70kgman NBA/NCAAB totals season long thread

[p00t13tan]

darn it missed your 2nd half play ended up playing dan kelly's 2nd quarter play and it was a bust! following no one but u for now on

[cc440unn]

hte overtime kill us....

[sbearc820]

Any model plays today 70kgman? Nice pick yesterday on the under - I missed it!

[SparJMU]

All right guys remember, we all agree one of the things that is great about this thread is that it isn't clogged up with a bunch of commentary about how the games unfold, little details about made three-pointer at the end of the 3rd quarter, and people chiming in whether they followed or missed the play.

[cint2cint]

Nice work once again

[70kgman]

Just FYI, there is really only one model play I am seriously considering playing tonight. It is one I will be waiting on until later in the day. Not the right time to lock it in yet.

[mrtsummit]

Thanks for update

[mrtsummit]

Is there any software thats available to help one model these games, such as yourself?
Or is this a model you developed on your own?

[70kgman]

Is there any software thats available to help one model these games, such as yourself?
Or is this a model you developed on your own?

I just use excel spreadsheet. So it is all specific numbers from the games that I manually input into the spreadsheet every day, along with original idea's involving the math to get the angle/numbers I want. It is not any kind of auto prediction software or anything like that. I wouldn't have a clue how to do anything like that.

[ManBearPig]

Hey 70KG...quick question. I've been tweaking my model, which I'll probably never stop doing, and I'm trying to determine the best metrics to include that will help predict or give an indication of what to expect when two teams get together. I've got numbers all over the place that I'm trying to organize and keeping adding more.

My latest thing is I'm wondering if incorporating the EFG% could be of benefit. The other day I put together a separate query that will spit out numbers based purely on pace and adjusted pace and I'm thinking I can use along with the EFG% to try and determine the final score.
Have you played around with or do you incorporate this in your calculations, just trying to see if I'm wasting my time with this as I'm sure I'll need to incorporate additional data as well, but I haven't gotten there yet. I just ordered Math on Paper and Mathletics today so I'm hoping there will be some stuff in there to spark more ideas as well.

Thanks again

[70kgman]

Hey 70KG...quick question. I've been tweaking my model, which I'll probably never stop doing, and I'm trying to determine the best metrics to include that will help predict or give an indication of what to expect when two teams get together. I've got numbers all over the place that I'm trying to organize and keeping adding more.

My latest thing is I'm wondering if incorporating the EFG% could be of benefit. The other day I put together a separate query that will spit out numbers based purely on pace and adjusted pace and I'm thinking I can use along with the EFG% to try and determine the final score.
Have you played around with or do you incorporate this in your calculations, just trying to see if I'm wasting my time with this as I'm sure I'll need to incorporate additional data as well, but I haven't gotten there yet. I just ordered Math on Paper and Mathletics today so I'm hoping there will be some stuff in there to spark more ideas as well.

Thanks again

For the most part, the lines on the O/U are compromises of the two teams style of play. A good way of predicting what to expect when the two specific teams get together would be keeping a history of the pace of each teams past games, along with a note of what the O/U line of the game was. That is kind of a roundabout way of finding out if a team regularly makes their own pace and forces their style of play onto the opponent, adapts to other teams pace, or usually just ends up in a compromise of both styles. Is that what you meant with that question?

As far as EFG%, I have never personally used it, I am not even sure where I would find that stat to be honest. I could see it being relevant in certain situations, teams with great EFG% would probably be a bad match-up for a team that plays a zone style of defense, or with small guards who don't defend well. But I guess that is more (extremely refined) situational capping than something I would know how to effectively plug into a model. You can only learn more from experimenting with it though.

Sounds like some elaborate stuff you are working on. To be honest, I would consider my model pretty elementary in it's math and spreadsheet knowledge. I would attribute it's success more toward the creativity involved. Kind of like a really great song that is based from nothing but two simple chords played over and over.

[Qtip]

All right guys remember, we all agree one of the things that is great about this thread is that it isn't clogged up with a bunch of commentary about how the games unfold, little details about made three-pointer at the end of the 3rd quarter, and people chiming in whether they followed or missed the play.

Looks like someone didn't pull the trigger, and missed out.

[70kgman]

Was waiting on this one to see if I could get a better number later in the day. But I am getting impatient keeping an eye on this stagnant line. Playing it now.

3/30

Dallas / LA Clippers Under 199

------
NBA Model: 118-77-3
2H wagers: 66-35

[aznknightt]

it's at 199.5 now

[70kgman]

Houston / Philadelphia Under 103.5 2H

[70kgman]

Cleveland / Charlotte Under 95 2H

[nickos86]

Good luck champ. First time I have remembered to check your thread at half. With you on these three

[70kgman]

Memphis / Golden State Under 103 2H

[ManBearPig]

For the most part, the lines on the O/U are compromises of the two teams style of play. A good way of predicting what to expect when the two specific teams get together would be keeping a history of the pace of each teams past games, along with a note of what the O/U line of the game was. That is kind of a roundabout way of finding out if a team regularly makes their own pace and forces their style of play onto the opponent, adapts to other teams pace, or usually just ends up in a compromise of both styles. Is that what you meant with that question? As far as EFG%, I have never personally used it, I am not even sure where I would find that stat to be honest. I could see it being relevant in certain situations, teams with great EFG% would probably be a bad match-up for a team that plays a zone style of defense, or with small guards who don't defend well. But I guess that is more (extremely refined) situational capping than something I would know how to effectively plug into a model. You can only learn more from experimenting with it though. Sounds like some elaborate stuff you are working on. To be honest, I would consider my model pretty elementary in it's math and spreadsheet knowledge. I would attribute it's success more toward the creativity involved. Kind of like a really great song that is based from nothing but two simple chords played over and over.

