Week 7 begins to tell the tale of my system... not sure it will work but here it is..

**usma1992** · 10-11-23, 07:11 PM

Follow On...

12:00 PM	ARKANSAS	8	ALABAMA	27	35	-19.5	48	2Close	Under
12:00 PM	TEMPLE	39	NORTHTEXAS	46	85	-7	68.5	2Close	2Close
12:00 PM	GEORGIA	55	VANDERBILT	0	55	31.5	57.5	2Close	2Close
2:00 PM	NAVY	33	CHARLOTTE	15	48	3	45	Away -	2Close
2:00 PM	TOLEDO	46	BALLST	6	52	16.5	49.5	Away -	2Close
3:00 PM	CALIFORNIA	26	UTAH	24	50	-13	44.5	Away +	2Close
3:30 PM	ILLINOIS	17	MARYLAND	41	58	-14	50.5	2Close	2Close
3:30 PM	TEXASA&M	21	TENNESSEE	29	50	-3.5	55.5	2Close	2Close
3:30 PM	WAKEFOREST	25	VIRGINIATECH	20	45	-1.5	49	2Close	2Close
3:30 PM	FLORIDA	30	SCAROLINA	19	49	-2	51	Away ML	2Close
3:30 PM	MASSACHUSETT	0	PENNST	71	71	-42.5	57.5	2Close	2Close
3:30 PM	TROY	24	ARMY	23	47	4.5	44.5	2Close	2Close
3:30 PM	FLAATLANTIC	32	SFLORIDA	40	72	-2.5	58.5	2Close	2Close
3:30 PM	BOWLINGGREEN	17	BUFFALO	30	47	-5.5	47.5	2Close	2Close
3:30 PM	AKRON	22	CMICHIGAN	26	48	-12	45	Away +	2Close
3:30 PM	MIAMIOHIO	26	WMICHIGAN	23	49	8.5	50	2Close	2Close
3:30 PM	OREGON	38	WASHINGTON	25	63	-2.5	67	Away ML	2Close
3:30 PM	BYU	20	TCU	40	60	-5	53.5	Home -	2Close
3:30 PM	KANSAS	27	OKLAHOMAST	25	52	3	57	2Close	2Close
4:00 PM	OHIOU	19	NILLINOIS	8	27	6.5	44.5	2Close	Under

**usma1992** · 10-11-23, 07:12 PM

Last Games of the Week...

4:00 PM	IOWA	6	WISCONSIN	31	37	-10	36	Home -	2Close
5:00 PM	UNLV	43	NEVADA	14	57	10	56	Away -	2Close
6:00 PM	SANJOSEST	35	NEWMEXICO	22	57	8	57	2Close	2Close
6:30 PM	LOUISVILLE	29	PITTSBURGH	6	35	8	47	Away -	2Close
7:00 PM	MARSHALL	42	GEORGIAST	23	65	1	56.5	Away -	2Close
7:00 PM	WYOMING	2	AIRFORCE	36	38	-10.5	42	Home -	2Close
7:00 PM	ARIZONA	27	WASHINGTONST	37	64	-7.5	60	2Close	2Close
7:00 PM	LAMONROE	15	TEXASST	52	67	-16.5	63.5	Home -	2Close
7:00 PM	AUBURN	24	LSU	41	65	-11	61.5	2Close	2Close
7:00 PM	KANSASST	35	TEXASTECH	30	65	-1	56.5	2Close	2Close
7:30 PM	MISSOURI	28	KENTUCKY	19	47	-2.5	52.5	Away ML	2Close
7:30 PM	MIAMI	25	NCAROLINA	33	58	-4	56	2Close	2Close
7:30 PM	USC	27	NOTREDAME	31	58	-2.5	60.5	2Close	2Close
8:00 PM	NCSTATE	15	DUKE	24	39	-3.5	46.5	2Close	2Close
8:00 PM	UCLA	25	OREGONST	22	47	-3.5	53.5	2Close	2Close
8:00 PM	UAB	39	UTSA	41	80	-9.5	64.5	Away +	2Close
9:45 PM	BOISEST	44	COLORADOST	37	81	7.5	60.5	2Close	2Close
11:00 PM	SANDIEGOST	25	HAWAII	27	52	6	52	2Close	2Close

**usma1992** · 10-11-23, 07:15 PM

So far...

