How do you calculate the no-vig line for MLs?

[PRC]

I think I can do it properly for spreads and totals but couldn't find anywhere he that explained how to calculate it for MLs. Also, what percentage do you need to gain from beating the closing line on spreads, totals, and moneylines for it to be +EV?

Thanks

[LT Profits]

Lets say Team A is -140 and Team B is +120

Team A: 140/240 = 0.5833
Team B: 100/220 = 0.4545
Overround = 0.5833 + 0.4545 = 1.0379

So:
0.5833/1.0379 = 0.5620 = -128.3
0.4545/1.0379 = 0.4380 = +128.3

[PRC]

Thanks LT. And how much would I need to beat the closing number by (in %) to make a bet +EV on spreads and moneylines? Do I need to beat the closing number by more than 4.55% every time?

[FreeFall]

beating the closing number by anything will return positive expected value... The greater the number the more you make?

[byronbb]

[Busterflywheel]

Lets say Team A is -140 and Team B is +120

Team A: 140/240 = 0.5833
Team B: 100/220 = 0.4545
Overround = 0.5833 + 0.4545 = 1.0379

So:
0.5833/1.0379 = 0.5620 = -128.3
0.4545/1.0379 = 0.4380 = +128.3

Perfect LT....

[WileOut]

Ok a quick question about Justin's video above. He says for every 2 cents you beat the closing number by, you have a 1% edge. So how do we translate this to spreads in the NBA, NFL, NCAAB, and NCAA FB? In other words how much is a point worth in NBA? NFL? NCAABB? NCAAFB?

edit: can we just use the half point converter, set the price to -100/+100, and then see what comes up? If so then for NBA, the first half point is worth about 7 cents. Then each additional half point you get away from the original spread is worth a little more than the last. If I am correct here (which I probably am not) then if you get +5.5 points instead of +4 (closing spread) on the dog, you are getting a 28 cent advantage and a 14% edge????

To get the above I set HPC to NBA, 4 point spread, -100/-100 (0% edge), and looked at the numbers. To buy to 5.5 at no vig it says it would cost 27.8 cents for every dollar risked.

[Dunder]

edit: can we just use the half point converter, set the price to -100/+100, and then see what comes up? If so then for NBA, the first half point is worth about 7 cents. Then each additional half point you get away from the original spread is worth a little more than the last. If I am correct here (which I probably am not) then if you get 5.5 points instead of 4 closing on the dog, you are getting a 28 cent advantage????

Yes, you have a 28 cent advantage but only if you you made the initial bet at +100.

In this situation you actual EV would be calculated as follows (funnily enough I have just posted this in another thread)

Probability of winning the bet
divided by
Implied probability at odds taken

minus 1

i.e 56.3%/50.0% = 1.127
minus 1
= 0.127

Expressed as a percentage, you have a +EV of 12.7%

[WileOut]

Yes, you have a 28 cent advantage but only if you you made the initial bet at +100.

Ok so if you got +5.5 -110 instead of the closing price of +4 -105, how would you calculate this egde?

And I'm sorry, where did you get 56.3% and 50% above? I know this is way below many people here but I don't post much here in TT, please bare with me. I won't clutter it up I promise.

[Dunder]

Ok so if you got +5.5 -110 instead of the closing price of +4 -105, how would you calculate this egde?

It seems to me that you are 90+% of the way to being able to do this for yourself.
Give it a go!

[Dunder]

Ok so if you got +5.5 -110 instead of the closing price of +4 -105, how would you calculate this egde?

And I'm sorry, where did you get 56.3% and 50% above? I know this is way below many people here but I don't post much here in TT, please bare with me. I won't clutter it up I promise.

For the 56.3%, consider that you got the result from the calculator whereby the true odds for your bet after the line movement are -129 (1.775 in decimals/euro odds). This equates to a 56.3% probability (i.e 1 divided by 1.775).

[WileOut]

I don't know how to account for the extra 10 cent difference from +5.5 -100 to +5.5 -110. Do I simply subtract 10 cents from +128 and make it +118, so I have 54.13. Then the -105 becomes 51.22. So 54.13/51.22 = 1.057-1 = 5.7% edge? I don't think that is right. It should be higher. I think I have to factor the fav prices in both spreads too somehow.

[Dunder]

I don't know how to account for the extra 10 cent difference from +5.5 -100 to +5.5 -110. Do I simply subtract 10 cents from +128 and make it +118, so I have 54.13. Then the -105 becomes 51.22. So 54.13/51.22 = 1.057-1 = 5.7% edge? I don't think that is right. It should be higher. I think I have to factor the fav prices in both spreads too somehow.

I have to step out for a couple of hours WileOut, but when I come back, we can go through it step by step. Your calculation above has some slight flaws and your reasoning is correct

[WileOut]

ok thanks man.

