Calculate Arbitrage/Scalp bet with different spreads?

[DeucesUp]

[ QUOTE ]
Let's try another example so you guys can tell me I'm wrong again. More obvious this time (for example sake)

Book1: PHI +7 -110
Book2: CHI -5 +120

Price conversions:
-110 = 10/11 = 90.91%
+120 = 12/10 = 120%

$113 on PHI gives $102.70 on win
$87 on CHI gives $104.40 on win

so you have a gauranteed minimum $2.70 profit.
1.78% average investment on you $200
PLUS if CHI wins by 5, 6, or 7 points you win huge with the hedge.

If anyone know how to factor in this "win huge with the hedge" factor into my mathematics example I am interested!

[/ QUOTE ]

You are really struggling with some basic math here, are you sure you want to pursue this?

Your minimum profit is not $2.70 in your example. You win $102.7 - 87 = $15.70 if Phi covers and lose $104.40-$113=-$8.60 if Chi covers. You really need to get a handle on this before you go any further, this is simple stuff.





[ QUOTE ]
PLUS if CHI wins by 5, 6, or 7 points you win huge with the hedge.

If anyone know how to factor in this "win huge with the hedge" factor into my mathematics example I am interested!

[/ QUOTE ]

As to this question, the Math is simple (though still harder than simple scalp above), you simply multiply the probabilty of a 5, 6 or 7 being the final score by the amount you'll win if those numbers hit (e.g. 6 wins both bets, 5 & 7 win one and push the other, since the bet amounts are different you will have a different outcome on each of 5, 6 & 7).

The trouble is there is no simple way to get the probabilities of landing on 5, 6 or 7 -- there is no mathematical rule you can use. An NBA game is different from NCAABB game, a game between Pho & Sea is different from a game between SA & Mem, a game lined at +2 is different from a game lined at +10, the chances of landing on 1,2 or 3 are different from the chances of 6,7 or 8, a home game for NY is different from an away game for NY, a game back in Dec is different from a game in March etc. The best you can do is use historical data, figure out which games are most similar to the one you're looking at and see how often 5,6 & 7 hit in those games and hope it's close to the same in your game. Craig gave a link above which is a probably about as good as your going to do with out a ton of work.

Good luck.

[Qbawler311]

[ QUOTE ]
[ QUOTE ]
Let's try another example so you guys can tell me I'm wrong again. More obvious this time (for example sake)

Book1: PHI +7 -110
Book2: CHI -5 +120

Price conversions:
-110 = 10/11 = 90.91%
+120 = 12/10 = 120%

$113 on PHI gives $102.70 on win
$87 on CHI gives $104.40 on win

so you have a gauranteed minimum $2.70 profit.
1.78% average investment on you $200
PLUS if CHI wins by 5, 6, or 7 points you win huge with the hedge.

If anyone know how to factor in this "win huge with the hedge" factor into my mathematics example I am interested!

[/ QUOTE ]

You are really struggling with some basic math here, are you sure you want to pursue this?

Your minimum profit is not $2.70 in your example. You win $102.7 - 87 = $15.70 if Phi covers and lose $104.40-$113=-$8.60 if Chi covers. You really need to get a handle on this before you go any further, this is simple stuff.





[ QUOTE ]
PLUS if CHI wins by 5, 6, or 7 points you win huge with the hedge.

If anyone know how to factor in this "win huge with the hedge" factor into my mathematics example I am interested!

[/ QUOTE ]

As to this question, the Math is simple (though still harder than simple scalp above), you simply multiply the probabilty of a 5, 6 or 7 being the final score by the amount you'll win if those numbers hit (e.g. 6 wins both bets, 5 & 7 win one and push the other, since the bet amounts are different you will have a different outcome on each of 5, 6 & 7).