Three games have transpired. Even thought my system didn't say to bet them... two of the three scores were really close. La Tech was driving and had 4 and 4 and didn't get it ... if they proceed to score that 30 to 30 score is dead on.

In addition, the Liberty game was really close. I had 30 to 19 ... I think it finished 30 to 13. Missed on Coastal Carolina game but my system said to close.

However, it did take UTEP ML and they are up 21-0.

Dave

**usma1992** · 10-12-23, 07:12 AM

Has anyone take notice of how close my numbers are to the actual scores.

Four out of five games my scores are almost dead on.

Dave

**A Quant** · 10-12-23, 08:38 AM

Good luck Dave----

Big card you have there.....

16 ATS sides, 4 ML, and 4 Totals.

Hope you kick some ass. Get after it!

**usma1992** · 10-13-23, 09:00 AM

Looks really tough so far.

Dave

**A Quant** · 10-15-23, 08:23 AM

Okay Dave...

The results......

12-12 on the weekend.

8-12 on sides
4-0 on totals.

I am also struggling, it really is difficult to model this stuff. Too many crazy variables. I find-- betting large cards makes it REALLY tough to win-- and the larger your card, the larger your Vig hit is going to be.

Keep grinding, hope this turns for you. Don't mail it in just yet.

**usma1992** · 10-16-23, 07:51 PM

Revamped to include more games...

Ready for week 8. Since my model is based solely on math... the more games the better. Btw ... the Vig... shouldn't be any different than if you bet five games. It is still -110.

I am not giving up and I am ready to post my Week 8 games. FWI... even though my system... hasn't been profitable yet.... I have been for the first time in a while. Broke even on college... with the ML bets.

Went 5-3 on NFL bets.

Dave

**usma1992** · 10-17-23, 06:48 AM

Trying a slightly different approach... Going with the law of large numbers.

AWAY TEAM	PROJ AWAY SCORE	HOME TEAM	PROJ HOME SCORE	TOTAL	VEGAS SPREAD	VEGAS O/U	SPREAD BET	O/U BET
MIDDLETENNST	17	LIBERTY	41	58	-14	55.5	2Close	2Close
WKENTUCKY	34	JACKSONVILLE	30	64	7	60	2Close	2Close
SOUTHERNMISS	21	SALABAMA	42	63	-18.5	52.5	2Close	2Close
FLAINTERNATI	15	SAMHOUSTONST	21	36	-5	39	2Close	2Close
NEWMEXICOST	22	UTEP	17	39	3.5	48.5	2Close	Under
JAMESMADISON	20	MARSHALL	31	51	3	52	Home +	2Close
RICE	30	TULSA	31	61	-3	57.5	2Close	2Close
SMU	43	TEMPLE	14	57	18	57.5	2Close	2Close
PENNST	27	OHIOST	18	45	-4	47	Away ML	2Close
BOSTONCOLLEG	29	GEORGIATECH	32	61	-5	58	2Close	2Close
BAYLOR	31	CINCINNATI	43	74	-2.5	52	2Close	2Close
RUTGERS	21	INDIANA	7	28	5	42.5	2Close	Under
AIRFORCE	34	NAVY	20	54	11.5	37.5	2Close	2Close
WMICHIGAN	14	OHIOU	35	49	-16.5	50.5	2Close	2Close
UCF	28	OKLAHOMA	53	81	-19	63	2Close	2Close
MISSISSIPPIS	19	ARKANSAS	25	44	-7	49.5	2Close	2Close
MEMPHIS	40	UAB	30	70	5.5	63	2Close	2Close

