If lions are -42.5 @ $1.90 and -47.5 @ $2.10
that is an average of 4c per point
but if you go the other way it doesnt work as easily because $1.90-$1.08(opening price)
= 72c which equals 18 pts which means that formula must change because $1.08 would mean
-24.5 start.
Do you have an idea how it changes as the start decreases.
Cheers
P.s. Your time is your own but I know you love to rack your brain with these things.

Hope you don't mind me fielding this one moeee - please correct me if I'm off the mark.

From my understand of the relationship between spread and odds, you need to look at three cases.

CASE 1: Home team cannot lose

CASE 2: Both teams are equal

CASE 3: Away team cannot lose

note: ? = infinity if it doesn't show up.

For Case 1, the start for the Home team would be -∞ (infinity), and for the away team +∞ (infinity). The odds for the home team would be $1.00, and the away team $∞ (inf).

For Case 2, the start for both homa and away would be zero. And in a 0% vig bookie, the odds would be $2.00 on each team, or $1.85 on each team at the TAB.

For Case 3, it is the revese of Case 1. +∞ for Home and -∞ for away. With $∞ for home and $1.00 for away.

OK, so what type of equation would best fit this data? Next, I got a whole load of sample odds and starts from the TAB, and plotted start vs. odds. (I've lost this since, but I could do it again with a season's worth of data).

The curve that fitted the data most closely was a TAN curve. There is an example here: Sample Tan chart

Now, just look at the middle curve. And imagine that along the x-axis we have the odds. With 90^ being $1.00 for the home team, and 180^ being $2.00 for the home team or the away team. Then 270^ again being $1.00 for the away team. Like this:

x-axis   |--------------+--------------|

HOME     $1.00          $2.00          $∞

AWAY     $∞             $2.00          $1.00

         90^            180^           270^

Now, let's take a look at the y-axis. Assume that this represents the start for the home team. Also, pretend that it is soccer (you just change the scale for different sports).

So zero on the y-axis represents a start of zero for the home team, and thus a start of zero for the away team. And from the lookup, you can see that the correct odds should be $2.00 for home, and $2.00 for away.

Let's look now at a start of -2 for the home team. This equals odds of about $1.20 for the home team. You can then work out the odds for the away team based on the bookies vig.

What you can see is that as the start for the home gets closer to -∞, the home team odds approches $1.00. And as the home team start approaches +∞, the home team odds approach $∞ .

When I did this years ago, I found that there were points in the graph where you got better odds staking on the win rather than the start in a few places (only for the fav.). It has something to do with the start only moving in 1 point steps, while the odds could move in smaller fractions.

I am sure I can do it again if someone could point me to a spot where I can find the start and the odds for about 1 season's worth. I could build a spreadsheet that will show you the correct odds based on the start.

Hope that helps.

JL.



[ This Message was edited by: jakelee on 2004-07-21 14:56 ]

Rabbitz!
I apologise for giving you the impression that I know something about mathematics.
That Tan theory there,as we all know,goes like this.
The tangent of an angle is equal to the opposite side divided by the adjacent.
DON'T RUN AWAY JAKELEE.
Rabbitz!
You don't need to know how a television works to enjoy watching your favourite program.
Do this for the more favoured team.
First decide how much more minus than the available line you want to go.
If you pick Brisbane at -24.5 then it is -18 less than the available line.
You add this 18 to 52.5,(special number),which gives you 70.5
Divide the 70.5 into 100 gives you about $1.42
If you went Brisbane by -0.5,or in other words just to win,It becomes 52.5,(special number again),plus 42 equals 94.5.
Divide this into 100 gives $1.06.
Yeah I know it's not the $1.08 offered but like I said it gets a bit tricky when you go too far off the line quoted.

[quote]
On 2004-07-21 14:55, jakelee wrote:

Let's look now at a start of -2 for the home team. This equals odds of about $1.20 for the home team.

[quote]

-2?
-2 what?
After re-reading your post,I imagine it must be a goal in soccer.
I now need to go back to your graph to see where the $1.20 came from.

Do me one.AFL.Both teams having equal chances of a victory.
What price the home team to win by 12 points?

first thankyou jakelee,but I dropped 3 unit maths in high school and I always have trouble getting a tan,maybe its a sine and I should give up.Anyway moee your system seems reasonly accurate.what I am going to though is copy down a range of starts this week so I'll have a better idea in future.
Do you know how to convert starts for NRL too
cheers

Hi Guys,

Seriously, I'm not that much of a propeller head :wink:

You're right, you don't need to know how starts relate to odds to enjoy the game -- or even to do well from betting. It's really just a curiosity. I'm pretty sure that the odds makers come up with the start first, then some number crunchers

I'm gonna hunt around for historical odds and starts for NRL teams tonight, and I'll produce a spreadsheet or put up a webpage that will let you convert from starts to odds, or the other way around.

It might make a bit more sense. I'll get some stats for AFL too.

So we can all get a good tan.

Onya jake
sined
Rabbitz
cosined
Moeee

So what Jakelee?Would you like an easier question?
y=tan x
Nice curves.

I myself prefer y=k^x
Where k equals 0.955.(special number).
I have lots of special numbers,because I'm a special person.

Well moeeeee
it seems that the special number 52 works
with roll your owns out ,it all fits
cheers