Attempting to model NBA over/unders

[Buried_PIRATE]

Mind me asking how much ur betting each game and what college you went to

I hope this doesn't hurt my credibility, but I am only betting $10-$15 per game. I would probably bet a little bit under whatever one "standard" bet is, because some nights I have 8 plays that could fit the guidelines for a bet under my model.

I'm liking my results so far and I have a decent record that I have been trying to meticulously track in this thread.

[Buried_PIRATE]

I guess I will also take the time and try to show what I think is a "luck" factor. Theoretically picking any over/under is a 50/50, probability = 0.5 ... so assuming I only picked games by flipping a coin, and came up with my record of 47-32, the probability that I would have 47 successes or MORE out of my 79 attempts is 5.7%

IE this means I am doing MUCH better than flipping a coin (comforting I know!)

With that said, 79 attempts is really only a drop in the bucket. I'll be touting myself more if I can keep up this pace at 200. For instance, if I put up a goose egg tonight and go 0-5, my "luck factor" would be 16%

You can mess around on a binomial calculator here: http://stattrek.com/Tables/Binomial.aspx

[Buried_PIRATE]

NJN@LAL / Line: 191.5
5Day: 198 / over
YTD: 196 / over
Total:188
Grade: LOSS

CLEA@UTA / Line: 196
5day: 209 / over
YTD: 199 / over
Total: 200+
Grade: WIN

CHAR@BOS / Line: 188
5day: 193 / over
YTD: 187 / under (playing the under on boston at home has burned me so far this year, but here we are again *playing under*)
Grade: LOSS

CHI@IND / line: 190
5day: 184 / under
total: 187
Grade: WIN

updated

[Buried_PIRATE]

also adding:
Por@phx / line: 195.5
5day: 200 / over
ytd: 205 / over
total: 226
grade: win

updated

[Buried_PIRATE]

Went 3-2 last night, honestly I am disappointed in the results considering we missed 2 games by about 3 points

Record goes to 50-34

[Nookx]

Sometimes its gonna go your way by 3 points so just gotta keep grinding. Keep it up. Looking forward to your plays today.

[Buried_PIRATE]

Sometimes its gonna go your way by 3 points so just gotta keep grinding. Keep it up. Looking forward to your plays today.

Haha you're definitely right. Look at the time I posted it (2 AM... I was drunk and bitter )

I remember one game in particular where I was going to lose by 1 pt but got saved by OT

Either way I'm going to keep on grinding it out, working on a model that takes into account some more factors that I will monitor

[ManBearPig]

I think if you apply an overlay buffer I think you will avoid some of these losses. If you make sure you have a 3pt padding that Boston play would'nt have been a play and you would've saved a loss...still killer results so far. I decided to revamp my db and come up with a new line a different way, it's 8-9 so far but we'll see how it goes. I think it will be more stable in the long run.

[Buried_PIRATE]

I know, I definitely got greedy on the Boston play. I have been looking at 2 more numbers that I am still monitoring their usefulness for and they seemed to indicate that it was a light play as an over... bottom line is I was greedy. Also still working on my max overlay, when I am off by 8-9 off the line I am doing good, over 10 doing terrible... came up with an over 10 play today so I guess we'll see...

Also I am looking into instead of doing past 5 games, past 5 games vs similar teams. I would need some help to come up with a list of which teams are similar though... any input here would be appreciated.

NO@CHAR / Line: 183.5 (I saw it already moved sorry... I think I would still play it tho!)
5Day: 193 / Over
YTD: 189 / Over

NJN@POR / Line: 183.5
5Day: 195 / Over
YTD: 195 / Over (this is the line I am talking about... historically when I am off by > 10 has not been so great... so maybe I am missing a KEY element in this game, but if not going over)


HOU@ATL / Line: 203.5
5Day: 201 / Under
YTD: 200 / Under

DAL@MEM / MIA@CHI are both no plays for me... both are about 1 point off the line. I am looking at a new model which pulls these both as plays but it is still in the testing phase

I am leaning toward adding

SAC@DET / LINE: 197.5
YTD: 197

New model I was looking at pulls this game up at 204... and I am verrrry curious. I think I will lay off it as a bet though, just something to monitor.

