Looking ahead to NFL 2003

It may be the off-season, but there's a lot of activity planned for the coming months as we prepare for the next season. Among the projects we will be working on are: Injury Research -- the early research on evaluating injuries shows a lot of promise, and so we're building a ten year database to investigate whether the "this week" and "change week" differences prove to be as telling over a longer time period. In addition, there will be more analysis of injuries by position and injuries by starter/backup roles. Weather Effects -- a lot is made about cold weather games, "sun belt teams" and similar references to temperature, humidity, and rainfall at game time. The question is what predictive value does the weather really have against the spread and the over/under line. We'll be doing some long term analysis on implementing weather as a handicapping factor. Time Frame for Stats -- we obviously already have the season-to-date and last five weeks breakouts as part of our existing coverage, but we will further investigate the benefits of using different data sets, as well as focusing on how to read the interation of multiple time period data (when they agree, disagree, etc). Home Field Advantage -- while we have thus far used a standardized number for home field advantage, we'll conduct some research on whether there is anything to be gained by variable team HFA's, home field advantage by week and situation and similar observations. Number Crunching -- our primary tools of drive chart stats and play-by-play ratings will be evaluated and tested for ways to improve predictive performance, particularly in terms of understanding when favorites can be trusted as overlays. In addition, the down-by-down stats will be readied for projection purposes for 2003. More Turnover Research -- continuing our series of articles on turnovers, look for several more pieces we've been working on to be published in the coming months, some with eye-opening results and tweaks on some of our current beliefs about turnovers as a wagering factor. New Columnists -- while some of our columnists had successful seasons in 2002 by any standards, we will be actively recruiting some new blood to improve the current quality of our writer stable. "Ultimate Ultimate Prediction Machine" -- the UPM is undoubtedly our most popular factor in terms of usage by subscribers, but in spite of some generally impressive results, we are well aware of some glaring weaknesses in the simulations. Hence we have two people dedicated in the off-season to creating our next level handicapping machine, the "UUPM" which is designed to incorporate many of the areas currently outside the scope of the existing UPM into one overall simulation engine that will have a similar output (eg team and confidence value) but be much more encompassing. Two more announcements of note: Subscriber SURVEY If you haven't already, please take a minute and fill out our subscriber survey. Your feedback is very important to us. Click Here. Early Bird 2003 Subscription Offer We are thankful to everyone who subscribed for this past season, and for those interested in re-subscribing for the 2003 campaign, we are offering a $99 price through February that will be the lowest available: 2003 Renewal.

Season in Review

The following table lists factors from high to low based on their final 2002 records from week five on (pushes are ignored). Feature Record Cover % Trends 32-18* 64% Play-By-Play 4+ Overlay Dogs 21-12 64% T/O Fumble Picks 46-28 62% PBP Last5 4+ Overlay Dogs 26-16 62% Over/Under Column 20-13 61% UPM Last5, 65%+ Confidence 21-14 60% C.S.M. 21-14 60% Tim's Picks 34-23 60% Turnover Difference 37-26 59% DC Last5 4+ Overlay Dogs 23-16 59% TMW Recommend Plays 40-28 59% Best Bets Column 78-58 57% Inside the 20 64-50 56% UPM 65%+ Confidence 10-8 56% Drive Chart 4+ Overlay Dogs 16-13 55% UPM Money-Line Large Edge plays +1.4 units n/a Teaser Tickets +0.7 units n/a Parlays +0.6 units n/a TMW Leans 36-32 53% O/D Ranks Overlays 22-20 52% Angles 75-70* 52% Injury Report 48-47 51% * record for the entire season (week one on)