TGoat's Total Plays

[doublec]

The latter.

That's really interesting. I find it very rare to see lines more than 5 points off the projected line, and anything more than 3 is considered a major play. Today for instance, I had Knicks/Hawks at 204, Sac/Lal at 212, and Orlando/Minn at 198 (my play of the day along with the Knicks, and fortunately both were winners).

My biggest issue is I can't seem to get my system to give any scenarios where a big favorite (Lakers and Hawks tonight) ever cover. I suppose the limited scope of my early project and the very few variables in it are to blame.

[TGoat]

That's really interesting. I find it very rare to see lines more than 5 points off the projected line, and anything more than 3 is considered a major play.

A few posts back I mentioned that I did much better earlier in the season. Not coincidentally perhaps, earlier in the season I found quite a few more large discrepancies between my line and the official line. Those "differences" have since tightened up considerably.

This could be one of the reasons (among many) that my system (for totals) has started to tail off.

The sides apsect of the system, something I didn't even bother looking at till recently, seems to be holding it's own. Here is a chart of how the sides have done according to various line discrepancies.

Only time will tell if the sides start to tail off as well.

[doublec]

Wow. That 20+ side is incredibly enviable. Clearly you are doing something correct. Nice work!

[TGoat]

1/2/10

12 Games 9 Plays

cle/nj 192.5, 199, 6.2 (Over)
char/mia 187, 192, 5.4 (Over)
san/wash 201.5, 208, 6.5 (Over)
min, ind 207.5, 198, 9.4 (Under)
tor/bos 193.5, 205, 11.2 (Over)
orl/chi 193, 199, 5.8 (Over)
okl/mil 196.5, 206, 9.7 (Over)
mem/pho 228, 222, 6.2 (Under)
gs/port 211.5, 206, 5.9 (Under)


SIDES:

char +6
min +4.5
orl -3.5
okl +1
gs +4.5

[doublec]

72 in the first quarter of the GS Portland game? Ridiculous!

[doublec]

But the play stays under! Pace (Poss per game) may be the most important stat in NBA betting. Nice call on the under.

[TGoat]

1/2/10

12 Games 9 Plays

cle/nj 192.5, 199, 6.2 (Over) (LOSS)
char/mia 187, 192, 5.4 (Over) (WIN)
san/wash 201.5, 208, 6.5 (Over) (LOSS)
min, ind 207.5, 198, 9.4 (Under) (LOSS)
tor/bos 193.5, 205, 11.2 (Over) (WIN)
orl/chi 193, 199, 5.8 (Over) (WIN)
okl/mil 196.5, 206, 9.7 (Over) (WIN)
mem/pho 228, 222, 6.2 (Under) (LOSS)
gs/port 211.5, 206, 5.9 (Under) (WIN)

5-4 (56%) +.55 Units

50-59-1 (46%) -13.55 Units


UNDERS:

1-2 (33%) -1.09 Units

9-14-1 (39%) -5.82 Units


SIDES:

char +6 (WIN)
min +4.5 (LOSS)
orl -3.5 (LOSS)
okl +1 (LOSS)
gs +4.5 (LOSS)

1-4 (20%) -3.09 Units

12-14-1 (46%) -3.09 Units

...

[TGoat]

But the play stays under! Pace (Poss per game) may be the most important stat in NBA betting. Nice call on the under.

You could be right. Where do you get your stats?

I'd also be curious to find out how many points each team scores in the paint. PIP, points in the paint, may be one of the most important things that separates winning teams from the also-rans.

The bad teams just can't (or don't try very hard) to score in the paint. The good teams work the ball around and try to take more high percentage shots.

If I were a basketball coach, 90% of my time would be spent working on plays that try to score inside. 3 point shots are fine when you're running out of time, or one of your best 3-point shooters is wide open, but a lot of teams don't even try very hard to work the ball inside. They just fire up a three or a bad two and hope it goes.

Bad teams win (or cover) if their outside shots are connecting. If they're not, they are really going to struggle against a good team.

