NBA adjusted for competition predicted score 4/13

[Kyleben]

cle-113.2(-1.6 w/o wallace)
ind-102.5

tor-104.8(-1.7 w/o baryani?)cant spell
was-100(-5.5 w/o arenas)

chi-103.7
det-102.1

char-98.7
njn-105.8(-3.8 w/o harris)

orl-90.2 ( this score includes magic w/o hedo,lewis)(-1.9 for alston)
mil-97

min-88.1
dal-113.4

noh-87.1(-1.9 w/o chandler)
hou-94.1

sac-103.9(-6.2 w/o martin)
den-113.9

lac-96.3
uta-106.4

mem-95.5
pho-106.5

okc- 89.6
por-113.7

san-102.4
gs-106.9 (-8.9 w/o crawford/ellis)

[B's Picks]

do you have a record of your score predictions vs. actual spread to solidify this? do you also include b2b, travel, rivalry?

[Kyleben]

Games that are 3+ points away from open line are hitting 61.3%, o/u are 58.6% started tracking record 12/28/08. I have it broken down into some other splits as well,for these posts i use the teams last five home games or road games,depending on where they are playing.

[spongerat]

you mean if the score you predict is 3 points away from the opening line your pick is winning 61% of the time? does this apply to o/u

[Kyleben]

you mean if the score you predict is 3 points away from the opening line your pick is winning 61% of the time? does this apply to o/u

example: today my system spat out toronto as a 4 point winner, vegas has at +2.5 that would be a spread play. same for o/u

[TORONTOBBALL]

ben wallace will not be playing for det just the heads up

[Kyleben]

ben wallace will not be playing for det just the heads up

thank you one of the few injured players i actually have decent data sample size for

[TORONTOBBALL]

not det meant to say cavs

[Kyleben]

not alot of data on the tor/was game with the injuries

[HoulihansTX]

How do you know how many points to subtract due to injuries, B2B's, road trips? Or is it your discretion, and totally arbitrary?

[Kyleben]

How do you know how many points to subtract due to injuries, B2B's, road trips? Or is it your discretion, and totally arbitrary?

Injures are calculated based on how many points they produced when they are injured in other games or dnp's(this data can sometimes be very small). I use the scores from the teams last five home games or road games for this post. I have other splits such as b2b, long road trips, b2b home an away, ect... Nothing I post on here is at my discretion

[HoulihansTX]

So you input the data, and wait for what the system spits out? If so interesting. You should give us a record and the avg amount of points the system is away from actual score. Would go a long way into giving us bettors an advantage against the books. I like the concept, keep it up.

[Kyleben]

So you input the data, and wait for what the system spits out? If so interesting. You should give us a record and the avg amount of points the system is away from actual score. Would go a long way into giving us bettors an advantage against the books. I like the concept, keep it up.

Yes I use a fairly simple algorithm that I have been testing for about 4 months, I will be keeping a running record on here for the rest of the season. from the results it has produced thus far 3+ points on o/u or spread has been the most efficient producer of bets, ot gives you a high volume and around 61% win rate thus far. But it is still in testing mode and should be used primarily as a tool for the time being, although i do typicall bet it every night

[spongerat]

you should check out justins blackbox modeling thread, it has some good info related to this type of system

[Kyleben]

21-12 3+points(9-3 today)

[King_Bookie]

Hmm this is an interesting angle on how to predict results, and its actually doing pretty good. Will watch you further on this one.

[spongerat]

no offense but its not like its unique or innovative. tons of people using modeling

[Kyleben]

not trying to be unique or innovative, just trying to win! I know alot of people use modeling, i learned it in school when they taught us how to apply it to finical markets and have been working on new models for two years to apply to betting markets