Super Bowl line of -4. Discussion on "key numbers" and their value in NFL....

The Super Bowl line is -4 at most places.  We often talk about "key numbers" when betting football - these are line numbers that are the most important because they are the most common result numbers.

The NFL has changed with the 2 point conversion and with more modern approaches to strategy decisions.  Have key numbers changed also?

Below is a chart (unfortunately, covers 1998 technology doesnt let me post a chart without it being a garbled mess so I will type it out and do my best) that shows the full seasons (including playoffs) for 2010 and 2021 (current), and the most common margin of victory:

Margin of victory, Times it happened in that season, percentage of that margin vs all games played that season

               2010                                                                   2021
1.           3, 47, 16.5%                                             3, 43, 15.1%
2.           7, 18, 6.3%                                               6, 20, 7.0%
3.           4, 16, 5.6%                                               7, 19, 6.7%
4.           10, 15, 5.3%                                             2, 16, 5.6%
5.           14, 13, 4.6%                                             10, 16, 5.6%
6.           6, 12, 4.2%                                               4, 11, 3.9%
7.           8, 10, 3.5%                                               8, 11, 3.9%
8.           18, 10, 3.5%                                             15, 10, 3.5%
9.           5, 10, 3.5%                                               12, 9, 3.2%
10.         2, 9, 3.2%                                                 14, 9, 3.2% 

(CONTINUED)

Some things jump out.  If I would have asked before posting this how many times an NFL game lands on 2 would anyone have said top 4?  Look at the increase from 2010 to 2021 for 2.  Who would have guessed that there were more games that landed on 6 (2nd in 2021) than 7, or even that 6 was close to 7?

But lets concentrate on 4.  Is it a key number?

4 has gone down over the years because of the 2 point conversion strategy.  Coaches are going for 2 earlier in the game, and more often.  The way 4 is achieved in modern football is a late TD when down by 3.  In the 11 games decided by 4 this season, 7 of them were on a TD that was the last score of the game when the winner was down by 3.  Otherwise, it is tougher to land on 4 than any time in the history of the modern NFL.

So how much should you pay to stay on 4, or should you possibly sell a half point to get 4.5 (rams), or to +3.5 (bengals)? 

Well, Nitrogen currently has -4.5 at 102.  If you like the Rams, are you better off paying -110 at -4, or -4.5 at +102?  If you use the results chart above solely to determine this (I understand the arguments against this, but just to illustrate a point), the SB has a 1 in 25 chance of the Rams winning by exactly 4.  That would earn you a push at -110, and a loss at +102.  But in every other scenario (24 times out of 25), you are better off laying the 4.5 at +102.  

So by looking at the Rams in the example above, and just using the table above, going from -4 to -4.5 should be a 3.8 cent discount - but you are getting 12 cents.

I know I got in to the weeds there....  But it is always important to remember PRICE and SPREAD = THE LINE.  They cannot be separated.  Betting -3.5 at -120 is not getting -3.5.  You ALWAYS have a choice to get off of the "primary" (-110) line the book is giving you - do you know when and why you should be doing it?  Important betting skill.

GL all.

great stuff.

@Skipster

   at least one person is interested in this stuff  

I made a bit of a mistake above when talking about the likelihood of the SB landing on the Rams -4.

4% of the games in the NFL this year landed on 4.  But that includes UNDERDOGS also winning by 4.

So it is even less than 1 in 25 for the FAVORITE to win by exactly 4, because it excludes the examples of when a dog wins by 4.

But for the sake of discussion....  Doesnt change the point.

wow love this 

Good stuff Van

@sportschat

@chewbacca

Great detail, van… always appreciate your insight. How do you then determine good prices on alternative spreads given this information?

I went for the value in this game. Got rams 

-1.5 (-205) did I do good? 

I seen the line I was Salivating I love the bangles I don’t miss many Super Bowls best of luck but I get on Cincinnati

3.5, 4, and 4.5 are pretty much the same.  I would take Cinci +3.5 or LA Rams -4.5

How does it land on 4? 

