A MLB season consists of 162 games for each team. Each team in a division plays the other division rivals 19 times each. The season starts in late March or Early April and regular season games end in September or early October making the season 7 or 8 months long. MLB has the most difficult standard of making the post season of any sport. It allows only division winners plus two wild card teams. The back story about the season, games, and opponents is important because we must understand and value (not underestimate) the nature of the game being played as a daily grind. Their are 81 away games so players and coaches will travel to multiple cities. They stay in nice places mind you so it mitigates wear on the body in the sense of sleep derivation or dealing with rough conditions.
The grind brings into play the importance of current performance and current conditions. Teams and players will go through periods of time when they are above their mean average and below, thus will have extended winning streaks and losing streaks. When handicapping this is vital knowledge. There many “old pros” who want accumulated data to ensure and verify the data is as accurate as possible. MLB is not the sport for that. Shorter trends in performance are more valuable than longer ones! Look at it this way. Let’s say SF Giants have a .440 winning percentage and we are in June. They have just come off a series where they lost 8 or their last 10 games. We hare handicapping their next game and they play COL who has a winning percentage of .400. COL has won 6 of their last 10 games. When comparing the teams, do we use the season long information or would it be more accurate to only include the shorter term data?
The answer has two parts. The overall team performance is something we need and utilizing the winning percentage accomplishes that. I use BaseRuns win/loss percentage (another subject for another time). We can give the team a base strength this way. Now comes the hard part. We do NOT analyze the players that way. We want to capture and use the nuances of the season to our advantage. We do this but analyzing data by time periods for players. Starting pitchers is last 30 days. The reason is using his last 5-7 starts will provide us the info we need to then compare to his season to date results giving us the ability to use the performance variations to our advantage. When looking a hitting data, I use last 7 days. I know that seems very short and it is, just remember we don’t care what this hitter did in April because we are handicapping a game in June. His April is gone and will not in anyway effect the outcome today. Seven games has been the right target to capture enough data to make it real yet still capture those variations we are looking for. I only use the same time period for relief pitchers. They do not throw everyday, but they do throw 3 or 4 times and week. Each area can have above and below performance outliers. The hitting maybe doing above average whereas the starting pitching may be below average for example.
We need to identify streaks not only the obvious winning and losing ones. Our edge will be getting on the streaks whether the team/player is going good/bad as soon as we see the variations. The sportsbooks take time to get their lines adjusted to these outlier performance variations. We want to ride the streak! Play on above-average performance and play against below-average performances. DO NOT think you can predict when the variations will stop. You can miss an opportunity if quit riding the streaks too early. Keep in mind the sportsbooks will gain knowledge and adjust the betting line to reflect streaks too. The longer one goes, the more juice to bet into there will be. Performance variations are expected and a normal part of an MLB season. They are our friends. Treat them well and they will reward you.