Time for a quick word association game: If I say, “college football analytics,” what do you think of? Something involving Moneyball or when to go for it on fourth down, right?
A standard definition for sports analytics is gathering information and applying it in a way that derives a competitive advantage. Translation: doing what football coaches do, only with more help from computers.
Analytics are a path toward a winning edge. They see every game. They separate emotion from reality. They separate what you can control from what you cannot.
You only have to hit 52.4% of your wagers to break even at -110 juice. A 50 percent handicapper is going to lose you a lot of money, while a 55% handicapper is going to win you a lot of money. That doesn’t seem like a very big difference, but in sports betting it’s absolutely enormous.
Some amateur bettors scoff at the idea of pulling down 5% of their bets, which is what your profit would be on a 55% winning percentage. I certainly don’t scoff at it and have had some of my best years profit-wise in picking 55%. Averaging just 5% each week you bet is going to make you a nice chunk of change over the course of a year.
Let’s do an example just to see how much you’d make at 55% winners. If your starting bankroll is $5,000 and you make 5% on that the first week, that’s $250. The next week it would be $262.50 and the next week after that would be $275.60. In a year, you would be up about $15,000. So, you’ve essentially tripled your starting bankroll of $5,000.
Instead of winning 5% on cash, let’s say you’re winning 55% of your betting units. If you play 10 units per day, you have 1 more unit every other day with a 5% return. You’d also make between $15,000 and $20,000 in a year. Not only is that a lot of money, but it’s realistic if you can find a consistent handicapper who hits 55%.
After you’ve found a handicapper that can hit 55%, then it’s on you to make sure you take advantage. You’re going to be betting multiple plays per day in various sports, so staying consistent with your wagers will be a must. You don’t want to up your bets when you’re winning or chase your losses when you’ve hit a cold streak. Stick to a dollar amount per unit and don’t waver from it.
You want to reinvest the cash you make raising your unit bet according to your growing bankroll. If you’re betting 2% on each play at $5,000, that would be $100. If your bankroll grows to $10,000, then your 2% would be $200. You could conceivably turn that $5,000 into over a million in less than 8 years if you find a handicapper hitting 55% and adjust your bets according to your bankroll size.
By treating sports betting as a business instead of a hobby, you can seriously turn a profit like a well-run business. It’s great if you can figure out a way to hit 55% for yourself, but if you can’t, there are plenty of handicappers here at Betfirm that will help you do just that. Check out the leaderboards and find the best handicappers over time who hit around 55% in all sports.
When constructing a ranking system of individuals or teams, some issues need to be addressed. Danehy and Lock (1995) say that any ranking system should order all teams, compare teams, adjust for the quality of the opponents, predict outcomes of games, and predict the game scores and differentials.1 Statistical methods to rank teams have been widely used in sports. The following is just a brief list of some of the work that has been done in the past. Harville applied linear-model methodology to the point spread of games in college football.2 This methodology can be applied to high school football. Stern (1995) used a least-squares approach to rank college football teams. Danehy and Lock (1995) used an additive least squares model and Lock and Danehy (1997) used a multiplicative Poisson model for collegiate hockey rankings.3 Glickman and Stern developed a predictive model based on a state-space model that assumes team strengths follow a first-order autoregressive process.4 Annis and Craig used a hybrid paired comparison model that incorporates wins and scores to rank college football teams.5 Kvam and Sokol presented a combined logistic regression/Markov chain model that was applied to NCAA basketball data in order to predict the NCAA basketball tournament games.6
The model used to produce the PING ratings is a variation of the method developed by Stern (1995) to rate college football teams as well as predict individual game outcomes. The goal of the ratings is to account for the team’s performance and their strength of schedule.