Austin's NFL Spreadsheet Trial Run

dwaechte · 10-27-07, 04:21 PM

Hey, not sure what all this involves but it certainly seems like you've done your homework. Maybe theres more of an explanation in another thread, but what exactly is the meaning of the number next to the team with the "edge"? Is it how much you're assuming they SHOULD be favoured by? That was my presumption but then the games you've bolded don't seem to back that up.

Anyways, as I said I'm very interested in hearing more on this and seeing how it works out.

austintx05 · 10-27-07, 09:34 PM

Originally posted by dwaechte

Hey, not sure what all this involves but it certainly seems like you've done your homework. Maybe theres more of an explanation in another thread, but what exactly is the meaning of the number next to the team with the "edge"? Is it how much you're assuming they SHOULD be favoured by? That was my presumption but then the games you've bolded don't seem to back that up.

Anyways, as I said I'm very interested in hearing more on this and seeing how it works out.

The edge is basically a predicted margin. I have a lot of factors that go into the spreadsheet and I don't know what you mean by that don't back that up?

austintx05 · 10-27-07, 09:37 PM

Week 8 Totals

1. Cleveland @ STL 45 ...48.53
2. Detroit @ Chicago 45 ...44.9
3. Indianapolis @ Carolina 44.5 ...43.42
4. NY Giants @ Miami 47.5 ...51.64
5. Oakland @ Tennessee 40 ...42.17
6. Philadelphia @ Minnesota 37.5 ...37.16
7. Pittsburgh @ Cincinnati 48 ...49.08
8. Buffalo @ NY Jets 38 ...41.43
9. Houston @ San Diego 45 ...45.83
10. Jacksonville @ Tampa Bay 32 ...32.88
11. New Orleans @ San Francisco 40 ...38.83
12. Washington @ New England 47.5 ...46
13. Green Bay @ Denver 42.5 ...43.25

Week 8 Final Card

1. Tennessee -7
2. Jacksonville +3.5
3. Green Bay +3
4. Cleveland @ St Louis Over 45
5. NY Giants @ Miami Over 47.5
6. Buffalo @ NY Jets Over 38

etothep · 10-27-07, 10:38 PM

doesn't appear as though you're taking jax's starting qb into consideration here

and if the edges are predicted margins, shouldn't you be going with the largest differences (cle, ind, & ten)?

austintx05 · 10-27-07, 10:51 PM

Originally posted by etothep

doesn't appear as though you're taking jax's starting qb into consideration here

and if the edges are predicted margins, shouldn't you be going with the largest differences (cle, ind, & ten)?

there are different road edge's and home edge's....different parameters
injury is already in the line and numbers

austintx05 · 10-28-07, 06:25 PM

2-3 with GB pending.....the interesting fact is if you played the sides where the sheet shows a 7 pt+ edge, it would be 4-3 so far....still taking notes and tweaking....

austintx05 · 10-28-07, 10:25 PM

one last thing I will add, is I am trying to determine exactly how many points off my line needs to be v the vegas line to declare it a play.

By saying 7 points or more, it would become more selective, but I know as handicappers if we feel the line is off by about 3 we deem it as a play...

again, still taking notes....

durito · 10-29-07, 01:20 AM

Originally posted by austintx05

one last thing I will add, is I am trying to determine exactly how many points off my line needs to be v the vegas line to declare it a play.

By saying 7 points or more, it would become more selective, but I know as handicappers if we feel the line is off by about 3 we deem it as a play...

again, still taking notes....

this is a question you will need time to answer. the amount the line is off is irrelevant if your predictions are not solid. once you've tracked your predictions for a decent amount of time you will have the data to see what amount of differential produces a big enough % play for you to make it.

the biggest issue with the nfl is there are so few games, it can take a few years to figure the accuracy of your rankings.

good job making a rating system, it's something i think every handicapper should have without a doubt.

austintx05 · 10-29-07, 01:52 AM

Originally posted by durito

this is a question you will need time to answer. the amount the line is off is irrelevant if your predictions are not solid. once you've tracked your predictions for a decent amount of time you will have the data to see what amount of differential produces a big enough % play for you to make it.

the biggest issue with the nfl is there are so few games, it can take a few years to figure the accuracy of your rankings.

good job making a rating system, it's something i think every handicapper should have without a doubt.

thanks, I have my hypothesis...I will be backtesting it over the last 10 years to give it 3 separate tests...we will see if it passes.

austintx05 · 10-29-07, 10:16 AM

*fyi, hypothesis - 6pt edge or greater for sides, 3 pt or greater for totals

dwaechte · 10-29-07, 01:11 PM

Originally posted by austintx05

The edge is basically a predicted margin. I have a lot of factors that go into the spreadsheet and I don't know what you mean by that don't back that up?

Ok thats what I thought. All I meant was what eptothep said, that the games you were playing weren't the ones with the largest difference.

austintx05 · 10-29-07, 03:30 PM

Well according to my hypothesis on sides 3-1(GB Pending), totals 1-2...still of lot of backtracking to do....

keep in mind, no money is being placed on these....

austintx05 · 10-29-07, 11:17 PM

well a good week for sides 4-1, totals 1-2...

regularguy · 10-30-07, 02:03 AM

Hey austin. I was keeping my eye on this. Excellent work. I would be interested in seeing your chart this week. I'll be looking around. Thanks.

This caught my eye, because I have been thinking about how I might reduce my jumble of stuff to a more numerical approach. If I ever get organized ....

austintx05 · 10-30-07, 02:07 AM

Originally posted by regularguy

Hey austin. I was keeping my eye on this. Excellent work. I would be interested in seeing your picks this week. I'll be looking around. Thanks.

thanks reg guy...I am interested as well...haha