How many Games per day shall I bet to minimize VARIANCE

[coolguy73739]

Hey Guys,
Looks like finally I found a method to bet NCAAB totals and win consistently.
I have been trying this for last 1 week and I have been winning consistently with an average ROI of 15%.
Some day its 5-4, sometimes 6-4, 5-5, 7-3 etc.
What I like to know is how many games shall I play everyday to minimize the variance on this one.
With so many games and my edge on a few of them, I am underwhelmed/overwhelmed at times and need guidance.
For the time being I am planning on to keep about 10 games per day.

Thoughts and valuable suggestions from SHARP minds here?

[Nappyx]

My scientific mind thinks you should test this this method for more than 1 week before you mention anything about consistency....

[TommieGunshot]

Betting every game will have less variance than any other number of games bet. In college basketball, if someone were to bet every game for the year, the actual results will come very close to expected results.

[coolguy73739]

I am interested to know how many games you wanna play each day before Variance bites you.
Also whether MORE GAMES PLAYED, LESS VARIANCE FACED holds any value.

[PaperTrail07]

If you believe your system has an edge-you want to play as many games that fits right? Without knowing the method...cant give more info...

Hey Guys,
Looks like finally I found a method to bet NCAAB totals and win consistently.
I have been trying this for last 1 week and I have been winning consistently with an average ROI of 15%.
Some day its 5-4, sometimes 6-4, 5-5, 7-3 etc.
What I like to know is how many games shall I play everyday to minimize the variance on this one.
With so many games and my edge on a few of them, I am underwhelmed/overwhelmed at times and need guidance.
For the time being I am planning on to keep about 10 games per day.

Thoughts and valuable suggestions from SHARP minds here?

[BeatTheJerk]

My scientific mind thinks you should test this this method for more than 1 week before you mention anything about consistency....

Word.

[semibluff]

Some basic numbers churned out through a free online binomial calculator:

You hit coin flips at 55%.
Your probability of being ahead after 11 flips is 63.31%
Your probability of being ahead after 31 flips is 71.32%
Your probability of being ahead after 101 flips is 84.38%
The answer to your basic question is: Yes, you should bet more games.

If you bet just your biggest advantage line of the week you will have the greatest chance of individual success. As you bet more games you will inevitably be betting games with a smaller advantage on your side. The more interesting question is at what size advantage will you stop betting?

You need to keep in mind that your objective is to win money, not to beat or avoid variance. Variance can't be avoided and eventually it will destroy you if you employ any standard flat-betting strategy where you are staking enough for it to be worth the time and effort should you win. This isn't an attempt to re-open a staking strategy debate with Kelly Criteria posters. It's merely a reminder that profit is the goal and that there is no safe way to reach that goal with a gambling advantage. As a lesson to all let me mention my disastrous NFL Pick'em campaign this year. After 150 games I was 58-92. Statistically it's almost impossible to be that bad. If I were to compete with 999 randomly generated picking bots I would statistically be in the bottom 3 at that point. Worse still these were not my picks! The picks were a strategy of picking whichever side was deemed to be +ve as per the betting line as of the Thursday Night kickoff. By definition every pick was either + market value or at worst neutral. Variance is unavoidable, even with a system that significantly beats the closing line.

[coolguy73739]

Thank you so much for your valuable input.

[coolguy73739]

If you believe your system has an edge-you want to play as many games that fits right? Without knowing the method...cant give more info...

Formula is simple. My system gives me number of points with probable low and high pace game for basketball. Let's say those numbers are 150 for high pace and 135 for low pace for college Basketball. If line falls between these two numbers, I skip. If line is below 135 or above 150, I play. It's working and like I said giving me ROI of about 15 average. Could be more if I could skip those games I'm losing but not sure how to skip and hence the question of variance.. Thank you..

[Bizanta]

Formula is simple. My system gives me number of points with probable low and high pace game for basketball. Let's say those numbers are 150 for high pace and 135 for low pace for college Basketball. If line falls between these two numbers, I skip. If line is below 135 or above 150, I play. It's working and like I said giving me ROI of about 15 average. Could be more if I could skip those games I'm losing but not sure how to skip and hence the question of variance.. Thank you..

If you have been testing this formula for a week now, you can't really say you have an edge to be honest.
And yes, 15% ROI is fantastic, no doubt about that.

There is something that's called the p-value, which gives you the chances of your record occuring 1-in-X time when picking random.

In your case I assumed the following: 70 picks, average -110 (and none very far off), 15% ROI.
According to the P-value calculator there is a 10% chance that your record would have occured by picking random, which means that if you had 10 people picking 70 random bets, you expect 1 of them to end up just like you. You can Google for 'Survivorship bias' and find some interesting reading material.