If a team is down 11 and scores a TD, they go for 2 every time so it will be 3 or 5

If a team is down 7, it is very unlikely they kick a FG (old school coaches used to do this a lot)

Team is up 1 and kicks a FG, could happen, but I can't remember many games end this way

Very aggressive coach is down 3 and plays for a TD and not OT.  Could happen but I think both coaches play for OT.  The LAC coach did this - if he was in the Superbowl and the favorite, 4 would be more valuable

I guess it is possible a team is down 12 and goes for 2 to cut it to 4....

@vanzack

 3 has always been the key number. About 1 out of 5.5 games land on this number. Has been going on for over 40 years.

Picking the winner of an NFL game you will cover the spread over 70% of the time. I do not think the point spread will come into play on this game.

Cincinnati wins or the Rams win by 7-21 points. Only my thoughts although anything can happen in any game.

M

When it drops all the way to 3 in the next few days like I said, how you gonna feel? How you going to feel after you telling everyone it will go to 5 before 3.5 lol

the joke will be your tickets after this Super Bowl. You have tried to convince everyone to go with the rams by slamming others. Can’t wait till you have to torch those tickets. 

I will be cashing mine 

don’t even need the 3

Bengals win

Yeah Van....how you gonna feel!?

Tbh the only numbers I worry about now are 3 and 6. 14 also but it rarely comes up.

If it's less than 3 I just take moneyline on a fav. If you lose the juice isn't too crazy, and a lot of games won by last second FGs are won by less than 3. A lot of bettors dont realize this and think -2 is just like a moneyline, why pay the juice on the actual moneyline.

And if it's less than 3 and I like the dog I dont take the moneyline for the same reasons (while many people would take the little plus money for the same reasons)

With all the stupid missed PATs and the fact half the games now feature 2 pt conversions for all kinds of reasons, the way running out the clock is taking precedence over running into the end zone, I hate laying ANY points, so I worry more about the size of the number more than it being a key number now.

Except for3 or 6, because that's all you can win by in OT now (and a lot of them are still won by 3 now too) and there is more games going to OT than ever before. I hate when OT happens and your bet is dead in the water.

The NFL has changed so many rules to try and make the game more exciting. The refs have their hands on the flag for PIs and roughings to give the losing team a chance when it comes down to the last drive. Teams are blowing bigger leads more often, going for 4th down more often (i HATE this trend, the analytics are based on when it wasnt the norm, now that it is the teams get stuffed on it all the time) going for 2 more often (usually cus of missed PATS, sometimes idiot coaching) and basically the game has improved for the casual fan and gotten worse for the bettor, every game is a sweat fest. Used to be you were getting killed in a playoff game 1st half you were toast. Look at what KC did the last few years then had done to them this year and then look at the Rams try to fumble away a playoff win.

They increase the games to 17 and the playoff teams to 7, decrease the bye week teams to 1. The wild card games are all blowouts and then the 1 seeds both lose. It blew up in their faces but they got extra games and fanbases involved.

I actually had my best season ever this year college and pro but I'm going to quit for a year after this year's SB because I am tired of watching games come down to the last drive all the time. Games I handicapped perfect, that were well in hand, suddenly become sweats. It's the same reason i quit the NBA, it all comes down to free throws and desperate 3s in garbage time. You bet a team and they are up 30 at Halftime used to be the best thing could happen to you, now it's the worst, all you are gonna see is them take the foot off the gas and the refs help the other team.

The key number thing is just part of a bigger trend to make the game more crazy and unpredictable. Like as if you were playing poker the usual way for years and suddenly deuces became wild. Well some donkeys might like it but I don't want my pair of kings getting beat by Ace Deuce.

Stop fixing what ain't broke. I am looking forward to not worrying about this BS any more, especially since I just had the best year ever I can go out on a high note. I'll be back at some point Im sure but for a while not caring about what overpaid dumb jocks do with a piece of inflated leather will allow me to focus on other things that are more meaningful 

But one last note on key numbers: 14 point favs in football and basketball tend to be awful bets. You're ever in need of a winner, look for a 14 or 14.5 pt dog.

Really great stuff, Van. Especially interesting re: 2 being a more prolific number. I wonder if the new math on 2 hitting more frequently takes some of the value out of Wong teasers when you are teasing down a favorite from 8.5 to 2.5. 

Great question.  Lets see if I can go through the math...

