Welcome to the 3rd NFL PT Awards. The purpose of these
awards is to help raise the publicity for those systems that
have been superior in various qualities of interest.
Most awards will be based entirely on the numbers I've
monitored with my prediction tracker web pages. In 2002 I
followed the weekly performance of 42 computer systems rating
the NFL.
BEST STRAIGHT UP WINNERS (Entire Season)
Winner: Matthews Grid, Herman Matthews
Herman Matthews' grid system was the leader in correct picks this season.
Matthews' record was 172-94, (64.66%). This was three games better than
the two systems that tied for second place.
There seems to be a downward trend over the years. Our prediction records
are getting worse each season. Last season ARGH was at 65.2%, the year
before that CPA was 66.4%, and the year before that the Vegas line was 67.7%.
This trend seems to be following the rise of mediocrity in the NFL.
2002 Winner: Herman Matthews
2001 Winner: JFM Power Ratings, ARGH Power Ratings
2000 Winner: CPA Rankings
1999 Winner: Kenneth Massey Ratings
BEST AGAINST THE SPREAD (Entire Season)
Winner: Free Sports Plays
This might not be the most useful measure of the quality of a football
rankings system but betting on the NFL is big business. This season
the best system picking against the spread came from FreeSportsPlay.com.
FreeSportsPlay had an excellent record of 126-94 (57.27%). This is up
quite a bit from CPA's 53.75% last season. It also sets a new record.
This was actually a very good season for systems against the spread.
Two of Jeff Sagarin's ratings were over 55%. My logistic regression ratings
actually matched FreeSportsPlay but were only measured for the second half
of the season.
winner 2002: FreeSportsPlay.com
winner 2001: CPA Rankings
winner 2000: Yourlinx
winner 1999: PerformanZ Ratings
SMALLEST MEAN ABSOLUTE ERROR (Entire Season)
Winner: Free Sports Plays
This is big news folks. For the first time ever we have a system win
this award and outperform the Vegas spread. FreeSportsPlay.com become
the first system, college or pro, to do this. Kenneth Massey also gets
a very special honorable mention because he also joins FreeSportsPlay
in beating the line. Congradulations to both of those systems. FreeSportsPlay
wins the award with an average error of 10.33 and Massey was right behind him
at 10.36. This season the line comes in 3rd at 10.44. The all time record
is 10.28 set by the line in 2001.
winner 2002: FreeSportsPlay.com
winner 2001: Vegas Line
winner 2000: Vegas Line
winner 1999: Vegas Line
SMALLEST BIAS (Entire Season)
Winner: Sagarin Points, Jeff Sagarin
The smallest bias over the entire season belonged to Jeff Sagarin and his
Predictor or Pure Points system. This system had a bias of -0.0229. This
means that on average Sagarin's predictions were 0.0229 points to low.
Sagarin also did well in this category last season coming in second.
This sets the record for full season bias. This was slightly better than
Pigskin Index's 0.0270 mark in 2000. I think overall we were much improved
this season as a whole.
winner 2002: Sagarin Pure Points
winner 2001: ARGH Power Ratings
winner 2000: Pigskin Index
winner 1999: Flyman Ratings
MOST ACCURATE SYSTEM - SMALLEST MEAN SQUARE ERROR (Entire Season)
Winner: Free Sports Plays
I never realized how good of a season FreeSportsPlay.com was having but
they end up taking home three out of five full season awards. FreeSportsPlay
is deemed the most accurate system of the year for having the lowest
mean square error, 179.3. This is actually the least accurate winner
that we have ever seen. Last season the Vegas line won with a mark of
173.5. Kenneth Massey was second and previous winner Pigskin index was
3rd. These three systems were the only ones to outperform the line over
the full season.
winner 2002: FreeSportsPlay.com
winner 2001: Vegas Line
winner 2000: Pigskin Index
SECOND HALF AWARDS
I like to look at the results over the second half of the season. The
systems have had time to become 'burned in' to the season's data.
Second half data consists of all games from week 10 through the superbowl.
BEST STRAIGHT UP WINNERS (Second Half)
Winner: Least Absolute Value Regression
Winner: JFM Power Ratings, JF McConnell
Winner: CPA Rankings, Steve Wrathell
Interestingly, we have a three way tie for the second year in a row and a tie
for the fourth year in a row. there have been 10 different winners over
the last 4 seasons. This year Least absolute value regrssion,
JFM Power Ratings and CPA Rankings tie for first place with a 94-42 record,
(69.12%). This is almost an identical record as the three leaders of last season.
For the second straight season Jeff Self put up great numbers but did
not have enough weeks to qualify.
Winner 2002: Least absolute value regression, JFM Power Ratings, CPA Rankings
Winner 2001: Pythagorean Ratings, Elo Ratings, Ken Massey Ratings
Winner 2000: PerformanZ Ratings, Scoring Effeciency
Winner 1999: Ken Massey, Ed Kambor, Pythagorean Ratings
BEST AGAINST THE SPREAD ( Second Half)
Winner: Least Absolute Value Regression
The second half of this season saw some amazing records against the
spread. Least absolute value regression had the best mark 78-52, (60%).
While seven other systems all finished better than 57%. The highest
percentage seen before was 56.25% which had been the winning number every
season until this one.
Winner 2002: Least absolute value regression
Winner 2001: Stat Fox
Winner 2000: Pythagorean Ratings
Winner 1999: Least squares using team specific home field advantages
SMALLEST MEAN ABSOLUTE ERROR (Second Half)
Winner: Free Sports Plays
FreeSportsPlays.com dominated this category this year with a mean error of
9.692. The closest behind was Ken Massey and 9.952. Evidently the line really
struggled the second half as seven systems managed to beat the line. The Pythagorean
ratings had an excellent score for the third season in a row and is one of the
consistantly good systems in this category. The line in 2001 has the record
in this category with 10.3588.
