Is it a good idea to bet the ML on favs then betting the spreads on dogs. You can only lose 1 bet but have the chance to win two. I did this last night with the Mizz/Iowa St. game. Has anyone had experience betting this way?
-
-
It will work if you get odds on your wagers that create a +EV situation.
Frankly, I do the exact opposite as one my hedge strategies for live wagering. My wagers before the game are on the ML dog and the spread favorite. Then during the game I pick my spot(s) for hedging those wagers further.
Joe.Comment -
How do u figure out +ev. Thanks for your help. I am very new to sports investing. Doing it your way don't you have a chance at losing two bets instead of only losing one my way.Comment -
Generally this will not work.
You need to determine the odds of each event winning. There are some middle opportunities, but most of the time there is no advantage (or a disadvantage) betting this.Comment -
It is not a matter of how many bets you will lose, it is a matter of after many wagers of this type, do the odds allow you to collect more on your winning hedges than you lose on your losing hedges.
As far as how I do it, by combining my pre-game wagers with wagers made live and dependent on how the game is going, I am able to "close" my hedge for a relatively small loss, more often for a small gain and occasionally for a relatively large gain.
Joe.Comment -
agree with Justin, maybe in some cases, but usually u will loseComment -
long term you will most likely lose on this because when the fav wins and dog does not cover you lose 10% of 1 unit, when the dog wins and fav loses you lose say 50-100% of 1 unit, and when they both win you win 2 units but the middle is the least common of the 3.
say the fav wins 60% of the time, the dog covers 30% of the time and they both win 10% you would win $200 10% of the time, lose $100 30% of the time (assuming a line of -200) and lose $10 60% of the time.
so in this case you would stand to lose $14 per every $100 bet.Comment -
Sharpcat, Justin, and others are right. You will lose if you make this a system and use it as something to automatically go to based on numbers instead of handicapping. I did a lot of backtesting on this before this basketball season started and it is a failure if overplayed. That said...
I do this semi-regular but not as a system. I don't purposefully find underdogs and favorites and put them together, but there are plenty of times when there are games where I think the dog will cover but not win outright. I don't wager both sides equally, putting more on the dog with spread and less on the favorite's ML. The only drawback is if the dog wins outright I sometimes lose .2 units overall, depending on how much I had to lay on the ML.
Don't just make this a system and toss it at any game. Only do it when you would already bet the dog plus the spread and already think the favorite will win outright. Tonight I did it with Wake Forest +7.5 and FSU on the ML, and it worked great. Earlier this year I took UCONN +9.5 and Nova on the ML, and lost $8.20 on $180 worth of bets because UCONN more than covered as they won straight up. Another option is if you have two games where you like a dog to cover but not win, take the 2 fav's and parlay them on the ML. As with pretty much everything, it will work if you pick the right spots. Easier said than done, of course.Comment -
A complete answer to this question is at least a chapter in a book.
With all due respect to others that have answered, yes, this can work. Of course you must pick your spots to apply such a strategy and make sure that you get odds that justify the wager. To give this strategy a better chance to work and to create more opportunities to apply it, I personally would combine it with live wagering. However, as I have already posted, I actually do the exact opposite of this because that is where I find more "mispricings" when combined with live wagering.
Sharpcat, I applaud your trying to justify your response with an example but your example is flawed IMO. The way I read your example, the favorite cashes at -200 70% of the time among other problems. If I have misread your example please clarify.
For me, it all comes down to the numbers. I try to find a way to cash more on my winning wagers than I lose on my losing wagers. When I find that way, I play it as long as it is working. Many times I find that doing the exact opposite of what I thought would work is the way to go and I have no hesitation playing that way as long as it is profitable. This is where I believe the fact that I am not a true handicapper comes to my advantage. I don't have a vested interest in any team and I can believe the "numbers" more so than someone who is sure that they should have won if only so and so didn't do this or that or if the ref was fair etc etc.
Please be aware that I am not trying to start an arguement, just discussion. If you currently handicap and/or have a "system" for finding profitable wagers, great, nothing I write denies that capability. I am just trying to give a different way of approaching the problem of finding profitable wagers.
Joe.Comment -
when the favorite covers the spread and wins the moneyline you would win $100 dollars on the fav for your ML bet and lose $110 on the underdog ATS bet for a loss of $10. 60%
When both teams cover you win $100 for the ML bet and $100 dollars for the dog ATS for a gain of $200. 10%
and when the dog wins the game out right you would win $100 and lose $200. for a loss of $100. 30%
I just did the math in my head on the first equation and got my %'s mixed up but this is what the equation would look like.