[Avatars]

are you filtring out back-2-back games? Would be worth a shot to try it, IMO
Anyway, happy to see your results
GL tonight

[ManBearPig]

Don't look at similar teams try and get past performances between the teams. I try to pull the average total for teams for the past 5 match-ups as part of the formula. It's not an exact always and I'm still trying to find the best way to use it, but it gives me a good idea of recent trends between these two teams to take into consideration.

It's been a few years since I took prob/stats in college so I'm really trying to brush up on using this in a relevant way to find any other edges. I just got the book Conquering Risk that I've heard is really good and I can see why, lots of helpful info that noobs like us don't think about...I would highly recommend.

[Buried_PIRATE]

are you filtring out back-2-back games? Would be worth a shot to try it, IMO
Anyway, happy to see your results
GL tonight

I'm not right now. Is there proven info that on average teams with B2B score lower etc? I'll have to look at the differences sometime and hopefully can find the stats readily available. Any detailed NBA stats sites are always good...

I am adding the Sacramento line as a play.

4 plays tonight, currently 50-34. Hopefully we get some!

[TWENTYFINGAZ]

Sac over sounds good to me

[joanapoker]

Pirate,

do you assume line movement during the day?

nice work so far!

[Buried_PIRATE]

Pirate,

do you assume line movement during the day?

nice work so far!

I usually make my picks the day before and try to get in on the line as soon as possible. My hope is that the line will move in my favor because I'm on the right side

Sometimes it does, sometimes it doesn't... I haven't been tracking it but I will start to in the future. For instance, I believe two days ago I was on MEM@DET on the line at 192, I predicted over 198 and took the over. When I revisited the lines again, it had moved to 195!

On the flip side, I was on the Hawks game under 205.5, because I predicted 201, the line moved against me to 206 and the total was 205. I won that bet but in that case, I would have also gotten in on it at 205,204.5,204, and those would have been losses.

So in short, not right now. I am adding more and more checks into how I decide plays though and I hope it is helping. I also can be a bit of an action junkie and I like to make as many plays as I can justify

[joanapoker]

thx for the fast reply

in the past days it's has been hard to be on the UNDER side!! damn it....today looks like an Over night again....

[Buried_PIRATE]

NO@CHAR / Line: 183.5
5Day: 193 / Over
YTD: 189 / Over
Total: 169
Grade: LOSS

NJN@POR / Line: 183.5 (If you got in at a later line this game could be a loss for you)
5Day: 195 / Over
YTD: 195 / Over
Total: 185
Grade: WIN

HOU@ATL / Line: 203.5
5Day: 201 / Under
YTD: 200 / Under
Total: 216
Grade: LOSS

SAC@DET / LINE: 197.5
YTD: 197 [Played Over]
Total: 224
Grade: WIN

2-2 on the day... brings the record to 52-36

[Buried_PIRATE]

LAL@LAC / Line: 197
5Day: 206 / Over
YTD: 199 / Over

Only one game today so I tried running a multiple regression adding in pace, that came out to 200. I'm going to back test that model on the past couple games when I have some time

BOL on your plays today

52-36 from 12/28

[Buried_PIRATE]

LAL@LAC / Line: 197
5Day: 206 / Over
YTD: 199 / Over
Grade: LOSS

52-37... I might take tomorrow off and not cap

MLK baby

[Nookx]

Lots of money to be made tomorrow. Why would u take it off?

[Buried_PIRATE]

Lots of money to be made tomorrow. Why would u take it off?

I was just lying to myself. Working on the predictions now

[Buried_PIRATE]

I'll take a stab at MLK day NBA game predictions... I really need to go back and back test some of the new numbers I'm looking at...

