By Mark "The King" Wald
Cold, Hard Football Facts Wise Ass Guy
There are few things in life harder to do than beat the NFL pointspread. Hit a major-league fastball is one. Get out of this corporate madhouse with a shred of your human dignity is another.
But we know the truth. You rock. We rock. Glad we got that settled.
If there's one area Cold, Hard Football Facts rocks, it's our picks against the spread. We hit a rough patch last week, but since last year our picks are like that part in Queen's "We Will Rock You" when Brian May's guitar rips into you like a Wusthof boning knife.
show video here
Yeah, we get goose bumps, too.
So we're kind of proud of ourselves. And it got us thinking about the NFL pointspread in general. How good are the wise guys when it comes to predicting margins of victory?
Well, here's what we found: they're not nearly as good as public perception would lead you to believe. In fact, their line is more likely to be off by double digits than is to be within a touchdown of the final margin.
But, like we mentioned a few days ago, the bookmakers' goal might not be to nail the final score. Their goal is to reduce risk and make a profit.
In that regard, they are much like your corporate Risk Management and HR departments. They're not there to help you, my naïve young careerist. They're there to hurt you.
The bookies probably want to bring in even money on both sides so they can make a tidy profit. But they still publish a number for almost all games, a number that theoretically reflects how much better one team is than another.
Our task: determine how accurate that number is.
That's a can-do, Rubber Ducky.
We have pointspreads for every regular-season and playoff game since 1969, most of the games from 1960s, about half the games from the 1950s, and a handful of games from the 1940s.
The source for most of the older games was newspaper archives, mainly the New York Times, Hartford Courant, Baltimore Sun, L.A. Times, and Washington Post.
The methodology
It's pretty simple. For each game we determined how much the pointspread differed from the final margin of victory. For example:
  • If a team favored by 6 points won by 6 points (a push), the variance is 0. The linemakers nailed the final margin of victory.
  • If a team favored by 3 points won the game by 7, the variance is 4.
  • If a team favored by 7 points won by 3, the variance is also 4.
  • If a dog getting 7 points won the game outright by 7, the variance is 14.
So how good are the linemakers when it comes to predicting margin of victory?
The Cold, Hard, Stunningly Overwhelming Facts
In 11,162 spreads covering 70 years, the wise guys nailed the final margin of victory ("pushed" the game) only 311 times (2.8%). But betting types know that a lot of lines end on half points, making a push impossible. So the fact that bookmakers haven't nailed the margin more often is partly by their own design.
In deference to their craft, we've decided to give the bookmakers credit when within a half point of the final margin. It's essentially a win for them if they're within a half point.
Based on our charitable, expanded parameters, Vegas nailed the final margin of victory 569 times (5.1%). They missed by more than 10 points 4,803 times (43.0%).
We've got tons more info. We've crunched and will crunch a lot more numbers based on our research, including:
  • The won-loss record against the spread of every team since 1969
  • The teams most likely to be favored (here's a hint, No. 1 is America's Team)
  • The biggest regular-season and postseason point spreads in history
We'll have a lot more news on all those numbers soon. In the meantime, here are a couple notable findings through the years.
Biggest line since 1969
Baltimore (-28) vs. Atlanta – 11/13/66
Oakland (-28) vs. Miami – 11/19/67
Result: Baltimore 19, Atlanta 7
Result: Oakland 31, Miami 17
Biggest line since 1978
New England (-24) vs. Philadelphia – 11/25/07
Result: New England 31, Philadelphia 28
Biggest playoff line
Chicago (-21) vs. Washington – 12/13/42
Result: Washington 14, Chicago 6
Biggest playoff line since 1978
San Francisco (-18.5) vs. San Diego – SB XXIX 1/29/95
Result: San Francisco 49, San Diego 26
Biggest bookie "miss"
78 points, Washington (-5) vs. Chicago – 12/8/40              
Result: Chicago 73-0
Most games "nailed" by bookies in a single week
Within a half point: Week 12, 2003 – 6 games
Nailed at 0 (pushes): Week 11, 1994 – 4 games