The trouble is there is no simple way to get the probabilities of landing on 5, 6 or 7 -- there is no mathematical rule you can use. An NBA game is different from NCAABB game, a game between Pho & Sea is different from a game between SA & Mem, a game lined at +2 is different from a game lined at +10, the chances of landing on 1,2 or 3 are different from the chances of 6,7 or 8, a home game for NY is different from an away game for NY, a game back in Dec is different from a game in March etc. The best you can do is use historical data, figure out which games are most similar to the one you're looking at and see how often 5,6 & 7 hit in those games and hope it's close to the same in your game. Craig gave a link above which is a probably about as good as your going to do with out a ton of work.

Good luck.

[/ QUOTE ]

Ouch yes, I see how I was very stupid. I am very knew to this. I really appreciate your post, you demonstrated how I was wrong. Plus, after seeing that I am a noob you continued to help.

Thank you

[Qbawler311]

I got this off a website, I'm not sure if its correct. Since my math skills are so weak I can't figure out what I am doing wrong. Thanks in advance to all members who choose and have chosen to help instead of ridicule.

"Remember that you must TAKE (T) the higher odds and LAY (L) the lower odds. The formula is based of off what odds you can get. In the above example the take (T) odds are 155, and the lay (L) odds are 145.
(T + 1) /( L + 1) x L = amount to lay per 1 dollar bet on the take odds.

So our take odd is 1.55, and our lay odd is 1.45

(1.55 + 1) / (1.45 + 1) X 1.45 = 1.51

For every 1 Dollar bet at +155, we bet 1.51 at –145.
The next formula is to determine the percentage profit we are guaranteed.

(T –L) / L (T+2) + 1 = percentage won.

(1.55) / 1.45(1.55 + 2) + 1 = 1.6%"


This is the line I am having trouble with:
<< (T –L) / L (T+2) + 1 = percentage won.

In his example (T-L) would not equal 1.55

I have changed the function around several different ways and I can't seem to get 1.6%?

I am interested in the correct function for gauranteed percent profit. I plan to have my programs GUI sort the highest percent gaurantee profits.

[DeucesUp]

[ QUOTE ]
This is the line I am having trouble with:
<< (T –L) / L (T+2) + 1 = percentage won.

In his example (T-L) would not equal 1.55

[/ QUOTE ]

You're right 1.55 is a typo:

(T –L) / L (T+2) + 1 = percentage won

(1.55-1.45) / 1.45(1.55 + 2) + 1
=
(.1) / 1.45(1.55 + 2) + 1
=
(.1) / 1.45(3.55) + 1
=
(.1) / 5.15 + 1
=
(.1) / 6.15
=
0.0162
=
1.6%


If you've balanced the amount bet such that you win the same amount no matter who wins by using (T + 1) /( L + 1) x L then you can use a simplier equation of:

%won = (amount won) / (amount wagered)

Amount won is 1.55 - 1.51 = 0.04 (this is amount won if the +155 team wins, but will be the same if the -145 team wins except for rounding error: (1.51/1.45)-1=0.04.

Amount wagered is 1.51 + 1 = 2.51

%won = (1.55 - 1.51) / (1.51 + 1)
=
0.04 / 2.51
=
1.6%


Note however that these are not these are not optimal bet sizes for this scalp. I'll let you see if you can figure out why. Here's a hint: The above gives you the highest guaranteed profit, but not the highest expected profit. You want to maximize expected profit.

[Qbawler311]

First, Thank you again DeucesUp.

[ QUOTE ]
Note however that these are not these are not optimal bet sizes for this scalp. I'll let you see if you can figure out why. Here's a hint: The above gives you the highest guaranteed profit, but not the highest expected profit. You want to maximize expected profit.


[/ QUOTE ]

I think you need to weight your bets more on who is expected to win some % of the time. You center the lines and find what percentage of the time the favorite is expected to win, then you weight your gauranteed bets in accordance with the expected probabilities?

center(155, -145) = 150
So fav is expected to win (or cover) 3 to 2 times or 60%-40%

So you want to portion the gauranteed winnings 60% on the side of the fav. covering?

[DeucesUp]

Got it!

Multiply the amount you'll win if teamA wins by the probability that teamA wins for your expected profit. Then do the same for teamB. The expected profit is the number you want to be the same on each side.

[Qbawler311]

Most of you were a great help. Thanks again.

Signed,

The idiot