The implied probability (% you would have to win to break even) on -110 is 52.38%, and on -120 is 54.55%.  Basically, there is a 2.2% difference in implied probability between a -110 bet and a -120 bet - or another way to think of it is that you have to win 2.2% more of your bets if you are laying -120 instead of -110 to break even.

So, the question at hand is....  If you buy a half point, and make your odds -120 instead of -110, what spreads does that make sense for by being able to make up the 2.2% difference?

If you used the results from 2021 as a sole data input (for the sake of illustration, I realize it has faults but it is actually pretty close) - Lets start by looking at 4 and ask the question....  If I like the Bengals, is it smart to pay -120 at +4.5, and if I like the Rams is it smart to lay -120 at 3.5

First off - out of the 11 games that ended on 4 this past season - not one had a line of -4 and the favorite won.  So just using this, you would have been 0-11 buying off of the 4 either way.  This is the distribution of the 11 games that landed on 4:

Closing Line, result difference in absolute value

6.5, 2.5  (example:  Chargers were favored by 6.5 over Steelers, and won by 4)
2, 2 
5.5, 1.5
14.5, 10.5
11.5, 7.5
2, 6
4, 8
7.5, 3.5
6, 2
1, 5
3, 7

So if you look at that table, the closest game that ended on 4 to the spread was 1.5 points away - meaning out of these 11 games buying a half point would have only cost you more vig - and only in one case would buying 1.5 points have benefitted you a push.

It is a little bit startling to see how few times those half points matter, when you think about how many times you see people post here about buying half points.

So back to the question..... If I like the Bengals, is it smart to pay -120 at +4.5, and if I like the Rams is it smart to lay -120 at 3.5?

In almost all cases, the answer is going to be never to buy points.  What most people don't consider is that if you buy a half point, you are only turning a loss in to a push, or a push in to win.  You are never making a loss a win.

So if a game has a 3.9% chance of landing on 4, it would be smart to "pay" only 2.2% to turn it from a loss to a win.  But since you are only turning it from a push to a win, you only take half of the 3.9% chance of landing on 4 which is 1.95%.  So paying 2.2% for a 1.95% gain is a bad proposition.

Here is the basic formula:

52.38 (implied probabilty at -110) + half of the value in the results chart in the original post = new implied probability

Then take the new implied probability and convert it to a price (like -120 for 54.5%) and you can see the fair price for that line.

In our case above:  52.38 + 1.95 = 54.33.    54.33% is an implied price of -119, and you have to pay -120

Hope that makes sense lol

Yeah, 7 of the 11 games that landed on 4 this season were a team down by 3 who scored the last TD of the game. 

Thats how it lands on 4.

You remind me of Chunk in the movie The Goonies.

The guys who is constantly "what are we doing today guys?", and hangs on always trying to make an impression while getting none of the desired attention you crave.

Dont try so hard pal.  Let it come to you.  If you do good over an extended period of time, people will gravitate towards you.  Everyone knows one result like this SB means nothing - so stop with all of the I told you so stuff.  This SB is a grain of sand on a beach.  

Relax Chunk.

@NutinButtLove

Very interesting stuff.

I think if you asked the average bettor what is a more likely NFL outcome, 6 or 7......    9 out of 10 would say 7.

It sure does.

@vanzack

The 2 point conversion was adopted 28 years ago.

IMO, to get a more accurate assessment as to the value of key numbers, you would need to have a larger sample size.

Just as an example, a 25 year period just before the 2 pointer was adopted, vs a 25 year period just after it was adopted.

I'm not asking anyone to run these numbers, simply pointing out that the percentages could be skewed by the variance within a single year.

Van, out of curiosity, why did you choose 2010 as your sample year?

Totally agree!  And acknowledge this (I hope) a lot throughout my ramblings.

I do think there is value to recency in your suggested data set though.  It isnt only the 2 point conversion, it is the seismic shift in game strategy also that is driving new key numbers.

I only brought all of this up to illustrate some points and bring out some math. 

I chose 2010 randomly.  I went back about 10 years.  I could run for another year also - it was totally random I promise - I dont have any reason to think there were outliers or anything to prove a point.  Fair question.

Good point