**usma1992** · 10-17-23, 06:49 AM

2:00 PM	CHARLOTTE	14	ECAROLINA	26	40	-7	41.5	2Close	2Close
2:00 PM	LAMONROE	10	GASOUTHERN	55	65	-16.5	58.5	Home -	2Close
2:00 PM	AKRON	12	BOWLINGGREEN	15	27	-7.5	40	Away +	Under
3:30 PM	PITTSBURGH	13	WAKEFOREST	25	38	-2	48	2Close	2Close
3:30 PM	SFLORIDA	43	CONNECTICUT	25	68	2.5	53.5	Away -	2Close
3:30 PM	OKLAHOMAST	29	WVIRGINIA	28	57	-3	50.5	Away +	2Close
3:30 PM	CMICHIGAN	26	BALLST	15	41	4.5	43	2Close	2Close
3:30 PM	BUFFALO	35	KENTST	13	48	6.5	46	Away -	2Close
3:30 PM	WISCONSIN	34	ILLINOIS	16	50	2.5	42	Away -	2Close
3:30 PM	EMICHIGAN	5	NILLINOIS	25	30	-13	43.5	2Close	Under
3:30 PM	MINNESOTA	10	IOWA	12	22	-5.5	33	2Close	Under
3:30 PM	WASHINGTONST	18	OREGON	58	76	-19.5	60.5	Home -	Over
3:30 PM	TENNESSEE	23	ALABAMA	16	39	-8.5	48	2Close	Under
3:30 PM	NORTHWESTERN	12	NEBRASKA	22	34	-12.5	42.5	2Close	Under
3:30 PM	SCAROLINA	22	MISSOURI	37	59	-7.5	60	2Close	2Close
3:30 PM	NORTHTEXAS	25	TULANE	39	64	-20.5	61	Away +	2Close
4:00 PM	TOLEDO	31	MIAMIOHIO	14	45	2	49	Away -	2Close
4:00 PM	TEXAS	49	HOUSTON	13	62	22.5	60	2Close	2Close
6:00 PM	UTSA	32	FLAATLANTIC	29	61	3.5	59	2Close	2Close
6:00 PM	HAWAII	32	NEWMEXICO	27	59	-2	59	Away +	2Close

**usma1992** · 10-17-23, 06:50 AM

6:30 PM	VIRGINIA	12	NCAROLINA	55	67	-23.5	56	Home -	2Close
7:00 PM	APPALACHIANS	47	OLDDOMINION	16	63	6.5	54	Away -	2Close
7:00 PM	TEXASTECH	39	BYU	18	57	4.5	52.5	Away -	2Close
7:00 PM	COLORADOST	39	UNLV	45	84	-7	61.5	2Close	2Close
7:00 PM	UTAHST	36	SANJOSEST	31	67	-5.5	66	Away +	2Close
7:00 PM	COASTALCAROL	43	ARKANSASST	14	57	10	60.5	Away -	2Close
7:00 PM	TCU	39	KANSASST	36	75	-7	57	Away +	2Close
7:00 PM	OLEMISS	37	AUBURN	18	55	7	57	2Close	2Close
7:30 PM	MICHIGAN	31	MICHIGANST	2	33	24.5	47.5	2Close	2Close
7:30 PM	DUKE	13	FLORIDAST	29	42	-13.5	49	2Close	Under
7:30 PM	ARMY	23	LSU	53	76	-30.5	57.5	2Close	2Close
8:00 PM	CLEMSON	33	MIAMI	25	58	2.5	50.5	2Close	2Close
8:00 PM	UTAH	16	USC	28	44	-7	56	2Close	Under
8:00 PM	GEORGIAST	27	LALAFAYETTE	32	59	-3	60.5	2Close	2Close
9:00 PM	NEVADA	23	SANDIEGOST	37	60	-13.5	49	2Close	2Close
10:30 PM	UCLA	47	STANFORD	10	57	16.5	54.5	Away -	2Close
10:30 PM	ARIZONAST	13	WASHINGTON	45	58	-28.5	58	2Close	2Close

**DJK** · 10-17-23, 07:42 AM

Why bother with any system?

