A couple of NBA systems I have learnt of elsewhere...

[happygooner]

Well, I tell a lie, one of the systems I first learned of here (although noted after that it was sweeping the forum world but being played ever so slightly differently) from Alense.

Seeing as he now no longer seems to be posting I see no harm in continuing.

From what I also understand the NBA is notoriously difficult to succeed at in the 2nd half of the season after all star break so this coupled with a progressive staking plan (that used by the infamous maria and her laying of horses) should prove to be the real acid test so to speak!

Going to try and get this out there for the first game tonight but not long now so entirely possible I will miss it.

PLEASE NOTE PEOPLE: I AM PAPER TRADING ONLY RIGHT NOW

[happygooner]

One of the systems is called TAR and below is an explanation of how it works, using NCAAB teams, that I have copied from another site.

Stats used to ascertain these are found at covers under matchups and home/away in depth.

The TAR System is a ratio-driven system that's been tweaked by some on this site for both NCAAB and NBA. TAR is an acronym for Turnovers-to-Assists Ratio. In the Elon example below, we look at the visiting team (Western Carolina in this case), and their TAR numbers on both Offense and Defense vs the NCAAB average. Then you look at the home team (Elon) and their Offense and Defense TAR against the NCAAB average. Note: you MUST look at the stats for the teams when they are Home/Away, NOT the Overall numbers, as that will skew (read: invalidate) the data. e.g. Does Virginia play the same when they’re at home AND on the road? No team really does. In our example, Elon gets a closer look because they fit the 4-0 TAR system:

	West Caro		NCAA	Elon
	Offense	Defense	Avg	Offense	Defense
Assist : TO	0.64	1.00	0.87	1.00	0.25
Assists	9	15	13.5	13	5
TO	14	15	15.6	13	20

Note: This doesn't display lined up properly, let me just tell you that the numbers in the middle (gray) column are the NCAA average (.87 = Assists-to-Turnover Ratio, 13.5 = Avg Assists, 15.6 = Avg TO's).

In the TAR system, we see that the West Carolina Offense is 23 points (hundredths) worse than the average in the NCAA for turnovers to assists (.87 - .64 = .23), and their defense is 13 points worse than average on the TAR (1.00 - .87 = .13). Again, these are ON THE ROAD numbers for West Caro, if you look at their Home stats, they will be different! Same with Elon, these are their HOME stats, as their ROAD stats are different. Clarifying the data we're seeing, a higher number than the average is better on offense (more assists to turnovers), and a lower number than the average is better on defense (more turnovers to assists, which, as a defense, you want to cause).

Now we also see that Elon is 13 points better than the NCAA average on Offense at home, and 62 points better on Defense.

That's four factors ALL to the benefit of Elon: WestCar 'O' worse + WestCar 'D' worse + Elon 'O' better + Elon 'D' better = 4-0 in favor of Elon.

Where this system has been tweaked in the past is in the ATS numbers. If Elon has won more games than lost Against The Spread, then they qualify as a play in the TAR system. They are currently 3-1 ATS, so they qualify. By extrapolation, if we find a team has a winning ATS record, we PLAY them. If they have a losing OR TIED ATS record, we FADE them, and call them "Dawgs", as shown above.

The perception here is that a tied or losing ATS team has a winning record SU, but they are OVER-valued by the lines makers. In this next example below, we see that it is indeed the case in this instance:

	W/L	ATS
East Illinois	1-3	0-2
Wis-GreenBay	3-2	1-2

Since this is very early in the season, this is a fair example of Wisc-GreenBay, but it would make anyone wary of taking an 0-2 ATS team like Eastern Illinois. East Illinois has both a bad W/L AND ATS record, BUT has played some tough teams so far this year (all on the road), AND returns ALL Five starters from last year's team that BEAT this Wisc-GB team. GB returns FOUR starters from last year's team, BUT the leading scorer from that game (and best player) graduated. Now you're giving me 14 points too? Gosh, I don't know... (note:sarcasm)

You'll see later in the season (closer to Christmas) a lot of teams that qualify as plays in this TAR System that, in spite of having a bad SU W/L record, do much better ATS.

I hope this clears up the TAR System for some of you. GL to all who use this info!

From this, I believe the first game of the evening between Detroit & Phoenix doesn't qualify.

From all of this I also found records of the system being used in the NBA but in a different way.

The best thing was to find the qualifying plays and play all single digit faves, back against the double digit faves and play every highlighted (under)dog.

[happygooner]

Ah no, I missed something. Both teams 2-0 TAR you take the dog.

[happygooner]

Detroit +2.5 @ 1.98 1 unit

[happygooner]

Originally Posted by happygooner

Detroit +2.5 @ 1.98 1 unit

+.98 so a nice start.

