[gojetsgomoxies]

something i'm trying to get my head around...

i've come up with or seen some big winning systems that are pretty reasonable as to volume of picks and not excessive amounts data mining

i have two basic questions,

so it's worked brilliantly for a long time but lo and behold either 2015 or 2016 was a big loser........ what's your thought process?

say again it's worked really well for a 6 or 7 year period........ can it be expected to work really well for a 12 year period? bettors catch on and/or the underlying dynamics change - second effect is maybe what i'm trying to get my head around.

an example would be non-big 5 conferences and UNDER in bowl games........... worked really really well.. i understand people catching on but what about just basic change? for one thing, the CUSA today is NOT the CUSA 5 years ago... credit to JTrain for bringing ths to my attention

thx in advance ............ sorry for the ramble

[gojetsgomoxies]

i will try to look up the odds of a 65% system going 1 and 5 or 0 and 6..... i think 0 and 6 is about 1 chance in 700 - did this in my head 1/729 - but i don't think that's the real world as alot of these results in your 65% win system were close so maybe it's not a 65% win system.

[Waterstpub87]

something i'm trying to get my head around...

i've come up with or seen some big winning systems that are pretty reasonable as to volume of picks and not excessive amounts data mining

i have two basic questions,

so it's worked brilliantly for a long time but lo and behold either 2015 or 2016 was a big loser........ what's your thought process?

say again it's worked really well for a 6 or 7 year period........ can it be expected to work really well for a 12 year period? bettors catch on and/or the underlying dynamics change - second effect is maybe what i'm trying to get my head around.

an example would be non-big 5 conferences and UNDER in bowl games........... worked really really well.. i understand people catching on but what about just basic change? for one thing, the CUSA today is NOT the CUSA 5 years ago... credit to JTrain for bringing ths to my attention

thx in advance ............ sorry for the ramble

If your system is beating the close, then you are still winning. If it is not, it is no longer a good system. If you won without beating the close, you were lucky.

The odds of going 0-6 with a 65% winning system is ~.002, or 1 out of 500, it is like .00184 or something. This seems very small, but consider that you bet 5 games a day for 365 days of the year, for a total number of 1825 bets. You would then have ~1815 6 bet series. so you would expect to go 0-6 3.65 times or so in a year.

It probably isn't a 65% system, I don't know if that is possible long term. Unless you are betting -150 chalk or more.

[gojetsgomoxies]

here's an example........... not sure it's system or what....

since and including 2008 bowl season... MAC is 14-30 ATS... so low 30% ATS........ no cherry picking of dates, teams are i think almost exactly the same for whole period (mass, temple and maybe SF moved?)

so here's the record

2008 0-5 0%
2009 0-4 0%
2010 2-2 50%
2011 3-2 60%
2012 2-5 29%
2013 0-5 0%
2014 3-2 60%
2015 4-3 57%

2016 0-2 0% but small number of games so far.

it's hard to think there isn't something strange going on with those numbers.....seems like excess kurtosis by season i.e. the seasons seem like too many outliers vs. the 33% average or similar.

[gojetsgomoxies]

sorry, i mixed ideas earlier...... a 65% ATS system would be based on bowl games which are small in number so small model but grades high % of bowl games.

high volume winning ats models i'd tend to think of in the 53% to 57% range........ and certainly at least moderate data mining. but 1% of less of the data mining that marc lawrence does and the models would make intuitive sense i.e. based on behavioural factors.

[Waterstpub87]

here's an example........... not sure it's system or what....

since and including 2008 bowl season... MAC is 14-30 ATS... so low 30% ATS........ no cherry picking of dates, teams are i think almost exactly the same for whole period (mass, temple and maybe SF moved?)

so here's the record

2008 0-5 0%
2009 0-4 0%
2010 2-2 50%
2011 3-2 60%
2012 2-5 29%
2013 0-5 0%
2014 3-2 60%
2015 4-3 57%

2016 0-2 0% but small number of games so far.

it's hard to think there isn't something strange going on with those numbers.....seems like excess kurtosis by season i.e. the seasons seem like too many outliers vs. the 33% average or similar.

It is easy to think there isn't something strange going on with those numbers. You can't really base anything off such a small sample size. Totally random, just looks like a pattern. Additionally, ATS performance is not a good indicator of anything. It is often dependent on where you got the line, are these lines closer or openers?

So, if I understand you, your system is to just fade the MAC schools in bowl games? Does that intuitively make sense to you, that to win you simply need to bet against a certain conference?

[tsty]

systems dont win lol

just luck

[gojetsgomoxies]

thanks for the feedback. much appreciated

i would say FADE MAC teams in bowls is fine for a system. you may want to do some qualitatitve tweaking, which of course is impossible to back test.......... AND i would NOT expect it to last too long. either bettors adjust or MAC gets better and people don't adjust enough.

it's funny i posted this as i am reading Michael Lewis' book about Daniel Kahneman/Behavioural econonics....... t

he next page he started on how people think small sample result converge to the population average far more quickly than they do...... i.e. people think after 20 flips of a coin that it'll very close to 50%/50% but it often isn't.

i found this out myself recently doing a challenge that the economist magazine put forth. even 40 flips of a coin i wasn't too confident that it was 50% chance instead of 60% chance - i.e. was it a loaded coin?

lastly, the one poster makes a good point. ATS or totals results are not really like flipping a coin. you know when you flip a coin whether it's heads or tails. whether X team covered depends at least 5% to 10% of the time on what line you got i.e. open vs. close. as the one poster mentioned.

i guess my original question was whether sports betting is a bit more susceptible to runs than a coin flip assuming both have 50% chance. game results somehow with small positive correlation?

