The Line - What the heck is that?

Submitted by Blazefire on November 18th, 2010 at 11:45 PM

Introduction: Why post this obviousness?

I noticed the other day a poster had mentioned that 'the line' for the Wisconsin at Michigan game was Wisconsin -4, and they were confused as to how Michigan could be favored by four. Several posters corrected him, informing him that the -4 line means that Wisconsin is favored by 4 for betting purposes. After thinking about this for a moment, it occured to me that while most of us know this, I'm not sure WHY most of us know this.

I have never in my life seen a "beginner's guide to sports betting". Of course, I don't bet on sports, so I'm sure they probably exist, but for the majority of us that do not bet on sports, it's not something we're likely to come across. As such, it is perfectly reasonable that somebody might not get the concepts behind "lines". I'm not so much talking about how they're set, which JamieMac of Just Cover has done a wonderful job of going over before, talking about the practices of vegas oddsmakers and so on. No, I mean what they, and other sports gambling terminology means. As such, I've created a little diary here I hope that anyone who never really thought about the concept of "the line" will take a moment to look over.

Okay, fine, so what is a line, and how does it work?

Well... do you play golf?

No. I'm just a temporary personification of your target audience. I don't even know what golf is unless you decide I'm supposed to.

Shut up. You now know what golf is.

I know kung-golf.

You're not Neo, though. Moving on, in golf, one often will play with a handicap. Different leagues may calculate handicap in different ways, but the basics are this: A golfer's handicap is the number of strokes over par he is expected to finish on a given 18 holes, divided by a certain number, which changes based on the courses you've been playing. The intention of a handicap is to provide a golfer of lower ability a chance to "beat" par on the golf course.

This isn't helping.

Just wait. This is going to be rough and dirty, because I don't want to include too much math. If one golfer in a twosome has a handicap of 16, and the other has a handicap of two, that would indicate that on a standard par 72 18 holes, the first golfer would be expected to finish with a score of approximately 88, and the other a score of 74.

Rough and dirty twosomes? Lines are awesome.

Quiet. Let me speak. When you subtract each golfer's handicap from their raw score, it produces their handicapped score, which is used to determine the winner. So, in those conditions, if golfer one shot an 87, while golfer two shot his expected 74, golfer one would actually win the round with a handicapped score of 71, one under par.

And this applies to football because?

Because the lines are a form of handicap! The lines are vegas oddsmakers saying that, if we gave team one an advantage of X number of points, then it puts the teams on equal footing.

I still don't get it.

Okay. Work with me here. Lets go back to the numbers for the Wisky @ Michigan game: Wisky -4; Michigan +4. This is not an expectation of the outcome. They are not predicting that either team will win or lose by that many points.

But I though-

No. They're not. What they are saying is, "If Wisconsin were put in a 4 point hole, or Michigan were spotted 4 points, then the teams should be approximatly "on par". The difference between this and golf is that the teams should be on par with each other, rather than on par with the golf course.

So the line not a prediction of who will win or really even by how much, but a prediction of how to make the teams even?

Sort of. The important thing to remember is, the oddsmakers aren't trying to predict the outcome of the game. They're trying to generate betting. So Essentially, they're saying, "We think that on this field, at this time, Wisconsin is a better team by 4 points than Michigan. Do you agree or disagree?"

So when you bet against the line, you're not... betting on the game?

Not really. YOu're actually betting on whether the oddsmakers are right or wrong about the teams. Someone betting on Wisconsin is actually saying, "Wisconsin just destroyed Indiana, a team Michigan barely beat. They're more than 4 points better than Michigan." Someone betting on Michigan is actually saying, "Michigan is way improved over earlier in the year. I don't think they're four points worse than Wisconsin."

But isn't the answer to the question decided in the game?

Yeah, it is, but you're still betting on who will win or lose with the handicap, so you're still betting against the oddsmakers. Not against one team or the other. And even more worrisome... the oddsmakers are liars.


Yes, shocking, I know. that line they set... that's not really what they think would be an accurate handicap for the teams. If I were going to set a true handicap for the teams, I'd probably give Michigan abou a 7 handicap versus Wisconsin. If you added a touchdown to Michigan's final score, I think they'd come out tied in an average game.

So then... the point of a line?

To get you to bet. To get you either nodding your head in agreement or shaking it in disagreement with the oddsmakers, and putting your money where your mouth is.

I don't have a mouth.

Uhh... shut it anyway? Lets sum up:

The line is not a means of picking a winner of a game, nor is it intended to do so.

The line is a form of handicap, suggesting that under given circumstances, adjusting the final score by X points (subtracting for the team expected to win or adding for the team expected to lose) would create a tie score. It is an attempt to put the teams "on par" with eachother.

The line is a betting tool, and is not intended to predict the score of the game.