If any system could beat sportsbetting, then someone would have come up with one already.

It's all a wasted time. There are too many tangibles those that cannot be predicted.

If you have it betting on sports, then you have it; otherwise don't bet on sports. Just my opinion.

**usma1992** · 10-17-23, 08:03 AM

Though I appreciate the feedback... I think it can be done. And maybe it has been done already. If it had we would probably not know about it because it would be kept a secret.

My system... hasn't completely worked yet, but it has helped many steer away from games. Because I provide mathematical totals... it allows you to see the numbers rather than just hear someone predict a game.

I am profitable this year for the first time in a while. I won on NFL back to back. I am betting all 29 games. The spread bets haven't normalized yet, but they should begin to this week. I thought they would have normalized last week. I was dead on 50% but won three ML bets.

Btw, the enigma machine was thought to be impossible too. Until it wasn't.

Dave

**A Quant** · 10-17-23, 09:34 AM

Originally posted by usma1992

Ready for week 8. Since my model is based solely on math... the more games the better. Btw ... the Vig... shouldn't be any different than if you bet five games. It is still -110.

I am not giving up and I am ready to post my Week 8 games. FWI... even though my system... hasn't been profitable yet.... I have been for the first time in a while. Broke even on college... with the ML bets.

Went 5-3 on NFL bets.

Dave

Dave, let me explain math to you.

I bet 6 games @100 a game, with -110 odds. I go 3-3. I lose $30.

Now I bet 30 games @100 a game, with -110 odds. I go 15-15. I lost $150.

$150 is more of a loss than $30. In fact, its 5x more.

**usma1992** · 10-17-23, 02:38 PM

True... but if ur betting ur bankroll or the same percentage of your roll on both it doesn't matter. Your betting more on the lesser games. You are making a massive assumption that I am wagering the same amount.

**usma1992** · 10-17-23, 02:41 PM

Yes when ur betting 3k vs.600 the vig will be higher... u are correct lol.

**VeggieDog** · 10-17-23, 02:52 PM

35% of the time it works every time.

**DJK** · 10-17-23, 03:18 PM

Originally posted by VeggieDog

35% of the time it works every time.

Probably a bit higher than 35% but not much higher.

**DJK** · 10-17-23, 03:25 PM

Dave,

How good are you with computers?

Are you a programmer?

If you are programming for your system, what are you using?

Joe

**usma1992** · 10-17-23, 04:27 PM

My degree is in electrical engineering with a computer science track. I am now a computer science teacher. Most of this program is simply built in VBA and excel. I know Java fairly well because that is what I teach.

What I am most proud of is that with one click of a button... I can reach out to several websites and grab all the statistical data that I need and put it in a very manageable format. This is harder than it looks because you need to keep all the names correct and websites call them by different names. Florida State ... Florida St. FSU

Then I teach math also and I use a combination of the spread... z score... standard deviation and averages to generate a projected outcome. Unfortunately, college football has such a diversity of games that you really need to wait until week 8 to look at spreads. The over/unders are far more consistent in the early weeks.

But, I am betting every game that I projected this week. The only item... I am really working on is the filtering of games. I had the filter at -10 points for a under bet. Which meant my system had to project a score 10 less than the total. I lowered that to 6.5 which lowered my accuracy but it spread the bets more.

I struggle with the concept of less games with a slightly higher hit rate versus more games with a solid hit rate. I less toward the latter. I know everyone else does not. I plan to grind out this next year and see what I can make happen.

Dave

**DJK** · 10-17-23, 05:08 PM

That is very interesting.

I did EE too for my undergraduate, but ended up with a programming job for the past 37 years. I've been around the computers ever since the micro computers were invented starting with Commodore Vic 20 or something like that and I even worked for Commodore for a couple of years as a software reviewer.