[TGoat]

1/3/10

5 Games 3 Plays

ind/ny 207.5, 182, 25.4 (Under)
char/cle 185.5, 200, 14.4 (Over)
dal/lake 200, 205, 5.1 (Over)

SIDES:

ind +9.5
char +11.5

[lcf]

Another important stat is FT. Check Boston's game box score against Toronto and you will easily find what made them winners...

[TGoat]

Another important stat is FT. Check Boston's game box score against Toronto and you will easily find what made them winners...

FT% (Pre matchup)

Tor 77.1
Bos 75.9

Am I missing something here?

[doublec]

You could be right. Where do you get your stats? I'd also be curious to find out how many points each team scores in the paint. PIP, points in the paint, may be one of the most important things that separates winning teams from the also-rans. The bad teams just can't (or don't try very hard) to score in the paint. The good teams work the ball around and try to take more high percentage shots. If I were a basketball coach, 90% of my time would be spent working on plays that try to score inside. 3 point shots are fine when you're running out of time, or one of your best 3-point shooters is wide open, but a lot of teams don't even try very hard to work the ball inside. They just fire up a three or a bad two and hope it goes. Bad teams win (or cover) if their outside shots are connecting. If they're not, they are really going to struggle against a good team.

Get my stats from teamrankings.com How about you?

Yeah, I certainly agree with higher percentage shots. Obviously, that leads to higher offensive efficiency, and thus more wins. Have you ever used effective FG%? I'm considering incorporating that soon. I'm sure you know, but for the benefit of those that don't, it adds the value of a 3 point shot to the overall FG percentage. Its certainly not a miracle stat, but when you see the top five teams in this stat (Phoenix, San Antonio, Boston, Cleveland, and Orlando) compared to the last 5 (NJ, Chicago, Detroit, Minnesota, and Indiana), its certainly a worthwhile sim stat.

[doublec]

And I don't know if you were watching the Portland/GS game last night, but if you didn't, the first quarter was a high possession quarter, with a very fast pace, and the Warriors were up ten (playing at their favorite pace). When Portland slowed the game into their pace, they took control and never looked back. Possessions are just so important.

[TGoat]

Get my stats from teamrankings.com How about you?

I looked at that site. Are you a "premium" user or do they offer free stats that you use?

I make my own stats.

[doublec]

I looked at that site. Are you a "premium" user or do they offer free stats that you use?

I make my own stats.

Free.

http://www.teamrankings.com/nba/stats/

This has points in the paint too. Just about everything.

I'm pretty impressed you make your own stats. That is some dedication.

[TGoat]

1/3/10

5 Games 3 Plays

ind/ny 207.5, 182, 25.4 (Under) (LOSS)
char/cle 185.5, 200, 14.4 (Over) (LOSS)
dal/lake 200, 205, 5.1 (Over) (WIN)

SIDES:

ind +9.5 (LOSS)
char +11.5 (WIN)

I'm officially closing down this process. I no longer see the point. The system isn't making any money, and seems to be a lesson in futility. Yes, it could turn around and start winning...if it does, it does, but I want to start looking into other things. [I will keep following it of course, but I won't be making any more public posts.]

What I consider to be the number one handicapping factor can't be programmed into a spreadsheet...motivation. I might write more on this subject later.

Feel free to post here if you have any questions or comments.

Thanks to all who followed along.

[doublec]

TGoat, why not keep this going with just your high percentage plays (see your first post of this page)? Clearly the system isn't worthless, you just might have too many games as "plays." And I'm also considering bringing in a subjective factor, determined by me depending on road-to-road games, or other factors. You should look into it.

If you do decide to pull your thread, I appreciate all the help you have given me and the other "systemers." Good luck with future projects!

[TGoat]

TGoat, why not keep this going with just your high percentage plays (see your first post of this page)? Clearly the system isn't worthless, you just might have too many games as "plays."

You're right, the high percentage plays are doing well, but there are so few of them...there hasn't been one in four or five days.

I think what I might do is just "update" the system from time to time--totals and sides. And then anyone who is interested can see how it is doing overall.

If you have any luck with your systems, keep me posted. I have an lot of different irons in the fire, and I'm interested in other peoples methods and ideas. It's a learning process, and that's why I'm here.

