|
|
Falcon Past Performance Sheets
Taking a concept from Horse Racing and applying it to pro football, the Falcon PP's aim to be the next big thing in football handicapping, and come complete with "NFL speed ratings" and game log stats.
We'll give some free samples during the year, but to see the Falcon PP's for every game you will need to subscribe.
|
Falcon PP's -- your ticket to success?
Maybe you're a good handicapper already. You make some money. Maybe you shop for lines and make a bit more. What if you could improve your base win percentage though, wouldn't that be worth doing? That's our goal -- to make you a better handicapper. Our PP's have performance figures that adjust for the quality of opponent, location and other factors to provide a one-number rating in a given game, ala a horse racing speed rating.
A longtime subscriber electing to go by the name of "The Falcon" has been working on a project to adapt conventions in horse racing to an analagous pro football prediction scheme...
Play the ponies? I do.
Heard of Andy Beyer? He's only the demigod of horse racing speed figures. Followed by Len Ragozin and a string of imitators.
One of Andy's famous quotes is that Speed Figures are for horse race predictions "the truth, the way, and the light."
Another famous quote is that "The race doesn't always go to the swiftest, or the fight to
the strongest, but that's the way to bet!"
My goal is simple, to create "Speed Figures for the NFL."
I want one number that tells me how well a team played in a game. I want PP's (Past Performances) for NFL teams that have the same structure as horse racing PP's you'd find in a Daily Racing Form. I want to use the same lingo and theories of horseplayers: new top, bounce, course specialist, bad trip. I am convinced the road to riches follows along a similar path in football that it does with horses.
Here is what I have in mind:
Week 13, 2004: Carolina at New Orleans (-1.5)
--> Carolina needs a "94" Fig to cover, New Orleans needs a "100" Fig to cover.
Carolina
| Week |
Opp |
Line |
For |
Agn |
Team
Pwr |
Opp
Pwr |
Opp
Fig |
Team
Fig |
| 12 |
TB |
+2.5 |
21 |
14 |
98.4 |
100.8 |
94 |
104 |
| 11 |
ARI |
-3 |
35 |
10 |
94.9 |
100.2 |
72 |
122 |
| 10 |
@SF |
+1 |
37 |
27 |
92.4 |
88.2 |
79 |
101 |
| 9 |
OAK |
-6.5 |
24 |
27 |
92.0 |
89.1 |
98 |
83 |
| 8 |
@SEA |
+8 |
17 |
23 |
92.3 |
104.5 |
95 |
101 |
| 7 |
SD |
-3 |
6 |
17 |
93.3 |
104.3 |
107 |
90 |
| 6 |
@PHI |
+8.5 |
8 |
30 |
96.8 |
115.8 |
116 |
97 |
| 5 |
@DEN |
+5.5 |
17 |
20 |
97.2 |
106.5 |
97 |
107 |
| Average |
+1.6 |
21 |
21 |
94.7 |
101.2 |
95 |
101 |
Away Ave. Fig = 102 | Common Opp. Ave. = 101
|
New Orleans
| Week |
Opp |
Line |
For |
Agn |
Team
Pwr |
Opp
Pwr |
Opp
Fig |
Team
Fig |
| 12 |
@ATL |
+9.5 |
21 |
24 |
91.8 |
99.0 |
92 |
99 |
| 11 |
DEN |
+4 |
13 |
34 |
92.6 |
106.4 |
117 |
82 |
| 10 |
KC |
+4 |
27 |
20 |
90.6 |
105.2 |
87 |
109 |
| 9 |
@SD |
+6 |
17 |
43 |
93.7 |
108.9 |
117 |
86 |
| 7 |
@OAK |
+3 |
31 |
26 |
92.6 |
91.0 |
85 |
99 |
| 6 |
MIN |
+3.5 |
31 |
38 |
93.4 |
103.9 |
103 |
94 |
| 5 |
TB |
-3 |
17 |
20 |
92.0 |
93.7 |
98 |
88 |
| 4 |
@ARI |
-3.5 |
10 |
34 |
97.8 |
93.0 |
119 |
72 |
| Average |
+2.9 |
21 |
30 |
93.1 |
100.1 |
102 |
91 |
Home Ave. Fig = 93 | Common Opp. Ave. = 85
|
Analysis:
In the last eight games Carolina has posted the "2Win" Figure SIX times. New Orleans has posted a 100+ fig only ONCE. Also the Panthers seem to be on the upswing with three straight hundred or better games.
