Players and coaches on non-playoff teams are more concered with tee times than defensive assignments. I don't think the posted totals will catch up to reality.
April=overs in the NBA
- Thread starter Valuist
- Start date
Another April trend: Golden State has lost 14 straight games in April going back a couple seasons.
Went back to the 1995-96 season thru 2000-2001:
336 overs 274 unders 55% over
You'd better be pretty sure on the unders in April (regular season of course) to bet them. And this composite includes teams like Miami, NY and Charlotte who still were playing defense in April of those seasons. Toss them out and I bet the number is probably at least 58%.
336 overs 274 unders 55% over
You'd better be pretty sure on the unders in April (regular season of course) to bet them. And this composite includes teams like Miami, NY and Charlotte who still were playing defense in April of those seasons. Toss them out and I bet the number is probably at least 58%.
I once did a study that showed 2/3 of games on the last day of the season involving teams that were either out of the playoffs or locked into a playoff spot went over. Sample size wasn't that big, maybe about 50 games. Haven't followed how it's done over the last few years.
I think 600 games is a legit sample size by anyone's criteria. Taking out Charlotte, Miami, New York and Utah, who were a combined 37-55 under, the new #'s are:
299 overs 219 unders
57.7%
I think most professionals would take 57.7% over a 500 plus game sample size; especially when no further handicapping is needed.
299 overs 219 unders
57.7%
I think most professionals would take 57.7% over a 500 plus game sample size; especially when no further handicapping is needed.
Sounds like your sample size is great but I have to HEAVILY criticize taking certain teams to increase your percentage. You can't just go back look at some sets of data that pulled the percentage down and just arbirarily dismiss them like that.
How do we know what teams to take out of the sample for this year coming up?
How do we know what teams to take out of the sample for this year coming up?
OK, toss nobody out. Its still 55% on EVERY OTHER TEAM. Like I said earlier, 55% with no other capping involved is pretty damn good. What kind of percentage do you expect. With a sample size this big I seriously doubt there's any angle thats over 60%.
Interesting angle.
While I agree we cannot LEAVE OUT certain teams just because they diminsh the returns, one method might be to NOT bet on the games that involve teams with the lowest scoring averages (or best defenses -- I'm not sure which statistic would be better).
Since the rotation of low-scoring teams changes from year to year, you could examine stats and see which teams have the lowest scoring games -- and throw them out.
I'm not sure why Utah and Charlotte were included in the above data (about leaving them out) since those are not generally known as low scoring teams (average I'd say). What is important is to see if you were to leave out the bottom few teams whould this kick the percentages from 55 percent up to 56 or 57 percent?
Just a thought.
Nolan Dalla
While I agree we cannot LEAVE OUT certain teams just because they diminsh the returns, one method might be to NOT bet on the games that involve teams with the lowest scoring averages (or best defenses -- I'm not sure which statistic would be better).
Since the rotation of low-scoring teams changes from year to year, you could examine stats and see which teams have the lowest scoring games -- and throw them out.
I'm not sure why Utah and Charlotte were included in the above data (about leaving them out) since those are not generally known as low scoring teams (average I'd say). What is important is to see if you were to leave out the bottom few teams whould this kick the percentages from 55 percent up to 56 or 57 percent?
Just a thought.
Nolan Dalla
Valuist --
Great info. I'm definitely playing this trend starting next Monday using ALL teams.
I agree with PerpetualCzech; from a statistical perspective I wouldn't try and take any teams out, in trying to go from 55% to 57% or whatever you'll probably ruin your model and end up with 45% or something. 55% is good enough for me.
Pigs get slaughtered.
Thanks for posting.
- Jon
Great info. I'm definitely playing this trend starting next Monday using ALL teams.
I agree with PerpetualCzech; from a statistical perspective I wouldn't try and take any teams out, in trying to go from 55% to 57% or whatever you'll probably ruin your model and end up with 45% or something. 55% is good enough for me.
Pigs get slaughtered.
Thanks for posting.
- Jon