[bolekblues]

Hopefully you will come up with something. I think about making my own excel-based model, then we could compare variables etc.
GL

[doublec]

You're right, the high percentage plays are doing well, but there are so few of them...there hasn't been one in four or five days.

I think what I might do is just "update" the system from time to time--totals and sides. And then anyone who is interested can see how it is doing overall.

If you have any luck with your systems, keep me posted. I have an lot of different irons in the fire, and I'm interested in other peoples methods and ideas. It's a learning process, and that's why I'm here.

If I were you I would just start a new thread with 20+ picks. I think you could become quite the legend.

[TGoat]

If I were you I would just start a new thread with 20+ picks.

Two strong plays last night, and they both won. There are three more for tonight. I'll post these strong plays for a while and see how it goes. The percentage number is the predicted chance of winning and the point differential next to that.

wash +13.5 (68%, 16)
mem +5.5 (74%, 23)
pho -4.5 (68%, 18)


[Last night's plays]

ind -9.5 (68%, 17)
lakers -7(68%, 16)

[doublec]

Keep up the hard work TGoat! Keep rolling with those strong plays.

[bustabook]

tgoat... check your gmail.. let me know what you think about the new model.

[TGoat]

wash +13.5 (68%, 16) (LOSS)
mem +5.5 (74%, 23) (LOSS)
pho -4.5 (68%, 18) (WIN)

Not too strong tonight. 1-2. I'll keep rolling with them for a while. I don't think there will be that many.

[doublec]

TGoat do you feel that any system with an objective component could be successful in the long term?

[TGoat]

TGoat do you feel that any system with an objective component could be successful in the long term?

I'm not quite sure what you mean by "objective component," but "long term" is the key phrase.

A system is usually just numbers, and athletes have emotions. You can't really put a number on emotion, because you don't know from day to day what those emotions will be. If two or three key players are down and out for some reason, that brings the whole team down. If they're up, every one else gets up too. Motivation is key, but I don't see how that can be entered into a system.

Systems always lag. Say a team gets hot and is on fire, a system is going to lag slighty with that information (especially if you're using some type of averages.) Same when they go cold. So at the outliers the system will perform poorly, because it hasn't "caught up," but in the middle stages it should do OK. In other words it's going to be slow to catch a trend--both when it begins and when it ends.

I have tracked dozens of different ideas on a multi-year basis only to find that it had one or two great years and then disaster in three or four years

A successful system (if there is such a thing) has to be grounded in up-to-date logic flow--it has to find errors in the line. Finding errors in the line these days is extremely difficult to do.

I think the best handicappers in the world understand and have information about the motivational underpinnings of a particular team. Are they up for a game? Down for a game? Looking past a game? Fighting amongst themselves? Out for revenge? Overly tired or broken down? Outclassed and intimidated? Nervous? Hated by their opponents? This is the information that can make you grab -7 because it should be -21 or take +16 when it should really be "pick-em." Bookies don't always know this stuff, especially in the smaller, less followed conferences. In the NFL most of this information is out there making the lines extremely efficient.

Maybe the best system is one that combines math and motivational factors that require judgment. When they coincide--you have a strong play.

Whatever method you use, I think it must be successful for a multi-year test period. Anybody doing anything can run good for one season. The real test is longevity.

[doublec]

I definitely agree. I think numbers by themselves can be relatively accurate, but combined with a person with good judgment ability, you could have a pretty successful system. The key is the person with those skills, and the patience to ride it out long term.

[TGoat]

Jan 8

mem -3.5 (66%, 17)
pho -6 (66%, 17)
den +3 (63%, 13)

[doublec]

This seemed like a bad system night. I was with you on Pho and Den, but unfortunately my big play of the night was on Milwaukee. I've gotta stay away from the Bulls, I think I have lost the last 4 I bet either for or against them.

Keep us posted.

[TGoat]

Jan 8

mem -3.5 (66%, 17) (LOSS)
pho -6 (66%, 17) (LOSS)
den +3 (63%, 13) (WIN)

...

[TGoat]

Jan 9

Mem +6

[TGoat]

Adding

Phil +4.5

[TGoat]

...

[TGoat]

...

[TGoat]

Horrible Internet connection. Is there a way to delete duplicated posts?