Wednesday/Friday Edition, September 18th-20th 2002

WEEK THREE PICKS Yuk, yuk, yuk! I hate the first month of the NFL season. What's up is down, what's down is up, and my annual newfangled optimism in my team, the Cincinnati Bengals, which has had a whole summer to grow, hasn't been fully crushed yet. Consider the Bengals' optimism fully crushed from here out. Since my advisories have been feeble (and I can't ride on the game of the century preseason pick for much longer), and I can't kick in with the big TwoMinuteWarning stat guns until week five, I've called in the calvary, the reinforcements, hit the panic button, broken the glass which says "break in case of emergency," put on my Zaphod Beeblebrox sunglasses that turn black at the first sign of trouble and gone around like Bill Murray in 'Groundhog Day' trying to set things right with everyone I know so my karma will be correctly aligned going forward. More importantly I talked to Scott, who was formerly I suppose head research mole at TMW but is now involved in another project (more on that another time), who had some sage words of advice for playing the early season...TRENDS! That's right, Scott wrote the Trends column last season for TMW and as you would expect from a number cruncher, he had numbers galore. His selections apparently finished at about 55% for the season, but he was a cracker 15-7 in the first four weeks when you might think the trend stuff would be most informative. I am never forget the day I first meet the great Lochevsky. In one word he told me secret of success in mathematics: Plagiarize! Plagiarize, Let no one else's work evade your eyes, Remember why the good Lord made your eyes, So don't shade your eyes, But plagiarize, plagiarize, plagiarize - Only be sure always to call it please 'research'. -- "Lobachevsky" by Tom Lehrer Could of also used the "I get by with a little help from friends" Beatles number, but here's my 'research' on the subject of week three trends (Danka Scott): This week in the Trends column I'll examine whether there are any specific conventional stat categories that have had strong predictive value in week three games. The following table lists various statistical factors and the results against the spread playing the team with the HIGHER numbers in the specific category. Note that this means that in something like "Points Allowed" the results are for playing the team with the most points allowed, not the team with the fewest points allowed as you would probably expect. In games where teams have equal stats, or one team had a week one bye, there is no play. Week 3 Stat Comparison results 1991-2000 (Results expressed as W-L vs the spread) Factor Home W-L Home W% Away W-L Away W% Total W-L Total W% Win % STD 22 - 15 59 % 21 - 9 70 % 43 - 24 64 % Win % Last Yr 30 - 35 46 % 24 - 17 58 % 54 - 52 50 % Win Last 19 - 10 65 % 20 - 7 74 % 39 - 17 69 % Win 2 Games Ago 11 - 18 37 % 15 - 11 57 % 26 - 29 47 % Spread % STD 20 - 21 48 % 18 - 18 50 % 38 - 39 49 % Spread % last Yr 30 - 30 50 % 23 - 20 53 % 53 - 50 51 % vs Spread last 19 - 17 52 % 17 - 13 56 % 36 - 30 54 % vs Spread 2 back 10 - 16 38 % 11 - 16 40 % 21 - 32 39 % Points for last 37 - 28 56 % 36 - 21 63 % 73 - 49 59 % Points Allowed last 29 - 27 51 % 35 - 32 52 % 64 - 59 52 % Net Pts STD 34 - 32 51 % 33 - 25 56 % 67 - 57 54 % Net Pts Last Yr 28 - 35 44 % 23 - 20 53 % 51 - 55 48 % Net Pts Last 33 - 29 53 % 37 - 25 59 % 70 - 54 56 % Net Pts 2 back 34 - 34 50 % 31 - 27 53 % 65 - 61 51 % Pts For STD 36 - 30 54 % 36 - 22 62 % 72 - 52 58 % Pts For Last Yr 34 - 30 53 % 29 - 17 63 % 63 - 47 57 % Pts Agn STD 32 - 28 53 % 34 - 28 54 % 66 - 56 54 % Pts Agn Last Yr 19 - 28 40 % 32 - 28 53 % 51 - 56 47 % Total Pts STD 36 - 31 53 % 33 - 24 57 % 69 - 55 55 % Total Pts Last Game 25 - 27 48 % 32 - 23 58 % 57 - 50 53 % Rush Yds STD 38 - 37 50 % 29 - 22 56 % 67 - 59 53 % Rush Yds Last Game 37 - 32 53 % 33 - 24 57 % 70 - 56 55 % Rush Yds 2 back 23 - 38 37 % 28 - 38 42 % 51 - 76 40 % Pass Yds STD 38 - 32 54 % 34 - 23 59 % 72 - 55 56 % Pass Yds Last game 32 - 27 54 % 39 - 29 57 % 71 - 56 55 % Pass Yds 2 back 29 - 38 43 % 28 - 32 46 % 57 - 70 44 % Rush Yds Agn STD 26 - 21 55 % 44 - 35 55 % 70 - 56 55 % Rush Yds Agn L Game 26 - 28 48 % 38 - 35 52 % 64 - 63 50 % Rush Yds Agn 2 back 29 - 28 50 % 38 - 32 54 % 67 - 60 52 % Pass Yds Agn STD 34 - 28 54 % 37 - 27 57 % 71 - 55 56 % Pass Yds Agn L Game 32 - 30 51 % 35 - 29 54 % 67 - 59 53 % Pass Yds Agn 2 Back 32 - 30 51 % 36 - 29 55 % 68 - 59 53 % Turnovers STD 26 - 28 48 % 32 - 26 55 % 58 - 54 51 % Turnovers L Game 21 - 27 43 % 27 - 28 49 % 48 - 55 46 % Forced Turnovers STD 31 - 28 52 % 27 - 27 50 % 58 - 55 51 % Forced Turnovers L Game 27 - 20 57 % 29 - 24 54 % 56 - 44 56 % Net T/O L Game 27 - 19 58 % 32 - 26 55 % 59 - 45 56 % Net T/O STD 30 - 35 46 % 28 - 27 50 % 58 - 62 48 % Penalties STD 34 - 36 48 % 23 - 24 48 % 57 - 60 48 % Penalties L Game 28 - 35 44 % 28 - 26 51 % 56 - 61 47 % Sacks OFF STD 25 - 28 47 % 29 - 28 50 % 54 - 56 49 % Sacks OFF L Game 22 - 27 44 % 27 - 29 48 % 49 - 56 46 % Sacks Def STD 26 - 25 50 % 32 - 28 53 % 58 - 53 52 % Sacks Def L Game 33 - 22 60 % 29 - 22 56 % 62 - 44 58 % Punts STD 22 - 39 36 % 26 - 29 47 % 48 - 68 41 % Punts L Game 22 - 34 39 % 21 - 34 38 % 43 - 68 38 % First Downs STD 26 - 33 44 % 25 - 29 46 % 51 - 62 45 % First Downs L Game 36 - 34 51 % 29 - 23 55 % 65 - 57 53 % First Downs Agn STD 30 - 26 53 % 37 - 24 60 % 67 - 50 57 % First Downs Agn L Game 27 - 29 48 % 33 - 27 55 % 60 - 56 51 % 3rd Down Conv STD 40 - 33 54 % 32 - 19 62 % 72 - 52 58 % 3rd Down Conv L Game 34 - 35 49 % 29 - 22 56 % 63 - 57 52 % 3rd Down Conv Agn STD 23 - 25 47 % 37 - 33 52 % 60 - 58 50 % 3rd Down Conv Agn L Game 28 - 29 49 % 35 - 29 54 % 63 - 58 52 % Time of Poss STD 35 - 37 48 % 28 - 26 51 % 63 - 63 50 % Time of Poss L Game 39 - 30 56 % 35 - 22 61 % 74 - 52 58 % 1st Qtr Net L Game 34 - 34 50 % 25 - 21 54 % 59 - 55 51 % 4th Down Conv Agn L Game 24 - 23 51 % 41 - 30 57 % 65 - 53 55 % For the data above the factors where home or away results were 60%+ are light green (57%+ for all plays) and the factors where home/away results were 40% or worse are dark green. As is usually the case when dealing with small sample sizes there have historically been a number of areas with results some ways from the norm. We will put together the following list of factors to examine for this week's games: Postive Factors Winning % Season-to-date Won last game Points Scored last game Points Scored STD Sacks (def) last game Time of Possession last game Negative (go against) Factors Won vs Spread 2 back Rush Yards 2 back Punts last game It is completely within reason to argue with the inclusion of any of the above and the omittal of other seemingly predictive factors, but we will concentrate on the nine factors with reasonable total effectiveness. That then was Scott's take on the matter last season at this junction, and while I don't have the wherewithal to run the data above to include the 2001 season ("if I only had a brain") we can assume it all did well enough as Scott's picks were 7-3 in week three. Therefore I will simply plug in the matchups this week against the program the big S Man wrote to calculate who's ahead in which department and this is what we get: Match-Up Line (Home) Nine Factor Scorecard Trend Pick Carolina at Minnesota -6 Minnesota 5-3 - Cleveland at Tennessee -4.5 Cleveland 7-1 Cleveland Dallas at Philadelphia -8.5 Philadelphia 6-0 Philadelphia Indianapolis at Houston +12.5 Indianapolis 4-3 - Kansas City at New England -8.5 New England 8-0 New England New Orleans at Chicago -2.5 New Orleans 4-2 - N.Y. Jets at Miami -6 Miami 6-1 Miami Buffalo at Denver -9 Tied 4-4 - San Diego at Arizona +1.5 San Diego 4-3 - Green Bay at Detroit +8 Green Bay 6-1 Green Bay Seattle at N.Y. Giants -6.5 N.Y. Giants 7-2 N.Y. Giants Washington at San Francisco -8 San Francisco 5-2 - Cincinnati at Atlanta -7.5 Atlanta 3-2 - St. Louis at Tampa Bay +1 Tampa Bay 6-2 Tampa Bay ANALYSIS: There are any number of ways you could decide whether something qualifies as a play or not. Scott used a 3-1 ratio or better and I ain't rocking the boat. The seven teams above are henceforth to be named fully certified WIN-O-RAMA picks. Last Week: 0-4 laugh it up while you can buddy-o, I'm coming back strong! Since I'm so darn confident, play 'em all in teaser parlays TOO. tim@twominutewarning.com Return to TwoMinuteWarning.com Copyright © 2002 by TwoMinuteWarning.com, All Right Reserved