dropped by 2 or more in both PR RANK and OYYP RANK, lost by 18 or more, and their net yardage was 100 or more yards below their season average, are 22-2 ATS in their following game. Favorites average covering by nearly 18 ppg whether on the road or at home.

    Combining this with the hybrid posted just above, the total is 38-4 ATS. There was one cancel out, reducing the record to 37-3, and one duplicate, reducing the record to 36-3 ATS.

    It's important to note that with all of these tech plays, they must have played last week. The point is that they are responding, one way or the other, to last week's performance.

    So if a team meets the criteria, say, for a PLAY ON in their next game, but then has a BYE after having met that criteria, you would not suspend the play for 2 weeks. You simply would not have a play. The team must have played last week, and is now playing this week, when it comes to these technical systems; which makes sense to me anyway.

    gl,
    dave
    __________________




    POWER PLAY Game by Game Results:

    2008

    Week 7
    Car -3 Win 30-7 (BLOWOUT PICK)
    Nwe -3 Win 41-7 (BLOWOUT PICK)
    Bal +3 Win 27-13
    Grb +2' Win 34-14
    Nyg -10' Win 29-17
    Oak +3 Win 16-13
    Ten -8' Win 34-10
    Sea +11' Win 10-20

    Week 8
    Stl +8' Win 16-23
    Nyg +2' Win 21-14
    Nor +3 Win 37-32

    Week 9
    Min -5' Win 28-21 (BLOWOUT PICK)

    Week 10
    Sea +8 Win 19-21
    Ind +3 Win 24-20
    Sdg -15 Lose 20-19

    Week 11
    Ten -2' Win 24-14
    Nor -6 Win 30-20
    Cin +9 Win 13-13

    Week 12
    Bal -1 Win 36-7 (BLOWOUT PICK)
    Chi -7 Win 27-3 (BLOWOUT PICK)

    Week 13
    Ten -11 Win 47-10 (BLOWOUT PICK)
    Phi -3 Win 48-20 (BLOWOUT PICK)
    Pit +2 Win 33-10

    Week 14
    Det +10' Win 16-20
    Cle +13' Lose 9-28
    Ari -15 Win 34-10
    Chi -6' Win 23-10

    Week 15
    No Plays


    2007
    Week 7: 5-0
    Week 8: 1-0
    Week 9: 1-0
    Week 10: 5-0 (BLOWOUT PICK Grb -5' Win 34-0)
    Week 11: 3-0 (BLOWOUT PICK Min -4' Win 29-22)
    Week 12: No Plays
    Week 13: 2-0 (BLOWOUT PICK Car -3 Win 31-14)
    Week 14: 5-0
    Week 15: 6-1 (BLOWOUT PICK Cle -4' Win 8-0)

    2006
    Week 7: 5-0
    Week 8: 5-0
    Week 9: 3-0
    Week 10: 2-1
    Week 11: No Plays
    Week 12: No Plays
    Week 13: 5-0 (BLOWOUT PICKS Pit -9 Win 20-3; Atl -1' Win 24-14)
    Week 14: No Plays
    Week 15: 4-0 (BLOWOUT PICK Nwe -12' Win 40-7)


    2005
    Week 7: 3-0
    Week 8: 3-1 (BLOWOUT PICK Nyg -1 Win 36-0)
    Week 9: 3-0
    Week 10: 2-0
    Week 11: 3-0
    Week 12: 0-1
    Week 13: No Plays
    Week 14: 1-1
    Week 15: 2-0

    4-Year Total: 94-7 ATS

    Another Free System: NFL XL PIX
    Contact: [email protected]

    gl,
    dave

    Comp: 49ers OVER 7.5 Wins -115

  • #2
    bump

    Comment

    Working...
    X