What are projected lines and odds?
Bookmakers odds are nothing else than probabilities turned into the odds. Those odds are the prices that we must pay. If we pay big price all the time, then we will lose in sports betting.
When bookmakers have a game, where they think, that both teams have 50% of chance, they will always include some margins, so the bettors will pay much more.
If the odds would be fair on 50% game, then the odds would be 2.00 and 2.00 (US odds +100). Right?
They usually offer us odds around 1.95 on 50% games.
My goal with my projected odds and lines is to find my winning percentages, which will be better than bookmakers odds and which will beat their margins too. I don’t pretend, that I have better models, than bookmakers, because this is not true for every single game. They hired the best experts from different fields and it is very hard to beat them. If you use only intuition or easy-to-find find information on the internet (which is normal case right now with the internet) you have close to zero chance to beat them on the long run.
However, the good thing is that they also don’t have correct odds (like in a flip coin experiment, where we all know that the chances are 50%), because sports is little bit more complex, every single game is a new story and of course they will also move the odds because of the market.
This is where I come in. I project my winning percentages for games and then I compare my odds with bookmakers odds. I have my own philosophy about this:
I usually wait for some numbers of games to be played in a new season and then I use current season numbers to project my odds. The formulas are still based on big number of games from past seasons (for example I researched last 10 seasons in NHL to get formulas how to rank teams and how to calculate win%).
I am selective about those games and I look for big difference between my odds/lines and bookmakers numbers. I use Kelly criterion for calculating the stakes and qualified bets on moneyline bets (MLB, NHL) and I look for big difference between my spreads and the totals in basketball (7 for spread and 10 for totals).
For example, if my betting model gives Boston 60% to win a game, my fair odds on this game would be 1.67. If bookmakers think, that Boston has a 50% of game, then they can set (with margins) the odds of 1.95 on Boston. In this case I expect at least $67 of profit if I bet on Boston (and if they win of course). But when I see, that bookmakers offer me the odds of 1.95, that means, that they are willing to pay me $95 instead of only $67 on Boston. And in this case I have a value with Boston.
Betting experts always say, that we must bet +EV and this is the only way how to beat a bookmaker. The problem is that most bettors don’t use analytics or any statistics at all. No matter if they sell picks or they bet for themselves. Even if you find some betting site, that use algorithms and betting models, they will not reveal formulas behind this. Some betting sites will use also very complex methods, using some complex programming languages and of course this is too complex to learn for most of us. Only if you go back to to University and start studying statistics, math and programming languages. Some of those sites will provide those lines and will charge huge amount of money for monthly subscriptions. Most bettors simply can not pay monthly subscriptions with their budgets.
I always recommend, that you use my numbers with other information, that you have. This is great tool for bettors from different sports to include my numbers into their analysis. Analytics is more and more important and I can not imagine betting without estimating and finding a value trough statistics.
My basketball method was one of the best in the world in 2017, using only my projected odds without any other research of basketball games: 236 – 149 – 7, +59.29 Units, Yield: +15.12% (1 Unit Flat/Bet Record Analysis). I project the lines for all kind of basketball leagues all-around the world. Then I compare my lines with bookmakers lines and when is the difference big enough between my lines and bookmakers lines, I bet. If you bet on basketball you can use my numbers are part of your analysis. I start betting on basketball somewhere at the end of January and bet until playoffs. Go to Basketball projected Lines.
Baseball is my favourite sport, because this is a sport of statistics and no other sports has so many games in one season. Blindly betting my model was very successful and if you played the odds between 1.75 and 2.85, the model has even better results (500 picks, yield: +12.24% in 2017). My projected odds on MLB starts in May, after first month, when I have enough stats to calculate win% and until the end of the season in October. Go to baseball projected odds.
NHL ICE HOCKEY
NHL is my third sport and I projected odds for every single game. I project how many goals will score both teams and then I project my own odds. The information about injuries and starting goalies usually come later during the day, so I adjust odds and win% manually later, because of this. It is recommended, that you use my numbers in combination with your information. I project NHL odds from November until the end of the season. Go to NHL Projected odds.
NFL AMERICAN FOOTBALL
American Football is a new sports for me and this is more a challenge for me. The problem with NFL is that there is not a lot of games and it is little bit harder to predict lines at least at the start of the season. This sports is great for bookmakers, who use all kind of historical data and much more complex models. NFL is also must popular sport in NFL, where a lot of gamblers bet based only on intuition, some trends or maybe some information, they have. There is still a lot of work for me.
Coming soon. I don’t bet soccer, but because there is big interest, I will create one model in for my followers. I will also show you how you can create this betting model and then you can calculate the odds for yourself for any league you want.