[Peeig]

a cleaned up version of this calculation should be in the FAQ (if there ever is one) for noooobz

[Dunder]

OK, for the purposes of this calculation you have taken +5.5 at -110
The closer was +4.0 at -105 (the other side would be -4.0 at -115).


If we use LTs calculation above to arrive at the no vig closer we get (-104.4/+104.4)

If we then take +4.0 at +104.4 into the half point calculator, we see that the correct odds for our bet at +5.5 should be -122.6
Expressed as a probability this is (122.6 / 222.6) 55.08%

This compares to the impied probability of the line (-110) we have taken which is (110 / 210) 52.38%

Our EV is therefore (55.08 / 52.38) -1
=+5.15%

[WileOut]

Thank you Dunder.

Sorry to clutter up the TT with my newbness.

[blix177]

How did LT get these numbers?

Team A: 140/240 = 0.5833
Team B: 100/220 = 0.4545

From these numbrs?
Lets say Team A is -140 and Team B is +120

[aggieshawn]

good points

[IrishTim]

How did LT get these numbers?

Team A: 140/240 = 0.5833
Team B: 100/220 = 0.4545

From these numbrs?
Lets say Team A is -140 and Team B is +120

It's the implied winning percentage of the respective moneylines.

For favorites, you calculate implied winning percentage: favorite moneyline divided by fave ML plus 100.

Team A: 140/(100 + 140) = 58.33% implied WP

For dogs: 100 divided by dog ML plus 100.

Team B: 100/(120 + 100) = 45.45% implied WP

Then you need to add these two numbers and anything over 100% is the "overround" (theoretical hold percentage): 58.33 + 45.45 = 103.78%

Once you divide both implied winning percentages by the overround (3.78%), you'll have the no-vig line as LT showed.

[blix177]

So a -900 and +800 would be,

Team A = 900/1000=.9

Team B = 100/900= .1111

Over Round = 1.0111

.9/1.011 = .89
.11111/1.01111 = .11

Now I am lost how do I convert .89 and .11 to (-100/+100)ish lines?

And what does a high overround vs a low overround implies if anything about the line?

[skrtelfan]

Low overround means less kept by the house, i.e. lower vig.

[Patrick McIrish]

I know this is way below many people here but I don't post much here in TT, please bare with me. I won't clutter it up I promise.

Don't worry about it, take advantage of this forum! Yeah some will give you a hard time but don't even worry with it. You got a question, ask. Everyone here had to learn somewhere. Plus there's probably another 100 guys that want to ask the same question but don't have the balls.

And BTW while this might not make sense immediately, after running these formulas a few times they will become 2nd nature. In the meantime keep these thread in a file somewhere for quick reference. Good luck buddy.

[benjy]

So a -900 and +800 would be,

Team A = 900/1000=.9

Team B = 100/900= .1111

Over Round = 1.0111

.9/1.011 = .89
.11111/1.01111 = .11

Now I am lost how do I convert .89 and .11 to (-100/+100)ish lines?

And what does a high overround vs a low overround implies if anything about the line?

After getting rid of the overround and verifying that your probabilities add to one (i.e. sum of percentages = 100)

For probabilities < .5

American Line = (1/P(x) - 1) * 100

For probailiteis > .5

American Line = (-1)*[1/(1-P(x)]*100

[Flight]

For probailiteis > .5

American Line = (-1)*[1/(1-P(x)]*100

I keep getting the wrong answer with that formula Benjy. Can you double check it? Here is the formula I usually use:

-100 / [(1/p) - 1]

[byronbb]

Also look for ganchrows excel functions in the tools section.

[Thremp]

beating the closing number by anything will return positive expected value... The greater the number the more you make?

If the line is efficient, yes.

[benjy]

I keep getting the wrong answer with that formula Benjy. Can you double check it? Here is the formula I usually use:

-100 / [(1/p) - 1]

I musta been 3/4 asleep. Sorry dude.

Your formula should work fine

[Newguyintown]

Hi gang,

Revisiting this and ran into a problem I was hoping someone could help me out with. So I'm trying to see how to calculate edge based on a closing line.

I'll stick with the same example from above. So let's take -140/+120. I see how we get 56.2% and 43.8% no-vig breakevens (and this translated to -128.3/+128.3). This is all precisely what Ganchrow's EXCEL chart says.

The problem is when I try to determine the edge of say a -130 wager. If I plug in -130 to the Ganchrow chart I get that its a -.56% wager. I'm basically trying to see how we get this.

So we need 56.52% to breakeven (130/230) and the true breakeven is 56.2%. So somehow its not just -.32 (56.2 - 56.52)? WHERE DOES THE -.56% COME FROM?

Any help please!

[LT Profits]

Divide the "true" breakeven by the breakeven on the odds you got and subtract 1. So, (.562/.5652)-1

[Newguyintown]

ty!