Tuesday, September 7, 2010

NFL 2010 Regular Season Recap

My Record

After 49-62-1 the final 7 weeks, I fell to 50.6% winners for the season or 126-123-7. I ran just behind a simple all-underdog strategy going 5-6 with "value overdog" picks. I did not have a winning week after week 10. All in all, rather a disappointing season after a promising start.

I have been participating in a season-long NFL pool picking winners against the point spread since 2001 when my then toy boy Reginald persuaded me that there was no better way for an English lass like me to immerse myself in North American culture. Over the past 10 years, I racked up a 52% winning record beating the pants off long-gone Reginald and most other foolish men who analyze each and every game in mind-numbing detail.

My Approach

It's no accident that my long-term record is right on the money with the 52% historical winning record for a simple all-underdog approach. I pick mostly underdogs leavened by a small selection of what I call "value overdogs". Most years, my approach will place you in the middle of the pack. But, statistical analysis indicates that 1 out of every 6 years on average, I will pick more than 55% of all games correctly (e.g., 141-115 or better). 55% or better should be enough to give you a chance to win a pool prize. If you are a contender once every 6 years, you will be doing well.

Wednesday, June 30, 2010

CFL 2010 Recap

For the 2nd year running, I correctly picked SASKATCHEWAN to cover over Montreal in the Canadian Football League championship game. Saskatchewan fell to 3-15 in Grey Cups. If there is another professional sports team with a worse record in finals, I can't think who it might be. (Apologies to the prairie family of offical Lady G toy boy and fiance Archibald.)

I finished 7-6 with my "value overdog" picks this year, but am 11-15 over 2 years. For the past 2 seasons as a whole including playoffs, I am 84-70 or 54.5%. A simple all-underdog strategy leads the way at 88-66 or 57.1%. Does anyone out there have any historical data on CFL underdogs?



TORONTO MAYOR RACE

Archibald wants the world to know his prediction that George Smitherman would win the race for Toronto Mayor IF former Deputy Mayor Joey Pants fell to less than 10% of the vote proved to be correct. Archibald prognosticated that if Joey Pants cracked the 10% threshold as indeed he did, high school football coach Rob Ford would be Mayor. Archibald, get your own blog!

Tuesday, May 25, 2010

Major League Baseball 2010 Recap

(Revised 2 November 2010)

I incorrectly picked Texas to take the World Series.

I based my pick on evidence of American League (AL) superiority. The AL has won 9 of the past 15 World Series and 12 of the past 19. Leaving aside the 11-inning tie in 2002, the AL has won 13 of the last 14 all-star games to reach a decision and 18 of the last 22.

AL teams finished 134-118 or 52.3% against their National League (NL) rivals in 2010 interleague play.

2010 marked the 7th consecutive year of AL superiority starting in 2004. However, 2010 was the best year for the NL since 2004 when they took 49.8% of all interleague contests. And, kudos to the NL for finally winning an all-star game for the first time since 1995. And, if we also consider San Fran's World Series win, perhaps there is preliminary evidence that the NL has found its long-lost mojo.

AL Rules

But, it remains true that the AL sports a 56% record against National League (NL) opponents in over 1,500 regular season and post-season inter-league contests from 2005 through 2010.

I calculate that the AL will win more than 50% of inter-league contests 19 out of every 20 years. I still think that management of NL teams is way behind the more scientific approach implemented by the more advanced and successful AL franchises – Red Sox, Rays, Yankees, Twins, As and even the Indians (who admittedly haven't been getting good results the past few years).

It's not just a matter of money. The LA Dodgers, Chicago Cubs or NY Mets could generate the same revenues as the Yankees and Red Sox if they put the right minds to it. Too bad the scarecrow's song in the Wizard of Oz is the motto for most NL franchises: "If I Only Had a Brain".

Interesting to note that San Diego's owners handed management over to a bright spark trained in the Red Sox organization and the Padres got far better results in 2010.

Good move for the Toronto Blue Jays to hire John Farrell, one of the Boston brains, as their new manager.

Implications for Pool Players

Official Godiva toy boy and fiance Archibald participates in a friendly, non-profit office pool picking major-league baseball winners in all games over the course of the season.

His pool works like this. He gets:
• 1 point for a win backing an overdog;
• -1 point for a loss backing an underdog;
• +1.x points for a win backing an underdog with 1.x based on the posted odds;
• -1.x points for a loss backing an overdog with 1.x based on the posted odds.

I advise Archibald to pick the AL team in all inter-league games.

Over 260+ games each year, my system generated a net gain for Archibald of 30.75 points in 2008 and 18.45 in 2009.

And, these gains were generated almost entirely by AL road teams visiting NL hosts – 20.95 points in 2008 when AL visitors won 55% of the time against their NL hosts and 20.45 points in 2009 when AL visitors won 54% of the time.

In other words, oddsmakers and their customers may not perceive that the AL is so much better than the NL that AL teams should be favoured most of the time even on the road against the NL.

I was so busy this summer that I did not have time to analyze Archibald's results in 2010 interleague games. Given the NL's slightly better performance this year, I suspect that he broke even at best with my all-AL strategy. But, my guess is that Archibald will do well most years with a consistent all-AL approach until NL managements show signs of catching up with their AL counterparts.

Archibald failed totally with his pre-playoff prediction that the Yankees would defeat the Braves in the 2010 World Series.

Monday, April 19, 2010

NBA Finals Game 7

I finished the 2010 playoffs 51-31 or 62.2% to the good against the point spread with a simple all-overdog strategy.