[happygooner]

I meant to say, starting bank of 100 points will apply for each system.

[JCtips]

It does look interesting happygooner !

Good luck with it and I will be following your picks/results closely.



JC

[happygooner]

Originally Posted by JCtips

It does look interesting happygooner !

Good luck with it and I will be following your picks/results closely.



JC

Thanks JC, that appears to be it tonight.

As yet I haven't managed to make one of these things work for me longterm but every day I learn a little more. Seen such a wealth of different systems now (including yours) that with discipline I can't fail to make this work in the end.

So shall we drink to early retirement

[happygooner]

Atlanta +12.5 @ 1.9 for 1.10 units
Detroit ML @ 2.06 for 1.10 units

I think this is correct...

[happygooner]

Right now for explanation of the other system which will begin tonight in the event I can find any qualifying games.

For an explanation lets go over to the guy who has publicised this over pretty much the entire interweb...

Copied and pasted from another source. The floor is yours Larry:


OK There seems to be alot of confusion concerning this formula. I doubt this thread will clear it all up, but I am going to give it a try.

First I will post the formula , then I will explain my interpretation of it, and then I will tell you the different variations.

Here we go. Buckle your seat belts.


Formula, as it was given to me by Walt many years ago. This is how I received word for word.


Example : Clev vs denver
Cle winning % is .600 and Denver is .057

Subtract the difference

Clev 600 - Den 057 = 543

Give 1 point for every 20% points Therefore 543 = 27, round to the nearest tenth

Home team is given 3 points

Subtract or add the difference
Clev is 37 and Den is home thus 27- 3 = 24

Then subtract or add point spread for the final number value

The spread for the game is Cle -9.5 points, thus 24 - 9.5 = 14.5

The final number must be 10 or greater for a PLAY. In this example, Clev is the PLAY.

If the spread is 10 or greater , do not play.


Another PLAY can be derived from a negative /positive computation. For example Utah/Chic 1/25/98

Chic % is 714 and Utah is 675. Thus 714 - 675 = 39 or 2 points for Chic. Chic was home 3 = 2 = 5 The spread had Chic by 5 or 5 - 6 = -1 or +1 for Utah. Utah won the game outright. The negative/positive computation must be -1, +1 or greater for the PLAY

10 steps to predicting the outcome of an NBA game:

1. Check % Clev 600 - Den 057
2. Subtract teams winning % 600 - 057 = 543
3. Find point equivalent 543 = 27 points
4. Give home team 3 pints Clev 27 - den 3 = 24 Add/subbtract difference
5. Add/subtract the point spread
6. Final number value must be 10 or greater Cle 24 - 9.5 = 14.5
7. Negative /positive computation Chic 714 - Utah 675 = 39
Chic 2 + 3 (home) = 5
Chic 5 - 6 = -1 or +1 Utah

Final number is "-" for a favorite and "+" for an underdog. Final number must be greater than 1 or less than -1

8. Don't play if selected team played the night before (No back to back games)
9. Dont play of one or more starters are out. Allow one week for return starters
10. Don't play 1st 20 games of the season or 1st 3 games after all star break
------------------------------------------------
The record for 1999 was 60-25 and for 2000 it was 73-30

Ok There it is.

Now this is how i figure it out.


If you have a home favorite with a better winning %, to me, this is pretty much straight up.

You take the higher % , subtract the lower %....divide by 20. Add 3 (for being at home) and then subtract the spread.

If this final number is above 10, it is a play on the favorite.
If this final number is negative 1 or less, then it is a play on the dog.
If the final number is anywhere in between (0-10), then it is a no play.

That's it. Straight up, simple as can be.

Now the tricky one

What to do if a home team with a lower winning % is favored.

This is where the confusion lies. THere has been much dicussion as to whether or not the dog need to come out with a number above 10 or just the fact they are a positive number is enough.

Nw I have had great difficulty interpreting how this should be played, and why.

I finally decided to base my plays selection on two emailed examples I had received from Walt.

Here they are:

-------------------------------------------------------------

Example 1:

Christmas Day 2000
Indiana at home minus 5 1/2 vs. Orlando

This is what he types , word for word, letter for letter:

Orl462-Ind429=33 or 1.5Orl-3(H)=1.5Ind-5.5(SP)= -4Ind or +4Orl -4 is > than-1 thus it is play. Orl (+5.5)Ind. Orl lost.