Either way I am on:

TOR@NO / Line: 193 **Playing Over**
5Day: 204
YTD: 197

MIL@HOU / Line: 193 **Playing Over**
5Day: 205
YTD: 195

SAC@ATL / Line: 201 **Playing Under**
5Day: 207
YTD: 196

CHAR@PHI / Line: 189 **Playing Over**
5Day: 200
YTD: 193

NJN@GS / Line: 205 [UNDER though very close to a no play... will be a game time decision based on line]
5Day: 211
YTD: 200

4 Overs, 1 Under.. hopefully the poster that claimed that all the games tomorrow said they were going UNDER is wrong (except about the ATL game )

Either way, we have been maintaining a strong record thus far, I hope we keep on rolling

52-37

EDIT: NJN@GS could be dangerously close to a no play for me. My numbers say the game has a moderate pace with both teams allowing tons of points. They aren't prolific scorers though... I could see this falling a bit short, around "200." If you are a river boat gambler... I think it could go either way...

[VTranX]

wow this is gold.

[Buried_PIRATE]

wow this is gold.

Haha idk if you are being facetious... but I reallllly hope I can keep hitting at around the 59-60% clip I have been hitting. No promises but we are almost nearing 100 games in

I think math is one of the ways to beat the game so I'm hoping I can push that edge...

[VTranX]

Haha idk if you are being facetious... but I reallllly hope I can keep hitting at around the 59-60% clip I have been hitting. No promises but we are almost nearing 100 games in

I think math is one of the ways to beat the game so I'm hoping I can push that edge...

keep up the good man, i'll be keeping a close eye out.

btw: what profession can you get into majoring in statistics? just wondering.

[joanapoker]

I also think that the majority of the games today will go UNDER!

let's see

[JD426]

btw: what profession can you get into majoring in statistics? just wondering.[/quote]

There's a ton of different ways you can go with statistics/math degrees.

I'm glad to see that you cut down on the number of games that you're tracking buried. If your model only predicts 1 or 2 points away from the actual line then I don't see that as being a sufficient reason to actually play the game, but 5-6 points shows that there may be some value in the play. Keep up the good work.

[Buried_PIRATE]

You can get into all fields. When I was looking for jobs I know the Cleveland Indians and CBS were looking for statisticians

I'm adding the NJN play as an Under

I might take a closer look at the 10 PM games but that's it for now.

[Buried_PIRATE]

There's a ton of different ways you can go with statistics/math degrees.

I'm glad to see that you cut down on the number of games that you're tracking buried. If your model only predicts 1 or 2 points away from the actual line then I don't see that as being a sufficient reason to actually play the game, but 5-6 points shows that there may be some value in the play. Keep up the good work.

Yup. Definitely cut down on the number of plays... As I tracked my results I noticed that when I was under 2 pts in my predictions it was just flipping a coin there. IE losing whatever edge I get from modeling.

For instances todays CHI@MEM game I come up with between 187-189... line is 187.5

@ JoAnna... I'll give you the SAC@ATL game as an under

BOL to you guys

[Nookx]

can u post your stats for the nets game rq?

[Buried_PIRATE]

can u post your stats for the nets game rq?

NJN@GS / Line: 205
YTD: 199.3
5Day: 210.6

New model I am working on gives it 217... I haven't back tested that model yet... so given that I feel like it is a much more marginal play. Based on my previous guidelines though (having between a 3-9 pt edge) I've been doing decent... so that would make it fall into the under.

I have the data to go back and back test that new model but I probably won't have time for another week or so to really get into it.

Edit: just noticed all the early lines moved in our favor by at least .5 -> 1.5 (starting to track it)

[timmyboy34243]

there is a MLK tradition of the afternoon games going under.................

[Buried_PIRATE]

there is a MLK tradition of the afternoon games going under.................

Thats what I heard... regardless I've been sticking to my predictions and hopefully math wins today instead of some possible random chance

[timmyboy34243]

keep posting your work and tracking your record, it's much appreciated.........

[JD426]

Thats what I heard... regardless I've been sticking to my predictions and hopefully math wins today instead of some possible random chance

Yea I agree, if it's only MLK day afternoon games that tend to go under then it's probably just coincidence. However, I can see why playing in the afternoon could throw players off of their daily schedules and force them to have an even quicker turnaround than a normal back-to-back. So I would be interested to know if all afternoon games tend to average fewer points.