Carolina only has one stinker while New Orleans has FOUR stinkers. Half the games shown on the PP's.
The home/away splits are big (102 for Carolina to 93 for New Orleans) and Carolina has done better against four of five common opponents.
Pattern lovers take note: every other game New Orleans gives up a monster Fig to its opponents (look at Opp.Fig) and guess what that suggests?
FAVOR CAROLINA +1.5
|
This is how I want to handicap, and through the powers of the TwoMinuteWarning group, I'll be getting my wish this season.
I have had to pass a rigorous series of tests to prove predictive value for this new method, and thankfully the Falcon PP idea has lived up to my expectations.
How to use the Falcon PP "Speed Figures"
There are many ways I believe one-number performance ratings on a game by game basis can be helpful. The most obvious approach is to simply use the average figure a team earns. The notion of "average speed rating" is very familiar to horse players so let's check it out on the football side. Take the Average Speed Rating for the appropriate data range and compare it to the other team's speed rating with a three point home field advantage.
Example from above: Carolina last 8 games average is 101, New Orleans last eight is 91, so the 'Falcon' prediction is 101-91-3 = Carolina by 7 points. With the line at NO -1.5 that means Carolina has an 8 1/2 point edge versus the spread.
| Data Set |
All Games |
Win% |
All Underdogs |
Win % |
4 Pt Edge Dogs |
Win % |
| Last Game |
772-754 |
.501 |
353-323 |
.522 |
284-270 |
.512 |
| Last 2 Games |
759-730 |
.509 |
350-306 |
.533 |
255-233 |
.522 |
| Last 3 Games |
692-681 |
.504 |
313-272 |
.535 |
217-178 |
.549 |
| Last 4 Games |
629-640 |
.495 |
280-260 |
.518 |
164-145 |
.530 |
| Last 8 Games |
457-390 |
.539 |
186-133 |
.583 |
86-49 |
.637 |
| Season Average (min 4 gms) |
657-613 |
.517 |
258-210 |
.551 |
99-72 |
.578 |
| Season Average (min 8 gms) |
455-392 |
.537 |
180-129 |
.582 |
60-37 |
.618 |
| Home/Away specific (min 8 total gms) |
445-402 |
.525 |
216-175 |
.552 |
137-99 |
.580 |
Using a small number of games to calculate an average is not very good, but when you get to the "Last 8" average it is powerful. You see the same thing in horse racing. A 'last 10' speed figure average will cream a last race average or last few races average in ROI (return on investment). We're batting .539 overall with the Falcon PP's but more importantly .583 with dogs only and .637 with Dogs with a 4 point edge. That grades out as a fine handicapping tool and would alone qualify the Falcon speed figures for your attention.
Yet this is such a simplistic way to use this information. Why stop there? We're talking an average season of 27-19 with all dog plays, 12-7 with 4 point edge dog plays. Good enough for you maybe, not total satisfaction for me.
Notice in the analysis for the sample game above that I refer to several things:
- how often a team has hit the "2win" number for the matchup
- how teams have done against common opponents
- patterns that are in effect for the game
And there's still more to look at:
- average line faced
- average opponent Fig yielded
- power rating trends over time
- Fig ratings vs outcome (losing with a big fig, winning with a low fig, etc.)
Find out more:
What the press says about TwoMinuteWarning
Comparing NFL Handicapping Services