You might be onto something as the start is promising but in my opinion it's too early to get too confident. A 15% ROI is really hard to sustain. To show the importance of the ROI in the p-value, I would like to change 15% to 7.5%. Suddenly there would be a 25% chance for this record to occur.

It's up to you what you consider to be statistically significant in order for you to bet with confidence and the risk that you are willing to accept.

You can look for p-value and play with the numbers to see how different numbers of bets, ROI and average odds will affect things.
To end things, in my opinion there is more to a pick just winning or losing. In theory you can go 6-4 on a day but the winners all being close while losers being blowouts. This specific example is quite extreme but I believe that is also something you should take into consideration.

[coolguy73739]

Thanks Bizanta for all your input. Its tough to measure chances to pick the game based my data using p-value.
Classic example is Dayton-Duquesne game tomorrow. Based on my system, total is 142.6 and 145.6 based on low/high pace. Line is set at 135.0. This is the kind of game I am focusing on. Line 135.0 is a BLUFF or REAL is what I usually look for. I will be digging deeper to ascertain how good DEFENCE will work for this game based on Teams' and Opponents' effective Possession ratio and other many defense stats. Finally I have to decide whether I play Over or Under for this game. Thoughts?

[Bizanta]

The problem is that even the most advanced analysis eventually comes down to winning long term. We can simply never now for sure what the exact chances are for a specific outcomes, so the best meassurement of success is a large sample size to make it statistically significant and proves your edge.

Also, let's say that Dayton - Duquesne will hit 152 points. In that case people tend to credit themselves for a great analysis: 'Easy high paced game', even if Dayton was shooting 7 from 12 from deep which is way above their average. Meanwhile, if it hits just 126 points they tend to lean towards 'unlucky', 'Dayton was just shooting 3 from 14 from outside the arc' and 'How did Duquesne just made 50% of their 18 FT's?'. That is why I give most credit to a long term record with a high statistical significance and care less about analysis on a given game.

Sometimes I see guys using the following thing as an angle: 'The Cavaliers are 8-2 against the spread on the road after two or more days rest versus teams with a losing record' I'm sure they are but it's just too random for me. Not bashing anyone here but I simply refuse to use those statistics. Also, if teams covering all the time, bookmakers might also adjust a little bit which will also decrease the value on that team if there is any to begin with.

I mean, your system might be great. Don't get me wrong. For me personally, the ultimate truth lies in the numbers.
Keep doing what you are doing and I am very curious how your record will look after the next couple of weeks.

[coolguy73739]

I have decided to play OVER 135 for Dayton Duquesne Game. Given lot of weightage to Opponents Total shooting %, 111.5% for Duquesne and 106.2% for Dayton to be precise. Verdict is out.. 135 is FAKE.. Lets see what happens.. Anything is possible with Variance Bitch..

[stevenash]

Betting every game will have less variance than any other number of games bet. In college basketball, if someone were to bet every game for the year, the actual results will come very close to expected results.

Couldn't have said it better myself.

[HedgeHog]

I am interested to know how many games you wanna play each day before Variance bites you.
Also whether MORE GAMES PLAYED, LESS VARIANCE FACED holds any value.

0 games in your case, as variance will then not be a problem for you. If you had an actual edge, then you shouldn't be concerned about how many plays you make. Bet ALL games that you have a legit advantage, and pass ALL games you don't. It's that simple.

[coolguy73739]

0 games in your case, as variance will then not be a problem for you. If you had an actual edge, then you shouldn't be concerned about how many plays you make. Bet ALL games that you have a legit advantage, and pass ALL games you don't. It's that simple.

Thank you

[d2bets]

Per day is not really a useful way to look at it. The answer is you should make every possible +EV bet that you can (given time and money constraints). No more, no less. No reason to try to get a certain amount per day. One day it might be 2 and then next it might be 20. Doesn't really matter, but the more the merrier.

[coolguy73739]

Per day is not really a useful way to look at it. The answer is you should make every possible +EV bet that you can (given time and money constraints). No more, no less. No reason to try to get a certain amount per day. One day it might be 2 and then next it might be 20. Doesn't really matter, but the more the merrier.

Thank you so much for your insight d2bets. Been trying to figure out defense data for teams for a while. In case of CBB, if line for totals is 135 when your system churns out 150 for same game, how would you analyze such game to decide UNder/Over for this Game? I have been focused on many data and Opponents True shooting percentage has been a bit helpful. Went 3-0 today for Under based on this important stat, of course have to try it out for some days before I can be sure of any data. Thoughts? Thanks..