For the past 3 years, I have followed what I call my Gibbs-Gipp system:
http://economics.stanford.edu/files/Theses/Theses_2007/Gibbs2007.pdf

and have called 55.7% of all playoff games correctly against the point spread.

And, over three years of picking NCAA March madness matches, I won 53.9% against the spread.

The statistically significant part of the Gibbs-Gipp system – backing all "heavy" underdogs receiving 12.5 or more points – came through with 61.8% winners over the 2009/10 NBA regular season after delivering a 54.7% success rate against the points in 2008/9.

My complete NBA record for the 2009/10 regular season + playoffs was 62% -- not too shabby if I do say so myself.

Official Godiva toy boy and fiance Archibald is still in mourning after the 1st ever Game 7 finals loss by his beloved Celtics – and to the hated LAkers to boot. When I mentioned that he seemed inconsolable, he begged to differ and got down on his knees to beg yours truly for the sweet consolation that only Lady G can provide.

World Cup Soccer Football

Archibald insists that I pass along his prediction that the Netherlands will win the World Cup. He's even more confident after their 1st 2 victories. (Get your own sports blog, Arch, and stop mooching off mine.)

As for me, you can take Lady Godiva out of Coventry, but you can't take my heart away from backing England. Too bad we can never find a good keeper or reliable scorer. (Until I met dear Archibald, I felt the same way about men.)

(Amended 21 June 2010)

Sunday, March 14, 2010

NCAA March Madness 2010 Recap

After losing on points with Duke -7 in the final, I finished even-steven at 31-31-2 against the point spread in this year's March madness.

My modified Gibbs system regressed to the mean over 3 years. In 2008, I finished 37-27 or 58% against the spread. In 2009, I declined to 34-29-1 or 54%. Over 3 full years from 2008 through 2010, I am 102-87-3 or 53.9%. Good, but not great and we won't have enough data to know whether my modified Gibbs system is a statistically significant winner for another 10 years or so of NCAA tournaments. I wonder if I will still be as great-looking 10 years from now.

Before I get carried away with my good looks and basketball-picking prowess, I should admit that just by picking favourites in every NCAA tournament game for the past 3 years, you would have finished ahead of me every year and would be sporting a 108-81-3 or 57.03% record.

Lady G vs. President Obama

President Obama picks winners straight-up with no point line for each bracket and round.

http://games.espn.go.com/tcmen/en/entry?entryID=3288143

He had a tough time this year with none of his final 4 picks coming through. But, when he picked upsets in the early rounds, I could identify games where Lady Godiva and the Baracketer-in-Chief disagreed. Since Obama took office, I am 7-9 womano-a-mano against the Baracketer. I won 5-3 last year, but the President can at least boast that he defeated me 6-2 this year.

And, speaking of mano-a-womano, kudos to the President for picking the women's NCAA brackets this year.

http://games.espn.go.com/tcwomen/en/entry?entryID=125259

Lady G's System

I myself take no interest whatsoever in American college basketball, but toy boy fiancé Archibald can't get enough roundball. Every year he organizes a friendly, non-profit office pool for the NCAA men's tourney. Instead of picking bracket winners, the object of his pool is to pick winners against the point spread each and every day as the tournament unfolds. The winner of his winner-take-all pool is the contestant who picks the most winners against the spread over 64 matches. All matches are weighted equally starting with the play-in game to determine the 64th team.

To humour Archibald, I participate by relying on economists' studies showing that "heavy" underdogs win more than their fair share of college basketball games against the point spread, while "non-heavy" overdogs win slightly more than 50% (although this particular result is not statistically significant).

http://www.entrepreneur.com/tradejournals/article/print/179073423.html

Monday, January 25, 2010

Super Bowl 2010

Kudos to President Obama for correctly calling the Saints. He's undefeated as President with his Super Sunday picks straight-up.

And, I have now lost 3 Super Bowls on the trot against the points.

At least, I ended up on the + side at 6-5 for the playoffs. Nothing great, but my first winning playoff season since 2002.

Underdogs vs. overdogs now even-up at 21-21-2 in 44 Super Bowls.

I will be less active with this blog until March madness aside from the occasional NBA pick. But, tune in for the NCAA tournament when I hope to repeat my feat of outpicking the Baracketer-in-Chief.

Wednesday, January 13, 2010

NFL Semi-final Picks

If your friendly, non-profit pool extends through the playoffs, stick with the home overdogs in the conference championship games even though there is no strong tendency to go on in the historical data.

Favourites have won 65% of 78 conference championships since 1970 – actually a touch below the 2/3 outright winning rate for overdogs in the regular season. I recommend taking the favourites and ignoring the points. My unproved hypothesis is that fewer playoff games turn on points because winning teams go all out to pile up the score in elimination games.

Speaking of piling up the score, Minnesota went beyond the bounds of good taste throwing for a TD on 4th down leading 27-3 late in the game.

Pick

+/-

Against

NEW ORLEANS

-4

Minnesota

INDIANAPOLIS

-7.5

NY Jets

Indy covering the points is my choice for best bet this week. Sorry, former Godiva toy boy Reginald and still diehard Jet fan. Not this year for your beloved Jets. Too bad you loved the Jets and football more than you loved me.


Playoff Picks So Far:

3-1 with the home overdogs last week was easy-peasy. Favourites playing with the extra week of rest have won nearly ¾ of 80 divisional round games since 1990. The godds of probability definitely tilt to home overdogs in the divisional round.

Picks

Right

Wrong

All

5

3

Best Bets

3

1