Example 2:
Feb 2 2001
ind home minus 4 vs Denv

This is what he wrote:

Keeping an eye on den(+4)Ind-this is the 3rd situation where one team is % better but is getting points and is not a -/+computation game. Den565-Ind444=121or 6Den-3(home-Ind)=3Den+4(spread Ind)=7Den (people would sub and think it was 3-4= -1, I add 4 because if Ind -4 then Den is +4) It seems logical that if a team is better % wise they shoulod be giving points not getting them so I have decided to watch this 3rd situation (1st-final #10 or more, 2nd - -/+comp)
---------------------------------------------------------------------

So what is the difference between these 2 plays . Both have road teams with better winning % and are dogs, but one is a play, and one he is "watching".

The only difference I can see is with example one, your final number before the spread is a negative number. In example 2 , it is not.

I ahve asked others to detrmine the difference between the 2 plays, and no one has come up with anythig.
So without Walt to answer the question (it is possible he screwed up), I have nothing else to go on , but the fact that he talks about +/- computation and in one example where it is a play, the number before the spread goes into the negative and then back to positive.

So this is how I have determined these plays. Others still disagree and believe both examples should be plays, and thus the reason I have decided to post and track them separtately rather than not at all. They have been winning, and I want people to benfit with as many winners as possible. But I also want to keep the system to its purest sense possible.
(also note...this system had only been licking out about 100 or so plays a year, and by playing every road dog with better winning %, there will be a ton of plays. Not what the formula had intended)


So these examples are how I determine the plays. One plain and simple and the other, up for debate. I may one day have a euphoria,, and all of a sudden see the equation in a whole different light and change my thoughts on this, but for now, this is how I am proceeding.

Now on to "Filtered vs. non filtered"

I am not sure if Dice is labeling the above debated plays as filtered or non filtered, but I am pretty sure what is being referred to as "non flitered" are plays on team that have injured starters , played back to back games, or are more than 10 point spread. He had found these were winnning anyway and decided to play them regardless of aformentioned filters.
I still am not.

Notes:
back to back. If the team that is a play, played last night, it is a NO PLAY. It does not care what thier opponent did yeterday. Nor does both of them playing last night cancel each other out.

Injuries:
the rules say do not play if a starter is out and to wait one week for return.
I never remember Walt waiting a week after a starter returned. And I do remember him playing on teams after they adjust to the starter being out.

So this I am playing as a judgement call.
Example, Artest is out and Sac is still covering, so I will now play them if they come up as a play. And I will not wait one week after he is back.


Maybe "wait one week for return starters" meant wait one week for them to return, and if they don't, then play as normal"


He didn't play spreads of more than 10 simply because he was afraid of "garbage time"

So there it is. This is how I have come to understand it, and this is how I am playing it.

You can dissect and twist and turn and find a 1000 different variations to come up with as many more plays as you wish. If it works, more power to us.

But for me, I prefer to keep it to its simplest and play less plays. I would rather play one 10 unit play than play 5, 2 unit plays. (At least for this part of my bankroll)


And I hate to sound like a broken record but, if this formula works for oyu and you are able to make a profit from it, please please please find it in your heart to pass some of it along to others less fortunate than yourself. Drop a few coins off at a church or charity, give a pair of gloves to a guy living on the street. Many ways to do this. And it makes all the difference in the world.
Oh, and also, make to show the ones you love, how much you love them.

Take care all,

Larry Legend

[happygooner]

Now from my experience of following these, the back to backs should definitely be avoided but injured starters not so much. Maybe more than 1 injured starter is a definite no though!!

[happygooner]

Appears to be nothing on tonight for the 2nd system.

[happygooner]

Originally Posted by happygooner

Atlanta +12.5 @ 1.9 for 1.10 units
Detroit ML @ 2.06 for 1.10 units

I think this is correct...

+.07

Todays start bank 101.05

[happygooner]

TAR System:

Utah -7.5 @ 2.12 for 1.10 units

[happygooner]

System 2: Mathematicians Formula:

Utah -7.5 @ 2.12 for 1 unit
Sacramento ML @ 1.92 for 1 unit

[happygooner]

Originally Posted by happygooner

TAR System:

Utah -7.5 @ 2.12 for 1.10 units

Well yesterdays didn't go too well did it....

TAR bank 99.95 units

[happygooner]

Originally Posted by happygooner

System 2: Mathematicians Formula:

Utah -7.5 @ 2.12 for 1 unit
Sacramento ML @ 1.92 for 1 unit

Rubbish start

0-2 so bank down to 98.00 units.

[happygooner]

TAR System

Clippers -2.5 @ 2.10 for 1.10 units

Mathematicians Formula

Charlotte +7.5 @ 1.90 for 1 unit

[happygooner]

Originally Posted by happygooner

TAR System

Clippers -2.5 @ 2.10 for 1.10 units

Mathematicians Formula

Charlotte +7.5 @ 1.90 for 1 unit

Another pish night

TAR 98.95 units

Mathematicians 97 units

Nothing for the Mathematicians tonight but maybe one for the TAR.