[d2bets]

Thank you so much for your insight d2bets. Been trying to figure out defense data for teams for a while. In case of CBB, if line for totals is 135 when your system churns out 150 for same game, how would you analyze such game to decide UNder/Over for this Game? I have been focused on many data and Opponents True shooting percentage has been a bit helpful. Went 3-0 today for Under based on this important stat, of course have to try it out for some days before I can be sure of any data. Thoughts? Thanks..

Uhhh, well if your model is actually correct then bet that hard. But I reckon to say that closing totals are never truly wrong by 15 points. If that's what your model is churning, then you need to take a hard look at that model.

My focus is on beating the closing/prevailing line. Not to say that there aren't other ways, but that's what I know is reliable and tried and true. If you get CLV over the long-run, you will win guaranteed.

[coolguy73739]

Thanks. Somehow never got convinced that CLV can be a true measure of your talent / success.

[Waterstpub87]

Thanks. Somehow never got convinced that CLV can be a true measure of your talent / success.

I always tell people, run the data. You'll be amazed how much money you would win if you always beat the closing line.

[GunShard]

My sharp experience sports betting since 2009. I seen other sports bettors attempt a variance system and their systems failed. The variance is worst in NBA and NHL. The sports that's easy for me to win are on the NFL and UFC favorites. The public loses on betting "short term" regular season especially NFL night game favorites. The public wins on betting "long term" sports futures. Money management is important, you should be flat betting each game, $100 on game 1 then $100 on game 2 etc.

Here's a futures bet that I won. I believe everyone on here can win sports futures bets if you study each team in detail.

[d2bets]

Thanks. Somehow never got convinced that CLV can be a true measure of your talent / success.

I could care less about "talent". More interested in success. You don't need to be convinced; it's a provable fact. Again, not to say it's the only way, but it's the surest way.

[d2bets]

I always tell people, run the data. You'll be amazed how much money you would win if you always beat the closing line.

If you need a way to think of it without all data, imagine you bet 20 hoops games ATS or totals in a day. If you've done this before, you know that you are going to be within a point of the number once, maybe even more. Let's say your best number helps you win 1 more out of 20. Well, just with coin flips that turns 10-10 into a profitable 11-9. Now keep doing that over and over and over. You just won.

[coolguy73739]

What drags the closing line to where it closes? Public Money based on their Perception? Sharp Money? Why Line moves for some games and not all? I always believe for certain games, Books purposely take line where they wanna take irrespective of Market forces.
You bet a game at OVER 152.5, Line closes at 155, are you a winner already? A lot happens after game starts, Live Line will move probably from 165 to all the way down to 140 may be.. How BTCL declares you a WINNER?? I never understood..

[d2bets]

What drags the closing line to where it closes? Public Money based on their Perception? Sharp Money? Why Line moves for some games and not all? I always believe for certain games, Books purposely take line where they wanna take irrespective of Market forces.
You bet a game at OVER 152.5, Line closes at 155, are you a winner already? A lot happens after game starts, Live Line will move probably from 165 to all the way down to 140 may be.. How BTCL declares you a WINNER?? I never understood..

No, but over time on those you have much better than a 52.4% chance to win. If that wasn't the case then you could just gobble the tail end of the move and bet the other side and be profitable.

Of course on any individual game, it might end up either way, but long-run results on this don't lie.

[coolguy73739]

What's the LONG RUN? 1000 GAMES? 2000 GAMES? I will be broke and on street by the time I hit that Milestone.. No wonders no one wins at this Crap..
I have nothing against your strong belief in CLV but I'm always sceptical about it.

[d2bets]

What's the LONG RUN? 1000 GAMES? 2000 GAMES? I will be broke and on street by the time I hit that Milestone.. No wonders no one wins at this Crap..
I have nothing against your strong belief in CLV but I'm always sceptical about it.

wtf man. There can be variance just like in any other system, but it's the most surefire way to win.

"No wonders no win wins at this Crap" is woefully inaccurate. Just tells me you're not winning. Maybe if you focused on CLV.

My results are proof. I say CLV but I massage that to say prevailing line, because my current favorite, given my stable of outs, is CBB halftime betting. I'm able to beat the prevailing line handily, so I know I am going to win.

[semibluff]

If you are winning you will almost certainly be beating the closing line. However, if you are beating the closing line it does NOT almost certainly mean that you will be winning. Focus on winning. In practice that will mean betting -110; -110 book lines where your evaluation says the true odds line should be -115; +115.

[d2bets]

If you are winning you will almost certainly be beating the closing line. However, if you are beating the closing line it does NOT almost certainly mean that you will be winning. Focus on winning. In practice that will mean betting -110; -110 book lines where your evaluation says the true odds line should be -115; +115.

