I just discovered a reliable system for betting NBA games

[suicidekings]

Betting sports isn't like playing the lottery, where it truly is random balls spinning around a container. Sports betting is predictable, it's just no one has figured it out yet. Until now?

You're off course here. Certain trends / tendencies are predictable and it is often possible to identify situations where there is a definite edge on one side of a line / spread / total. It all looks great on paper, but on the court, random things happen all the time.

You shouldn't be trying to have your system pick winners, but rather identify and quantify the situations where one side has the advantage. That means picking up totals and spreads that are off from what they should be.

And you're doing this by hand calcs? Excel is a handicapper's best friend. Learn to use it.

[RoagBettor]

Next time I speak with Stephen Hawking I'll run this by him. There has to be something in Quantum Mechanics that can help us beat the books!

[Jerm3462]

Orlando -3.5 (WIN)
Denver -2.5 (LOSS)
Memphis +6.5 (LOSS)
Sacramento +2 (WIN)
Miami -8 (WIN)
Houston +6.5 (LOSS)

3-3

sigh

I'll mess with it tomorrow..I dont want to quit.

[kahnner]

You might be on to something!
Keep testing it and see how it does!
Anything over 60% is great!

[n1co35]

will be on a look out again 2morow. like they said anything 60% will be great.

[texhooper]

backtest the mother****er already. i say that in the nicest way possible. it's the best way to know if you're wasting your time going forward or not. of course, no one knows what your formula is, but if what you need is more than basic past results someone here probably has what you're looking for.

i agree with everyone else in that it's very commendable what you're doing, and it's awesome you would share these results so truthfully, but this forward-testing approach is not worth it if it's destined to lose. you have shared no past results or anything to that effect, so i only assume you haven't backtested this.

[johncrud]

systems

[Jerm3462]

I'm backtesting the last 3 nights. I don't have time or the excel skills to sit here and go thru previous years.

I studied last nights games and my "system", and tried to figure out a variable to add or subtract to make the losses turn into wins, and same for previous night; without making the wins it selected, losses....a tough task.

So I woke up early, tinkered with this, tinkered with that....looked at stats...added more weight to this...less weight to that.......received fellatio, and BAM...the system proves TRUE on the last 21 NBA games ATS

I don't know how much more manipulating I can do to the system to keep all games TRUE so I think I might just have to settle on this system I have now.

For any "system" to pick 21 games ATS is a phenominal feat, even if I changed it to do so.

A lot of games tonight, this will be a real good test. I'll have em up around 6pm EST or so.

Thanks for the support.

[MrDomination]

Good luck finding a system that picks almost every game right, there are humans playing this game, and as we know, humans tend to be predictable in some cases, but in others, not so much..

I can think of a system that would be perfect.. Only play teams in the top half of the standings.. then always buy 40 points.. It would problably have an amazing record, right? :P Still bet you can find some losses over the last few years... No system is perfect, but some are profitable, and that's good enough.

[jgiun1]

Damn, even a system that loses at 50% is better than some people that actually know the sport.........LOL

[jgiun1]

Looking forward to your card tonight with the huge slate of games

[mojomaker11]

good luck tonight.

[Jerm3462]

Wednesday Nov 11,2009 NBA selections

Here they are:

Toronto -5
Indiana -5
Boston -11
Detroit -4
Philadelphia -4.5
Atlanta-5
Portland -7.5
Orlando -1
Milwaukee +4
Houston -9
LA Clippers -1.5
Phoenix -6.5





Seems to lean on the favorites tonight, but it did spit out Milwaukee +4 if your a dog kinda guy.

Let's see how it goes. Again, this one is backtested on 21 games (3 days worth)

P.S The San Antonio/Dallas matchup is a NO BET game.

[kahnner]

HOPEFULLY GOES WELL!
I LIKE A LOT OF FAVORITES TONIGHT ALSO!

[mojomaker11]

like them all but atlanta.

[Jerm3462]

like them all but atlanta.

I wish I could discuss this with you, but I don't know jack about basketball.
My system picked it.
I did learn one thing watching my first NBA game in about a dozen years last night.
That guy Wade on Miami is really good.

[williams22]

You need to either backtest this for at least 3 seasons, or just quit wasting your time now. Taking it day by day and then changing the formula each day to turn losers into winners will get you no where. Set up a simulation in Excel.

[Waiting4Godot]

If you aren't Excel savvy, send a PM to one of the folks in this thread who's recommending Excel and he/she will likely help get the formulas setup.

[Sekrah]

Guys.. He's jerking all of you around. None of this is serious.