[Waterstpub87]

thanks for the feedback. much appreciated

i would say FADE MAC teams in bowls is fine for a system. you may want to do some qualitatitve tweaking, which of course is impossible to back test.......... AND i would NOT expect it to last too long. either bettors adjust or MAC gets better and people don't adjust enough.

it's funny i posted this as i am reading Michael Lewis' book about Daniel Kahneman/Behavioural econonics....... t

he next page he started on how people think small sample result converge to the population average far more quickly than they do...... i.e. people think after 20 flips of a coin that it'll very close to 50%/50% but it often isn't.

i found this out myself recently doing a challenge that the economist magazine put forth. even 40 flips of a coin i wasn't too confident that it was 50% chance instead of 60% chance - i.e. was it a loaded coin?

lastly, the one poster makes a good point. ATS or totals results are not really like flipping a coin. you know when you flip a coin whether it's heads or tails. whether X team covered depends at least 5% to 10% of the time on what line you got i.e. open vs. close. as the one poster mentioned.

i guess my original question was whether sports betting is a bit more susceptible to runs than a coin flip assuming both have 50% chance. game results somehow with small positive correlation?

Bettors/Market does not adjust, the system never worked in the first place is the more likely explanation. I doubt there is any correlation between independent, unrelated bets, or the performance of different teams.

[the shadow]

Always base a system on a minimum of 5000 games . You may have to wait 10 years to see if it successful! Same thing with handicapping.

[Buffalo Nickle]

Simple ideas like bet under in the bowl games or bet against the MAC are not "systems." Those are most likely aberrations in statistics. Just noise.

Games are being bet with extreme statistical measures adjusting yards per quality of opponent. There is not much gets past the computer guys so the lines are very hard to beat. There is not going to be a simple bet like bet against Conference USA because the computer guys know pretty much exactly how good those teams are.

You've got to have an angle that is better than their numbers or you've got to be better at crunching numbers. Betting these days is largely a battle of the computer geeks.

If you want to win at gambling, you've got to be able to pick winners in head-to-head matchups. There is not going to be an easy way around that.

[QuantumLeap]

The volatility of any betting system, whether it be sports betting or casino games or card games is different for each system as is the success of each system. Each system has its own statistical probability of winning along with its own volatility of the system.

I do believe there can be systems that are successful but I think the one you mention can be validated as a true system due to the small sample size as was mentioned before.

[tsty]

The volatility of any betting system, whether it be sports betting or casino games or card games is different for each system as is the success of each system. Each system has its own statistical probability of winning along with its own volatility of the system.

I do believe there can be systems that are successful but I think the one you mention can be validated as a true system due to the small sample size as was mentioned before.

lol

[StackinGreen]

You've got to have an angle that is better than their numbers or you've got to be better at crunching numbers. Betting these days is largely a battle of the computer geeks.

It distills down to this. Nickle is right on.

There is a LOT of work for real long term winning. All in all you have to choose one of the following:

1. High edge low volume
2. Low edge high volume

Guess which most "winners" end up utilizing? I'll explain more when you answer.

The biggest eye opening thing for me, and recently, is that market factors and angles (why do they offer that bet?) are the only way to hone in on where you can find the opportunities. Then, once you find where it is even possible, there are other factors that start to get in your way.

Hint, hint: it's where the "system" is lazy. But even then you have to exploit it and understand why it happens to be lazy.

I know I'm sorta speaking in code but those who know know what I'm talking about.

Back to 1 and 2: How many guys do you know have enough knowledge and enough humility (read: no ego) to put in all that time and only bet a few games at high edge? Not most competitive sports guys. So I just gave you the answer.

Walters didn't have any more knowledge in any single game than anyone else. Think about that for a long time. He had knowledge elsewhere.

[evo34]

systems dont win lol

just luck

That's exactly what a skill-less person would say.

[evo34]

It distills down to this. Nickle is right on.

There is a LOT of work for real long term winning. All in all you have to choose one of the following:

1. High edge low volume
2. Low edge high volume

Guess which most "winners" end up utilizing? I'll explain more when you answer.

The biggest eye opening thing for me, and recently, is that market factors and angles (why do they offer that bet?) are the only way to hone in on where you can find the opportunities. Then, once you find where it is even possible, there are other factors that start to get in your way.

Hint, hint: it's where the "system" is lazy. But even then you have to exploit it and understand why it happens to be lazy.

I know I'm sorta speaking in code but those who know know what I'm talking about.

Back to 1 and 2: How many guys do you know have enough knowledge and enough humility (read: no ego) to put in all that time and only bet a few games at high edge? Not most competitive sports guys. So I just gave you the answer.

Walters didn't have any more knowledge in any single game than anyone else. Think about that for a long time. He had knowledge elsewhere.

Don't confuse quantity of work with quality of work. Most successful gamblers (as few as they are) have never been on 60 Minutes, do not use Twitter, and do their research in an efficient way -- not via 80-hour weeks.

[StackinGreen]

What % would you guess are "successful" evo? Define success any way you want, then please estimate it, I'd like to know your thoughts.

[tsty]

That's exactly what a skill-less person would say.

No it's not. It's what people who bet more than $100 on a game say.

It actually amazes me that you are not a troll and think you are winning.

Enjoy positive variance.

[StackinGreen]

No it's not. It's what people who bet more than $100 on a game say.

It actually amazes me that you are not a troll and think you are winning.

Enjoy positive variance.

What has you so convinced that you are a loser? If you are convinced, why do you bet? Why don't you try another angle or quit?