Hey, thanks for answering my question, what you say about the pace makes sense and I think the only change I need to make is either saving past pace numbers or calculating it on my own so that I can back test and run a regression on it.

I'm mainly trying to see if there's any correlation between the EFG% (you can calculate or pull from a number of sites) and the predicted pace that I've come up with. At the same time you bring up a good point about applying the KISS principle and only using what you need an not cluttering your model up, which is what I may have right now. I've got a lot of stuff going on and I'm trying to find the data that makes the most sense and improving on that.

[70kgman]

These games are getting really difficult to predict down the stretch. Probably because so many of them are essentially meaningless for a lot of the teams. Seems like a lot of these games have turned into street ball games with hardly any defense. I have noticed so many games over the last week or so where both teams consistently shoot over 50% from start to finish. That was a rarity earlier in the year.

Might be a good idea to pull the plug on these model/2H plays pretty soon and sit the last 7-10 days or so of the regular season out. I do remember going 0-5 the last week of the season last year with this model. These erratic last couple weeks of the season seem better suited for the non math capper.

[Dan Kelly]

These games are getting really difficult to predict down the stretch. Probably because so many of them are essentially meaningless for a lot of the teams. Seems like a lot of these games have turned into street ball games with hardly any defense. I have noticed so many games over the last week or so where both teams consistently shoot over 50% from start to finish. That was a rarity earlier in the year.

Might be a good idea to pull the plug on these model/2H plays pretty soon and sit the last 7-10 days or so of the regular season out. I do remember going 0-5 the last week of the season last year with this model. These erratic last couple weeks of the season seem better suited for the non math capper.

That is a great observation - nothing wrong with watching a little baseball.

[70kgman]

Creighton / Oregon Under 74.5 2H

[ManBearPig]

Creighton / Oregon Under 74.5 2H

Dam...missed it

[dume walker]

These games are getting really difficult to predict down the stretch. Probably because so many of them are essentially meaningless for a lot of the teams. Seems like a lot of these games have turned into street ball games with hardly any defense. I have noticed so many games over the last week or so where both teams consistently shoot over 50% from start to finish. That was a rarity earlier in the year.
Might be a good idea to pull the plug on these model/2H plays pretty soon and sit the last 7-10 days or so of the regular season out. I do remember going 0-5 the last week of the season last year with this model. These erratic last couple weeks of the season seem better suited for the non math capper.

Thanks for the heads-up, K. I've been raising my units because of your recent runs. Realized I now need to lower them for this last stretch of regulation. Unfortunately, I started lowering them with your Chreighton/Oregon pick. Nice call there.

[70kgman]

Thanks for the heads-up, K. I've been raising my units because of your recent runs. Realized I now need to lower them for this last stretch of regulation. Unfortunately, I started lowering them with your Chreighton/Oregon pick. Nice call there.

I guess I will say my tentative plan at the moment is to cautiously continue through the weekend, and then sit out those last 8 days of the regular season. I have still been winning for the most part, I just cant say I particularly like the way I have been winning games lately. I have a feeling if I keep trying to predict these unpredictable last couple weeks of the season games, I could potentially hit a losing patch.

[DemoralizdDreamr]

Clips under is going down to the wire, i got 3 teamer hangin g on it.

[DemoralizdDreamr]

i couldnt hedge it in time another tough loss

[70kgman]

61 point 4th quarter killed it after projecting to go under for 95% of that game.

3-2 overall on the day. No plays tomorrow.

[ManBearPig]

Your comment about the scoring lately got me thinking and I think this is spot on...

Below is the average scoring by month since 2006 - I plugged in different years but the numbers remain consistent.

Notice the scoring drop during the playoffs. I might even consider using these numbers in my calculations like one would use averages because I think they are important.

MONTH_NUM AVG_PTS
--------- ----------------------
01 200.48
02 200.26
03 201.76
04 198.17
05 194.19
06 187.72
10 195.03
11 198.69
12 197.92

Wanna know the Days of the week? Pay attention to Wed...since 01/2008. Again changing the dates the data remained consistent and points fluctuated by the tenths of a pt.

DAY_NUM WEEK_DAY AVG_PTS
------- --------- ----------------------
1 Sunday 197.76
2 Monday 198.24
3 Tuesday 199.51
4 Wednesday 202.01
5 Thursday 196.57
6 Friday 198.08
7 Saturday 198.64

[70kgman]

Interesting stuff Manbear. It almost seems like the linesmakers are aware of that fact as well. Lots of overs have been showing up in my model lately. But when I compare the O/U line against what I come up with for a rough estimate, they almost all seem way too adjusted to be playable. I voided five 1H over model plays yesterday because they all seemed way too adjusted. They still went 4-1 though, lol.

[cint2cint]

nice job , keep it up.

[70kgman]

LA Kings / Vancouver under 5 (+110) is a good spot play tonight in my opinion. This game has a 2-1 type of final score written all over it.

[JOHON8]

LA Kings / Vancouver under 5 (+110) is a good spot play tonight in my opinion. This game has a 2-1 type of final score written all over it.

Thanks for the tip...

[sbearc820]

Thanks 70kgman! Will you have model plays for the NBA playoffs too?

[70kgman]

Thanks 70kgman! Will you have model plays for the NBA playoffs too?

I don't think the model would really translate too well in the playoffs where the same teams are playing each other over and over. I was just planning on just looking for 2H plays throughout the playoffs.