[Jerm3462]

Toronto -5 (WIN)
Indiana -5 (WIN)
Boston -11 (WIN)
Detroit -4 (WIN)
Atlanta-5 (WIN)
Portland -7.5 (WIN)
Milwaukee +4 (WIN) nice dog pick . Won S/U
Houston -9 (WIN)
Philadelphia -4.5 (LOSS)
Orlando -1 (LOSS)
LA Clippers -1.5
Phoenix -6.5

[mojomaker11]

i don't know, looks like at least 6 of the first 8 are gonna cash.

[kahnner]

Alot of these look like winners so far!
Like i said before anything over 60% is awesome!
Good work so far!
I will be watching how this goes!

[Pnutz84]

Not looking too bad at all, Wish I seen this sooner.

[suicidekings]

I'm backtesting the last 3 nights. I don't have time or the excel skills to sit here and go thru previous years.

I studied last nights games and my "system", and tried to figure out a variable to add or subtract to make the losses turn into wins, and same for previous night; without making the wins it selected, losses....a tough task.

So I woke up early, tinkered with this, tinkered with that....looked at stats...added more weight to this...less weight to that.......received fellatio, and BAM...the system proves TRUE on the last 21 NBA games ATS

I don't know how much more manipulating I can do to the system to keep all games TRUE so I think I might just have to settle on this system I have now.

For any "system" to pick 21 games ATS is a phenominal feat, even if I changed it to do so.

A lot of games tonight, this will be a real good test. I'll have em up around 6pm EST or so.

Thanks for the support.

This right here is also a problem. I'm assuming you're using current season data (correct me if I'm wrong), which to this point is a very small sample size, with each team playing 7-8 games. Taking a system based on those stats and modifying it such that the results correlate better with the actual results is only fitting the model to the data, not improving it's accuracy in predicting future events.

[suicidekings]

That being said, good work on the early games tonight.

[Jerm3462]

fitting the model to the data, not improving it's accuracy in predicting future events.

Hmm, yea that's what I did. But I'm not sure how it doesn't better the accuracy of future games. Are you saying what teams did the past 3 nights mean zero?

[netinfo]

Congrats on a 9-3 night

netinfo

[Jerm3462]

Toronto -5 (WIN)
Indiana -5 (WIN)
Boston -11 (WIN)
Detroit -4 (WIN)
Atlanta-5 (WIN)
Portland -7.5 (WIN)
Milwaukee +4 (WIN)
Houston -9 (WIN)
Phoenix -6.5 (WIN)
Philadelphia -4.5 (LOSS)
Orlando -1 (LOSS)
LA Clippers -1.5 (LOSS)
9 out of 12..not bad...not bad 75%..I would have made good $ if I straight bet each.
I'll run this system on Thursdays games when I get home from college. Peace all

[jgiun1]

I like to see another..good work bro

[TGHMak]

Tonight most of the favorites covered so it is somewhat deviated.

[suicidekings]

Hmm, yea that's what I did. But I'm not sure how it doesn't better the accuracy of future games. Are you saying what teams did the past 3 nights mean zero?

No, but you're assuming that the small sample is actually representative of how the teams will play in the future. An analogy would be cooking for a small group a few times, adjusting the menu to suit their individual tastes until you have recipes that the group all love. You try those same recipes out on a larger population and you're unlikely to match that original approval rating.

If you're unwilling to collect and process past season data, your model can only go so far. When you're only looking at 3 days worth of games, you're not sure if a team won because they played better or because the other team played below their skill level. The variability in performance of players and teams over different periods of time is huge. Maybe Wade scores 45 points against Cleveland but Miami loses the game, and in the next one he only puts up 20 points against Utah, but the Heat win easily. Which game is more representative of what will happen in the next game? Or look at Denver, going 5-0 to start the season and becoming public betting darlings, only to falter when they started facing stiffer competition. Is Denver the 5-0 team or the 1-3 team? Which set of games represents them better?

Add to that: injuries, homestands and roadtrips, bad reffing, teams' playing styles and how they match up, etc... There's so much variability in game outcomes, you need that larger pool of data to smooth out the wrinkles and develop a model that is more representative of a team's actual ability rather than the team's W-L record.

[suicidekings]

If you don't mind me asking, what stats are you using to make your selections? Points per game for and against? What else?

[n1co35]

good stuff, i guess for now u found the right way to change ur system plays.
hope u keep this up.

[mojomaker11]

Good job, system or no system the results are great!

[TANGIBLE]

Any plays for today?