I programmed in VBA and Excel for as far as I can remember among other languages until my company decided to go away from Windows and went to Linux, so I now program in Python mostly but I still program in VBA and Excel for certain requirements for the clients.

I do have a VBA program in Excel to show me the past game results with a click of button, but honestly it's not that useful in my opinion as what's going on current is what matters the most; although the historical data does matter in some games.

Anyway, at the moment, I'm writing a program in VBA and Excel to track baccarat game results to see if a trend that I noticed recently is profitable. If it is, then it's going to be free money if there is such a thing.

Good luck to you in your endeavor. I hope it works out for you.

**A Quant** · 10-17-23, 07:38 PM

Originally posted by usma1992

True... but if ur betting ur bankroll or the same percentage of your roll on both it doesn't matter. Your betting more on the lesser games. You are making a massive assumption that I am wagering the same amount.

You have NOT ONCE, posted that you bet more on some plays, so yeah, you can see why I would make that assumption.

And, for someone who created a math model, you should know and understand, the more games you play, --- inevitably, variance will always bring you back to 50%--which is gambler's hell.

Every successful gambler I have known or met--- understood, the ONLY way to beat this game is finding 4-7 games you feel you have an advantage on, and hammer them.

**usma1992** · 10-18-23, 05:31 AM

First, I think you were too busy teaching me math and somehow made the assumption that I was going to increase my bankroll by 5x to bet games. I think it is way easier to make the assumption that you maintain current bankroll and spread over more games. Nonetheless, enough of that.

I truly do understand the prevailing wisdom of bet less games and pound them hard. The reality is that is why you double down in black jack when the odds are in your favor.

I also understand there is a regression back to 50%. However, since we only have to achieve 52.38%... Keeping it above that mark seems very achievable in a 13 week college football season. That is what I am attempting to do. Up to this point, my system has been very strong on over/unders but not on spreads. I hope week 8 proves differently.

Sports betting is intriguing to me because everyone believes you can't out math Vegas. I am not of that opinion. We have several advantages which don't exist in games like black jack.

1). They have to act first. Vegas has to produce a line for every game and we get to make a decision after it is produced. In black jack, you must decide your fate prior to the dealer.
2). We can pick and choose the games based on our analysis. We don't have to bet a single game if we don't feel right.

My models advantages:
1). Mathematically driven and can produce bets within 3 minutes of the lines being produced.
2). If I can produce a 50% run rate on a week basis... I can also use the spread of games to my advantage. If I do poorly in the morning games... I can bet the afternoon heavier or vice versa.
3). Zero bias other than teams playing a wide variation of mismatches upfront. Strictly math driven.

Trends on upcoming games.

North Texas +19.5 minimum negative against North Texas bet.
Tenn Under Strong trend against my under bet.
Eastern Michigan Under Strong trend against my under bet.

Northwestern Under Strong trend for my under bet.
UTEP Under Moderate trend for my under bet.
Coastal +10 Moderate trend for my spread bet.
Utah State +5.5 Strong trend for my spread bet.
Hawaii +2 Strong trend for my spread bet.
Oregon -19 minor trend for my spread bet.
Rutgers Under Strong trend for my under bet.
Toledo -1.5 minor trend for my spread bet.
Oregon Over Strong trend for my over bet.
Penn St. ML Moderate trend for my ML bet.
Georgia Southern -16 Moderate trend for my spread bet.

All other games have no trend. 3 lines moved against me. 11 lines moved with me.

Dave

**A Quant** · 10-18-23, 07:33 AM

Originally posted by usma1992

First, I think you were too busy teaching me math and somehow made the assumption that I was going to increase my bankroll by 5x to bet games. I think it is way easier to make the assumption that you maintain current bankroll and spread over more games. Nonetheless, enough of that.

I truly do understand the prevailing wisdom of bet less games and my system has been very strong on over/unders but not on spreads. I hope week 8 proves differently.
pound them hard. The reality is that is why you double down in black jack when the odds are in your favor.

