Team vs Team NFL win totals

[a4u2fear]

My book offers
Atl -1 vs Hou season win totals

i have calculated expected wins for each team

is there a better way to evaluate the bet than:
Atl exp wins 8.4
hou exp wins 9.5

about two full games of value?

[danshan11]

where do you see the edge? I mean you think Houston wins more
Houston is favored in 9 games at this point about -21 ATS for the season
Atlanta is favored in 9 games at this point about -24 ATS for the season
so this says for the season they both should win 9 and atlanta is slightly favored by 3 points on the season
so what is that worth? and what is the offering at -1 , -110????

[semibluff]

My book offers
Atl -1 vs Hou season win totals

i have calculated expected wins for each team

is there a better way to evaluate the bet than:
Atl exp wins 8.4
hou exp wins 9.5

about two full games of value?

What you seem to be asking is whether it's better to bet 2 events where you have an edge as singles or effectively parlay them. There is a thread about that in this forum. I can see the argument for NOT doing it based on Kelly returns but on balance i'm i favour of the increased edge parlay. Obviously there's nothing to stop you doing a bit of both. It depends on your view of resource maximisation and whether the 'pleasure' of betting and watching both is worth the loss of marginal theoretical profit.

There are 2 things to be aware of:
1) Generally there's a higher -115, -115 level of juice on totals rather than -110, -110 so potential profit margins are squeezed.
2) Books aren't laying the market as a flat playing field. Logic dictates there will be 256 wins and the combined Over price should match the Unders. This isn't the case. Every book i've seen is laying the market at around 255 wins with shorter prices on the Overs compared to the Unders. Effectively you're looking league-wide at 8.5 wins with -120 Over and -110 Under. From a purely logical perspective and 0 analysis of any specific team there ought to be more NET value on Unders and less value on Overs.

So while I can see the strong logical argument for pitting an overvalued Over line against an undervalued Under line as an effective parlay i'd probably prefer to just bet the undervalued Under line as a single. If there are 2 games difference worth of value on the 2 teams, (and i'm not necessarily agreeing with that analysis), then there's more likely to be 1.5 games worth of value on Atlanta and only 0.5 games worth of value on Houston.

[danshan11]

I think the issue is how do you get those win estimates, the vegas lines say different, technically right now both teams are expected to win 9 games and atlanta is slightly more favored to win the 9 but favored to win 9

[a4u2fear]

I came up with a method to bet team totals and developed my own lines (reviewed with prior years and so far so good)

i have already evaluated the solo team totals when I just noticed my book offers team va team total wins

so in the above bet it does not matter how many wins either team gets to but rather Atlanta has to win two more games than Houston to win that side of the bet.

so ideally, my numbers tell me I have about a 2 game advantage on this bet with Houston being the team predicted with more wins.

mY estimated predictions include the Vegas provided over under wins bet in the regression formula so it is accounted for

[Waterstpub87]

I came up with a method to bet team totals and developed my own lines (reviewed with prior years and so far so good)

i have already evaluated the solo team totals when I just noticed my book offers team va team total wins

so in the above bet it does not matter how many wins either team gets to but rather Atlanta has to win two more games than Houston to win that side of the bet.

so ideally, my numbers tell me I have about a 2 game advantage on this bet with Houston being the team predicted with more wins.

mY estimated predictions include the Vegas provided over under wins bet in the regression formula so it is accounted for

It is odd to have the line itself as a regressor. I would be concerned about over fitting.

The thing with estimating it that way is you don't have a standard deviation or a probability formula, so you could simulate 1,000 seasons, or paramatrically figure out what the win % is.

You could use poisson.

[a4u2fear]

I don’t feel like poisson would be a good indicator because the probablility is 0.5 as opposed to 0.10 or less as required by poisson

[tsty]

stop betting long futures
it's a waste of money and time

you need to have such a big edge for it to be even worth it locking up money that long

in short : it's penetrating retarded so stop looking

[a4u2fear]

[QUOTE=tsty;27987985]stop betting long futures
it's a waste of money and time

you need to have such a big edge for it to be even worth it locking up money that long

in short : it's penetrating retarded so stop looking[/QUOTE

its not invested capital. I can make “good faith” bets before any money is exchanged. Just tying up future funds. Not an issue

[tsty]

[QUOTE=a4u2fear;27988006]

stop betting long futures
it's a waste of money and time

you need to have such a big edge for it to be even worth it locking up money that long

in short : it's penetrating retarded so stop looking[/QUOTE

its not invested capital. I can make “good faith” bets before any money is exchanged. Just tying up future funds. Not an issue

so who's letting you bet on credit and why can't you just use that on other games?

the facts don't change just because you're using credit

[tsty]

http://www.aussportsbetting.com/2010/07/21/kelly-criterion-accounting-for-the-time-value-of-money/

just muck around with the numbers and you can see why it sucks

[danshan11]

I think anytime your model is saying there is a 2 or 3 game difference from Vegas, you probably need to work on your model a bit

[a4u2fear]

All plays listed for this season:

https://www.sportsbookreview.com/forum/nfl-betting/3486234-over-under-win-totals.html

well see how this goes