(-100 x .30) - (-10 x .60) - (+200 x .10) =
-30 - -6 - +20 = -14
So you would actually be losing 14 dollars on average per each 100 dollar bet of this type you made ( assuming the percentages are correct, which I would assume that your middle would probably hit much less than 10% so these numbers are being generous). Not saying that it can not be profitable but as stated before it is something you need to pick your spots in, like say betting one side of the line tonight and waiting till game time tomorrow to bet the other side anticipating a line moment.
As far as reversing this scenario goes, you are putting yourself at more risk because you are giving the book the chance to middle you whereas you could lose both your bets, break even or hit the +moneyline. If I were to do this I would give myself the advantage of having the middle and bet the fav on the ML and the dog ATS.Last edited by sharpcat; 03-04-10, 01:00 PM.Comment -
this is clearly not a good idea unless both lines are +EV which is severely unlikelyComment -
Sharpcat, again thank you but let me show you where I think you are going astray with your assumptions.
Your first assumption is that the favorite will cover 60% of the time ATS with the dog being -110. I then assume that the Fav would also be -110 ATS in your scenario. If I had a selection system that selected 60% at -110, there would be no need to make any other wagers. With a no vig line of 100 the correct % would be 50%. Please also note that I would never ever lay -110 on a no vig line of 100.
Your next point was that you felt that you were being generous with your middle % of 10%.
If the no vig line ATS is 100 and the ML is 200 and assuming these are accurate, then here are the probabilities:
Favorite covers ATS: 50%
Dog wins: 33.3%
The middle therefore must be 16.7%.
If the no vig ML was 150 (40%), then there would only be a 10% chance for a middle. However, if the no vig ML was 300 (25%), then there would be a 25% chance for a middle.
Your next statement is on the right track IMO: "Not saying that it can not be profitable but as stated before it is something you need to pick your spots in, like say betting one side of the line tonight and waiting till game time tomorrow to bet the other side anticipating a line moment".
Varely rarely do I actually do it this way but I manufacture the same concept with the use of live wagering.
I will address your last paragraph in another post later.
Joe.Comment -
I think we are getting confused here Joe
My scenario was based on the OP assuming taking your favorite to win straight up, ML say -200, I am just throwing out some numbers here off the top of my head there is no research put into this 
.
You than would take the dog ATS at -110 vig.
so your bet would end up looking like this:
#!) Lakers ML -200
#2) Celtics ATS +4 1/2 (-110)
With bet #1 being on a moneyline of -200 it would be safe to assume they would win the game heads up 67% of the time. (67% is the Break even point for this bet alone)
bet # 2 should theoretically win 50% of the time
And as far as the middle goes, on a 4 1/2 point middle I do not know the exact percentage of what it would hit but I would imagine it would be less than 10%.
I really do not care about giving the perfect mathematical simulation here on this one which is why I just plugged in some odd ball number, but you are correct in the sense that the dog to cover the spread should have been 50%, as I stated before I was simply just using an example of how the scenario "COULD" break down not trying to give the perfect answer.
since our numbers can only add up to 100% we will assume that 10% of the time the middle hits will come out of the remaining 50% of the time that the favorite will cover the spread.
(50% x -$100)-(10% x +$200)-(40% x -$10)
(-$50)-(+$20)-(-$4)= -$26
Regardless its most likely -EV unless the middle were to hit 25% of the time which is why I just threw some odd ball numbers off the top of my head (I simply took the % for the middle from the ATS team when I should have deducted it from the ML favorite) Although I agree that if you pick careful spots it can be profitable.
Not sure where you are getting the "no vig moneyline" betting a favorite at -200 is 100% vig and the dog ATS is -110 10%vig
Last edited by sharpcat; 03-05-10, 10:02 AM.Comment -
I made a lot of hedging wagers with intention of cutting my losses and exactly one time I hit a middle and that was with a 3 point margin. I guess it's quite rare for that to happen.Comment -
This might work, but I think it needs practice. If you do it a lot, you will start noticing patterns. Some patterns might be coincidences of just having a large enough sample. Some patterns might have value. But if you are just randomly betting two team and letting it ride, then it will have no value.
What I suggest would be using the ML point calculator as your starting point. Then on matchbook finds games with spread and money line that are slightly off.