Winner: 2002: FreeSportsPlays.com
Winner: 2001: Vegas Line
Winner: 2000: Pythagorean Ratings
SMALLEST BIAS (2nd Half)
Winner: StatFox.com
It looks like the awards this season are mostly going to the dot coms.
The ratings of StatFox.com had the smallest bias (-0.146) over the second
half of the season. Overall, the numbers here were terrible. CPA Rankings
and Flyman were close behind the leader but most were off by over a full point.
During the second half we generally gave a point too few to the home team.
In three of the four years this has been the case. Home field seems to play
a bigger part later in the season. Flyman has been a consistant performer
in this category, having another good second half this year.
Winner 2002: StatFox.com
Winner 2001: Monte Carlo Markov Chain
Winner 2000: Flyman Performance Ratings
Winner 1999: Flyman Performance Ratings
MOST ACCURATE SYSTEM - SMALLEST MEAN SQUARE ERROR (2nd Half)
Winner: Free Sports Plays
FreeSportsPlays gets another award for accuracy. FSP had a mean square
error of 172.845. This is a significant lead over several other systems
all around 179. This was the highest total to ever win this award. The
best mean square error was 165.954 by Pythagorean Ratings back in 2000.
Grid Iron Gold did very well in this category, actually beating FSP by
a small amount but GIG did not publish ratings consistantly.
Winner 2002: FreeSportsPlay.com
Winner 2001: Hank's Power Ratings
Winner 2000: Pythagorean Ratings
BEST PREDICTIVE SYSTEM (Entire Season)
Winner: Matthews Grid, Herman Matthews
For best overall awards I gave points to a system in each
category based on how well they do. I then sum the points. Hopefully this
gives an accurate read on who is the best system across the board.
This year this award seemed to be more up for grabs than ever. The best
overall predictive system this year was Herman Matthews' Grid system.
1. Matthews Grid 71
2. Sagarin Predictive 60
3. Yourlinx Retro 60
Winner 2002: Matthews Grid
Winner 2001: CPA Rankings
Winner 2000: Ed Kambour
BEST PREDICTIVE SYSTEM (Second Half)
Winner: Massey Ratings, Kenneth Massey
Unlike for the full season where nobody seemed to stand out from the crowd,
during the second half several systems really kicked into a high gear.
Kenneth Massey's system lead the way.
1. Kenneth Massey 115
2. Free Sports Play 106
3. Least absolute value regression 99
Winner 2002: Kenneth Massey
Winner 2001: Monte Carlo Markov Chain
WInner 2000: Pythagorean Ratings
RETRO AWARDS
The retrodictive categories are based on each systems final
standings and measure things by applying those final ratings
to the entire season in retrospect.
MOST RETRODICTIVE WINS
Winner: The Sports Report
The best retrodiction record goes to The Sports Report with a record
of 199-67 (74.8%). I'm giving this to their SLOTS system. Their main
system actually finished at the top of the list but it was based on a
modification made after the season. During the regular system I thought
The Sports Report was always way out in front so I was surprised to see
they did not set a new record. Last year the system average had the best
ever percentage at 76.1%.
Winner 2002: The Sports Report SLOTS
Winner 2001: The system average
Winner 2000: CPA Rankings
SMALLEST RETRODICTIVE ABSOLUTE ERROR
Winner: Least Squares with team home field advantages
Least squares using team specific home field advantage wins this award
for the third straight year. It has a winning average error of 8.76 points.
This was almost identical to last year's 8.77.
To date no body has really ever challenged this system in this category.
Winner 2002: Least squares with team specific home field advantage
Winner 2001: Least squares with team specific home field advantage
Winner 2000: Least squares with team specific home field advantage
SMALLEST RETRODICTIVE BIAS
Winner: PerformanZ Ratings, Todd Beck
It's always nice when I can claim an award myself because I don't
get very many of them. My PerformanZ Ratings had the smallest retro
bias this year at -0.01635. This year these numbers tened to range
from about .10 to .40. I'm not certain why more were not closer
to zero. Last year saw three finish below .01.
Winner 2002: PerformanZ Ratings
Winner 2001: Herman Matthews
Winner 2000: Massey Ratings
MOST ACCURATE SYSTEM - SMALLEST MEAN SQUARE ERROR (2nd Half)
Winner: Least Squares with team home field advantages
As with absolute error Least squares with team specific home field advantage
takes this award for the third straight season. It barely hung on to win last
year. This year nobody else was even close. The systems sets a new
best with a average mean square error of 126.085.
Winner 2002: Least squares with team specific home field advantage
Winner 2001: Least squares with team specific home field advantage
Winner 2000: Least squares with team specific home field advantage
BEST OVERALL RETRODICTIVE SYSTEM IN 2002
Winner: Least Squares with team home field advantages
I guess it is important to mention that this award, as with the
best predictive systems, go to the system that is the best overall,
not in just one category. We have a new winner for 2002.
Least squares with team home field advantage has surpassed
CPA Rankings this season to take the crown. This system always
does extremely well in the error measurements but for really the
first time it posted a good win/loss record as well.
As with the other overall awards, points are assigned to systems
in each category then summed up. The system with the most points
wins.
LS w/ HFA 111
CPA Rankings 102
System Average 94
Winner 2002: Least Squares with team home field advantages
Winner 2001: CPA Rankings
Winner 2000: CPA Rankings