It certainly does it you beat it by enough.

If you bet O132 and it closes 132.5, meh. If you bet O132 and it closes 135, you are going to win (long-term).

[dante1]

some very decent explanations in this thread. allow me to add my two cents concerning the number of games one might play.

I booked for more than two decades and I just loved when a player would give me 15 or 20 pieces. I could almost guarantee he will be a loser at the end of the day. it is simple math. let's say a player tosses in 20 games on a college Saturday at the usual 110. do the math, check the possibilities and the W/L chances. just for the hell of it let's say 110/100. now compare a 8-12 result vs the exact mirror opposite of 12-8. this should give you an example of why betting a ton of games is never a good strategy.

one more comment. before you buy into any system at all make sure you have a huge amount of data. thinking a system is working wonderfully because in week one it went 59 and 41 is almost worthless data.

bottom line (and many will disagree with me) winning in the long run, meaning a life time of betting is very very rare.

[d2bets]

some very decent explanations in this thread. allow me to add my two cents concerning the number of games one might play.

I booked for more than two decades and I just loved when a player would give me 15 or 20 pieces. I could almost guarantee he will be a loser at the end of the day. it is simple math. let's say a player tosses in 20 games on a college Saturday at the usual 110. do the math, check the possibilities and the W/L chances. just for the hell of it let's say 110/100. now compare a 8-12 result vs the exact mirror opposite of 12-8. this should give you an example of why betting a ton of games is never a good strategy.

one more comment. before you buy into any system at all make sure you have a huge amount of data. thinking a system is working wonderfully because in week one it went 59 and 41 is almost worthless data.

bottom line (and many will disagree with me) winning in the long run, meaning a life time of betting is very very rare.

It is when you have an edge, just like how it's good for the book when you don't have an edge. Maybe you were a good bookmaker. There are plenty of shit bookamkers out there for winners to feast upon. The more I bet, the more I win. I like to bet as many different things as possible. When I make only 5 or 6 bets in a day I have a better chance to lose than when I bet 50 or 60.

[dante1]

It is when you have an edge, just like how it's good for the book when you don't have an edge. Maybe you were a good bookmaker. There are plenty of shit bookamkers out there for winners to feast upon. The more I bet, the more I win. I like to bet as many different things as possible. When I make only 5 or 6 bets in a day I have a better chance to lose than when I bet 50 or 60.

hey D2 you know one thing I am positive concerning this reply. it is honest you have a reputation of integrity and are credible. my friend maybe you are one of the lucky 1% that might win in the long run. in fact, I believe the actual percentage is lower than 1%.

of all my players in all those years I.cannot say with any accuracy that even one was a long term winner. the math is against you however if you (and maybe you have) found a method I am happy for you, legitimately happy. I personally have bought into a ton of theories/methods and some of them worked until they didn't. probably the one I had the most success with was line changes. that worked for some time but in the end it failed.

continued good luck to you my friend. I wish you the best of luck and the best of health.

[d2bets]

hey D2 you know one thing I am positive concerning this reply. it is honest you have a reputation of integrity and are credible. my friend maybe you are one of the lucky 1% that might win in the long run. in fact, I believe the actual percentage is lower than 1%.

of all my players in all those years I.cannot say with any accuracy that even one was a long term winner. the math is against you however if you (and maybe you have) found a method I am happy for you, legitimately happy. I personally have bought into a ton of theories/methods and some of them worked until they didn't. probably the one I had the most success with was line changes. that worked for some time but in the end it failed.

continued good luck to you my friend. I wish you the best of luck and the best of health.

Has zero to do with luck. All about knowledge, discipline and access. I guess if there's a luck aspect it's having access to useful books (and sufficient cash).

[dante1]

Has zero to do with luck. All about knowledge, discipline and access. I guess if there's a luck aspect it's having access to useful books (and sufficient cash).

well my friend if you can win wagering 50 to 60 plays I have some good counsel. start selling those games.

let's look at the math if you play 60 games. just one example. let's say you hit 55% of those games on a particular day. if my math is correct at $100/110 per piece and a 55% W rate you would profit $330. however the same mirror image of 55% as a loser would produce a loss of $930.



I just finished cleaning up the kitchen so I only spent a few minutes on these calculations. maybe I am mistaken, if so you can help me out. but if those numbers are correct playing 60 games too often will find you in the shit hole very soon. that is unless you can select at 57% or higher daily. and if so sell those picks. or a famous line from BONNY AND CLYDE--DONT SELL THAT COW.

now the moral of this story is a 55% win will give you a small profit--but a 55% loss gives you almost three time greater loss.