Betting with or against the line is an agreement or disagreement with Vegas, not a belief or disbelief in one of the teams.

Got it?

Sorta... kinda...

Good. I'm going to bed. If you have more questions, talk to one of our resident sports gamblers. At least now you won't be confused when you see Minn +16 against an opponent and wonder how they could be favored.



November 19th, 2010 at 10:50 AM ^

The money line is the same "Handicap" idea, again with the goal to get equal money on both teams.  Instead of giving up points to bet the favorite, you give up money.  If the money line is Wisc - 180, you would have to bet $180 to win $100.  Conversely the underdog (Michigan) may be + 140, meaning by betting $100 you would win $140.  But both of these bets are STRAIGHT UP, with no points given of received


November 19th, 2010 at 5:01 PM ^

Money lines are easier to handle in most people's minds rather than odds. Betting $180 to win $100 is easier than 1.8 to 1 odds. A -200 is just a 1 to 2 - you'd have to bet $200 on Wisconsin to win 100. And Michigan would be +200 or a 2 to 1 - bet 100 to make 200.

Betting lines and propositions try to be (1) simple ideas to grasp, (2) clear cut wagering - bet x to win y, (3) seemingly based on a disparity in strength - when it is really about a disparity in perception.  Does anyone know who the Utah upset of Alabama did a few years ago? That is considered an upset with Alabama on paper being favored. ? Did the line cost the books money? Or did the line split the betting pool despite a large Alabama fan base probably picking the tide?


November 19th, 2010 at 12:47 AM ^

The only thing I disagree with is, I think, a common fallacy among sports betters:


"Yes, shocking, I know. that line they set... that's not really what they think would be an accurate handicap for the teams. If I were going to set a true handicap for the teams, I'd probably give Michigan abou a 7 handicap versus Wisconsin. If you added a touchdown to Michigan's final score, I think they'd come out tied in an average game.Yes, shocking, I know. that line they set... that's not really what they think would be an accurate handicap for the teams. If I were going to set a true handicap for the teams, I'd probably give Michigan abou a 7 handicap versus Wisconsin. If you added a touchdown to Michigan's final score, I think they'd come out tied in an average game."


The flaw in this logic is that, if this were really the case, many many more people could beat sports betting than actually do. What you're really saying is that the betting public will tend to bet on Michigan's side more than on Wisconsin's side, for whatever reason. This might actually be true if you talk to a small town bookie in Ann Arbor (or Columbus in the case of OSU).

The problem with this logic is that if the line gets too far out of whack, smart sports bettors will absolutely kill Vegas by betting a lot of money on Wisconsin. If these bettors are correct that Wisconsin should be a 7 point favorite over Michigan, they win and Vegas loses. And if they do that, the spread will naturally go more and more towards Michigan's favor until the amount of money bet on each team is approximately equal. This is why spreads move during the week: they are indications of the market.

All of this is to say that, while it's true that you're betting against the odds makers, the odds makers make their money, not by reading public opinion, but by being more correct than the public. Also by charging vig.


November 19th, 2010 at 9:01 AM ^

I like you, and your thoughts.  People always gloss over this, but lines move throughout the week based on single bets and betting trends.  I believe Miami Ohio/Akron was on tuesday (akron and somebody), the line moved 4 points toward akron and they hadn't won a game.  The spread was under 10 by game time and Akron covered.

Personally (feel free to disagree), I see Vegas as the best power ranking.  The people willing to drop tons of cash on these games have algorithms and other methods to determine the value of XYZ team. They take all emotion out of it and look at the numbers.  Reverse trend lines are generally a huge indication of this.  Mich/Wisc is an example of this, majority of the public is putting money on Wisconsin (favorite, 9-1, etc), while the spread is moving toward michigan.  Obviously, someone dropped serious cash on Michigan at 6, 5.5, 5, and 4.5.


November 19th, 2010 at 7:27 AM ^

First, talking to yourself is not a waste of time, because you get to talk to someone who really knows what he's talking about. Second, he really knows how to listen.

Magnum P.I.

November 19th, 2010 at 9:52 AM ^

it would be a supercool feature if Brian et al. could create an algorithm that would allow MGoBloggers to bet MGoPoints on games and have the "line" automatically adjust based on betting patterns.

People care enough about their MGoPoints that it would actually generate some realistic betting methinks. I've got 200 points on U-M +4!


November 19th, 2010 at 11:26 AM ^

In most sports the wagering is often based on picking the winner or order of finish: 1st, 2nd or 3rd. Games like football where there is incremental scoring the result isn't just that team A beat team B, but there is a record of by how many points, runs, goals, etc.