I also understand there is a regression back to 50%. However, since we only have to achieve 52.38%... Keeping it above that mark seems very achievable in a 13 week college football season. That is what I am attempting to do. Up to this point,
Sports betting is intriguing to me because everyone believes you can't out math Vegas. I am not of that opinion. We have several advantages which don't exist in games like black jack.

Dave-- with all due respect... what?

You totals are hitting at 54.2%.

Overall your posted plays are hitting at 43%.

Your sides are hitting at 34%, and MLs at 36%.

I really really hope you can turn this around, but I think you are learning that creating a model for the entire college board is proving to be very very difficult.

**usma1992** · 10-19-23, 05:41 AM

Actually, I had a few epiphanies last night on my approach. The epiphanies were not about how many games to bet or bankroll.

My epiphany came from examining my projected scores. After watching many, many games I made the assumption that they were coming in reasonably close so I there for went back to the individual z scores of the teams for several of the statistics to figure out if any specific scenarios matched up. When my projected totals and the actual totals were the same... what z scores and stats mattered.

I stumbled upon a specific z score of a specific statistic. If that statistic was extremely negative yet my projections were positive ... my results were above 64%. I back tested it and then I tested it on 2023 to be sure.

My philosophy has always been to try to find the combination of stats that matter to win. What if? I am just looking for a stat or too that Vegas has to take into account that doesn't matter as much as they have to account for in their totals. If that stat happens to normalize or go the other direction for that game I should win.

My projected totals already take that stat into account as is but if the stat becomes stronger toward my output. I believe that not only am I on to something... it will change my approach to all of my programs. Even though NFL is winning and has made me money I also now know I need to revamp that program because it is taking z scores into account too much. I will be ready for college basketball and NBA.

Enjoy... I probably did a poor job explaining.

The only additional game it said to bet was Oklahoma OVER. The initial line was 63 now... it is 66. It still meets my system requirements.
Dave

**ChuckyTheGoat** · 10-19-23, 09:12 PM

Originally posted by usma1992

My degree is in electrical engineering with a computer science track. I am now a computer science teacher. Most of this program is simply built in VBA and excel. I know Java fairly well because that is what I teach.

What I am most proud of is that with one click of a button... I can reach out to several websites and grab all the statistical data that I need and put it in a very manageable format. This is harder than it looks because you need to keep all the names correct and websites call them by different names. Florida State ... Florida St. FSU

Then I teach math also and I use a combination of the spread... z score... standard deviation and averages to generate a projected outcome. Unfortunately, college football has such a diversity of games that you really need to wait until week 8 to look at spreads. The over/unders are far more consistent in the early weeks.

But, I am betting every game that I projected this week. The only item... I am really working on is the filtering of games. I had the filter at -10 points for a under bet. Which meant my system had to project a score 10 less than the total. I lowered that to 6.5 which lowered my accuracy but it spread the bets more.

I struggle with the concept of less games with a slightly higher hit rate versus more games with a solid hit rate. I less toward the latter. I know everyone else does not. I plan to grind out this next year and see what I can make happen.

Dave

us, very impressive. You're a talented guy. I may in fact reach out to you in the future to help me out on a few data-mining tasks.

I'd give just a few general responses:
1) Think about the strengths of your skill-set. Being able to retrieve data is a talent.
2) Integrating this skill into the betting markets is NOT that easy. It takes testing, be patient.
3) Think about the betting markets from the perspective of the ODDS-MAKER. Think about what their rubric is, and think about whether they might have some flaws.

Those are my general comments. Will talk to you more.

**usma1992** · 10-20-23, 07:30 PM

Lol... don't be too impressed. My strengths are persistence... and thinking outside the box and analysis. I'm still learning on all three fronts. However, I am really proud of my adjustments and starting to understand how I can use my projected totals.

This lesson will be applied to all my models. Vegas is vulnerable... lol. Not really but I am going to think that way.

Dave

**Rrwolverine** · 10-20-23, 07:36 PM

Thanks for sharing this