College Basketball Picks & College Basketball Predictions

First off this will be on the longer side but time and effort is what it takes to win when betting on every game, all season long.
I envy those who can look at a matchup between teams in any sport and know more or less how they will matchup. For NBA this means knowing how teams will build their lineups depending on their opponents and which players can counter stars on the other teams. For MLB this could mean taking into account variables like stadium, weather, batting order, and what not. Generally across all sports being able to look at a certain matchup and the previous games between the two teams and using that knowledge to make an educated pick is a highly valuable skill. Being the degenerate that I am, I want to be betting all year round and as my favourite sport, basketball, only spans the winter months I wanted a way to break into the MLB betting (without losing).
My goal was to make a compeletely mathematical model in arguably the sport that has not the most available stats but the most meaningful stats. My first step was to build my own database on google sheets (not excel). Using javascript I managed to make sheets that could download the box scores for every game and the roster for every team.
The next phase of creating a system is the tough part. You have the hits, runs, innings pitched, batter faces, and pitches thrown for each pitcher. This allows me to calculate ERA, WHIP, OBP, B/IP, and P/IP on my own. That saves me a lot of lines of code otherwise needed to import all those stats into my own spreadsheets. The problem at hand is now figuring out which stats will predict what outcome. Big three bets are full game: spread, run total, moneyline. For someone who doesn't want to use high level calculus to get picks the immidiate solution is to reduce everything to a graph with a line. For example: ERA vs Hits/Inning, the possibilities are endless. Excel and google sheets have CORREL (correlation between two data sets) and LINEST (gives you an line of best fit for two sets of data) which makes it easy to pour out hundreds correlations between stats. So I set my correlations and equations between various pitcher (ERA, WHIP, OBP) stats as my X variable and game stats (At Bats, Runs, Hits) as my Y value. With all these equations and correlations I was now set to find what I call the "trail" of stats that would lead me to best results.
The trail is like a flowchart of stats that ultimately leads to what you want to predict, in my case, runs per team. The best system isn't the one that has all the right trails. A little side note I have been betting on the NBA spread and point totals for two seasons now. The odds are always -110 to -120 so a 55% win rate guarantees profit. This past regular season I was 210-143 (59%). What I noticed was that the trails change during the course of a season. At the beginning I could reliably predict each players individual points and add them up but towards the end I started predicting points by position instead. ANYWAYS, the goal is to always have many trails available and to track each one constantly to see if any of them become "hot". An example of a trail that I use for MLB is this :
have a graph and equation for opposing teams Pitcher's ERA in the last 2 weeks for all Yankees games vs the number of at bats for the Yankees. I then have graphs for At bats vs Hits and Hits vs Runs. You use AB to calculate the H's and then H's to calculate the R's.
Now this isn't the usual premise of a gambling system "78% WIN RATE" those "systems" are only at 78% because their definition of a win differs from ours. This is a system where you bet 5$ on 3 bets per 15 games ($225 risk) and you win 30 bets and 270$. Its not the thrill of winning a super parlay but it turns your hobby into an investment. Now that I have set up about 30 ways (trails) to predict runs I'm all set to go out and make some money.
I spent all last season betting on just run totals and went 60% with amount bet totaling just over 10k and amount won 15k (5k profit) I have little interest in making money off my system by selling because I make money off betting from it (that's what i don't get about sites selling you systems, why would they sell it if they have a 78% win rate LOL), so I decided to take to reddit to share what I do and how I do it. With all that being said I am just a college kid with no job (I quit my serving job a while ago) and not everyone has the time to spend 3 hours a day minimum collecting and organizing data. Some people may not have the excel, javascript, python, c++ knowledge I do or the time to learn it. The way I look at it a betting system is like a drug (pharmaceutical). If it solves a problem for you then you should make it available to everyone and for free.
That's why I will be posting daily MLB picks when I have the time (most days) and hope that you follow them. The idea is to bet the same on every bet and bet all bets. The MLB gives us the luxury other leagues don't, frequency. In NBA i would hit 7/7 spreads on day and 1/4 another. With 8-15 games every night this lessens the chance of a crazy loss but also of a crazy win. Hopefully we can all pocket some money this season and learn a thing or two while we are at it. I will be posting picks on the daily threads all season long but eventually my goal is to have a free app that will source my database directly. Since the lines never chance but only the odds on spread and run totals its okay to post picks early in the day.
If you've read this far thanks and even if you don't want to do the volume betting that I do track the picks for a week then decide again. I'm always spending time refining my system, I believe if you don't put in the hours you won't see results so I refuse to sit back and just let what I have made already run.
Looking forward to a plentiful season :)