As you nicely point out the betting line is way of gauging the relative strength of the two sides in a manner consistent with the game. Key to these lines however is that the betting lines aren't just about an accurate assessment of the relative strength - it is also a way of judging the wagering public's perception of the strength. And perception in this case may be the most important factor. The sports books ideally want a line which would draw equal amounts of wagering on both sides. They make their money off of the percentage charged for each wager, so a neatly split betting pool nearly eliminates any risk they have. They win and lose equally and profit on their wagering fee - this is no small potatoes when millions of dollars are involved each weekend.

In some sports where there isn't incremental scoring or where the scoring is harder to figure out (ex: tennis, one doesn't see a wagering line that Roger Federer is -15 games or -2 sets versus Andy Murray) one usually sees odds given. Team A for example might be cited as a 3 to 1 favorite to beat Team B. Again the point is to make the line tantalizing enough to attract an equal wagering pool. Team A may be clearly better, but a 3 to 1 return may be high enough to tempt betters to bet on Team B.

The line usually shifts as the money is bet so an opening line may gradually shift again in order ideally to balance out the bets.

I apologize if this is all old hat to the experienced bettors here, but it is worth emphasizing that in betting it is the appearance or perception of the betting line which is critical and the actual accuracy of the line is less so. For the statisticians, the situation is reversed; they are seeking accurate prognostic stats to help them better predict the actual results and don't really worry when those results may surprise conventional perception.


November 19th, 2010 at 11:24 AM ^

You asked why most of us know this.  Perhaps, they had the same conversation as I did:

Dad: Finished riding your Big Wheel?

Me: Yep.  Say, why do you have that pile of money?

Dad: Gotta go to Jordans and give it to my investment counselor.

Me:  Oh.  Jets didn't cover, huh.

Dad: Yep.

Me:  You should have bought the hook.


November 19th, 2010 at 11:43 AM ^

If the goal of Vegas in setting the line was to generate betting, they would set the line way out of whack, which would get a ton of people to bet (what betting man wouldn't at, say, Michigan -20?) Problem is, doing so will kill Vegas since all those people would likely, y'know, win. No, that's not what Vegas is doing. Instead, they're trying to balance the money on either side of the bet - that way, regardless of the way the actual game goes, half the bets win and half the bets lose. Vegas makes their cut, nothing less, nothing more. Even though each wager is against Vegas, by balancing their bets Vegas actually has minimal risk with essentially nothing on the line. As such, the line represents not what Vegas thinks (about anything, neither the game nor public opinion) but rather where the betting "market" places the game.


November 19th, 2010 at 1:15 PM ^

vegas doesnt do this. while they would like to have equal money on both sides, they cant move a line just to do that.

 It will get hammered by big bettors who know what they are doing and vegas will wind up with a couple hundred thousand dollar bets as liabilities on one side if their line just plays to public perception.


November 19th, 2010 at 2:48 PM ^

Some of the bigger betters even "set up" some of the bookies. They'll place a "small" bet on one side to move the line one way and then drop 4x much on the other.


For example, line starts at 5.5 for mich/wisc

Person X puts 2500 on Wisconsin, and gets the line to 6.5, then goes back and drops 10k on michigan.


Horace Prettyman

November 19th, 2010 at 3:34 PM ^

The expected value of and extra point or two on the line does not outweight the expected losses of betting both sides of the game.

To if you drop 20% of your money on the other team to improve a line, you would need your ecpected value to win the big bet to increase by over 20%.  1 or 2 points on the line doesn't equate to that big of advantage in winning.  You would lose lots of money doing this.


November 19th, 2010 at 8:08 PM ^

Sorry, but your analysis is not correct. The initial bet to move the line is not a complete throwaway and has almost as much chance of paying off as the main bet; so in addition to serving the purpose of moving the line to a more favorable position, it also serves as something of an insurance policy against loss.

In the previous example, let's suppose you estimate that the probability of Michigan winning or Wisconsin winning by 5 or less is 55%, the probability of a 6 point Wisconsin win is 10%, and the probability of a Wisconsin win by 7 or more is 35%. Then a straight $12,500 bet against the -5.5 line would yield an expected return of $13,062.50 (a $562.50 average profit, assuming a $1.90 payoff for each dollar bet, counting house cut). Now if an initial $2500 bet on Wisconsin moved the line to -6.5, the new expected return would be $2137.50 on the Wisconsin bet (a $312.50 loss) and $12,350 (a $2,350 profit) on the $10000 Michigan bet (against a -6.5 line). So your net expected profit goes from $562.50 to $2037.50!

Now there are lots of caveats here: it is unlikely that a $2500 bet will move a Vegas line by a full point, a 10% probability for an exact 6 point margin is somewhat high, and the ability to accurately assess probabilities significantly better than the Vegas line requires great skill. Still, the point is that betting on both sides of a moving line can present some profit opportunities for skilled gamblers, especially if that movement is by several points.