One example currently on matchbook:
Mil: -4.5 ML:-166 / Laker -3.5 ML -166
Was: +4.5 ML" +156 / Bobcats +3.5 ML +162
As you can see from the two games one game dog is +4.5, while the other game dog is +3.5. But their ML spread is the same. If you do a spread point to ML ratio (if there is such a thing), you would find more value with the Mil/Was game. At one point, there should be +EV. Now what point is that? You will have to gain it from experience of base your number off the ML to point calculator.Comment -
I think its always best to be on the money line insted of the puck lineComment -
Let's clear up the confusion if we can.
If the Favorite is -200 on the ML, then it should win approx 67% of the time. That also means that the Dog will win approx 33% of the time. If the Favorite covers the spread approx 50% of the time we get the following:
Favorite covers ATS: 50%
Dog wins: 33%
What is left? The middle, which has to be 17%. No guessing required. If one does not understand this, they probably have no chance of hedging profitably. Also remember that the middle % goes higher or lower based on the ML and/or ATS odds.
In addition, while I know that it is sometimes semantics, no vig line (as well as vig) has a specific meaning. If one does not understand what no vig line (or vig) means, then one has little or no chance of understanding the concept being discussed.
Joe.Comment -
still -EV which is why I just quickly threw out some odd ball numbers, I did not have interest in solving the equation for OP, but obviously this is something that you have been working on and I do agree that your numbers are pretty accurate. Please understand unlike yourself I have never invested much time looking into this because the only way to get value in this is to buy one side of the bet and wait and hope for the line to move at a later time. I feel that there are many better options out there if predicting line movements is your strategy. 
Last edited by sharpcat; 03-06-10, 12:36 PM.Comment -
sometimes, people should try to visualize the continuum of bets and look at how the different pcts line upComment -
Several years ago I did an extensive study of this using past NBA and NCAA lines. Given the spread, one can determine from past results the odds of each event occurring. Given these and the lines, one can then mathematically compute the expected return on the different combination of wagers.
I built a table of probabilities for each spread. I then wrote a program that will look at the current spreads and lines, generating the expected return for each combination of wagers. The goal is to find wagers with an expected return greater than 1.
For the purposes of responding to this post, I ran the program with lines for tonight's games from standard books that I had downloaded earlier in the day. It found two combinations with expected returns greater than 1.
1-Golden State to win +450 and Charlotte -10 at -110. This is what I call an edge (some call it a Polish middle). The expected return was 1.005.
2-Pacific to win -145 and Cal St Northridge +3 -110. This is a middle and the type of wager suggested in the original question. The expected return was 1.009.
I normally do not wager on a pure mathematical model with expected return less than 1.05. In the NHL I do lower the standard to 1.03 for combinations supported by simulations.
Hence the expected returns I was finding in basketball were too low to wager. So I quit running and using this system.
As a footnote, the advent of matchbook with it's much better lines did change the environment a bit. For instance now of Matchbook you can wager 100 on Cornell at -11.5 to win 101, and 100 on Yale to win the game returning 620. This has an expected return of 1.018. That's better, but if you look at the three potential outcomes, it doesn't seem worth doing.
+520 if Yale wins outright
+ 1 if Cornell wins and covers
-200 if Cornell wins but doesn't cover
I suppose one could look at varying the wager sizes and maybe find a more optimal combination, but I've never taken the time to do that.Comment -
how do u determine the odds of winning a event?Comment -
read his post. He called it a polish middle and evaluates it in terms of edge.Comment -
I view a money line bet as a bet with a spread of +0. So the wager I proposed was Pacific-0 and Northridge+3. If the results fall between these then both bets are winners. You have to win one of the wagers, but might win both. I call this a middle for my purposes.
Had I bet the other way, Pacific-3 and Northridge to win then you can only win one of the bets and could lose both. I call this an edge (opposite of middle) while others call it a Polish middle. I assume this term is used in reference to Polish jokes. On the surface it appears to be a dumb combination of bets. But it can be long term profitable if the lines on the two wagers have an appropriate relationship.Comment -
I think most of us understand what you're doing here.
However, I'll warn you about using old data like that. You can't just crunch the numbers and then wait 3 years and think you'll get the same result. The research youve done in evaluating a fair ML given a spread is extremely common, and it's not so simple anymore as just doing that and thinking you have an edge.Comment -
yeah i would def go with that ideaComment
