We should realize MLB has segments of seasonality that have effects on game outcomes, and all MLB teams have spurts of success and failure over a 162-game season. These factors are areas we can exploit if we look a data properly.
A complete MLB season starts in late March or early April depending on the year. It is 162 games long completed in October making it 6 months long. During those 6 months, the weather changes can be dramatic. The start of a season will have many games where the temperatures are cold and later in the season, the temperatures will be hot. There is a noticeable difference between the games in cold weather and hot weather. It applies mainly to the runs scored. In general, run-scoring is less when it is cold and more when it is hot. This is valuable data as the total for games needs to be adjusted as the season progresses in the cold climate cities. Be careful blindly wagering on cold games. It has become more of a known factor, so sportsbooks have adjusted the totals. The following chart indicates temperatures and runs scored.
In general, run-scoring is less when it is cold and more when it is hot. This is valuable data as the total for games needs to be adjusted as the season progresses in the cold climate cities. Be careful blindly wagering on cold games. It has become more of a known factor, so sportsbooks have adjusted the totals. The following chart indicates temperatures and runs scored.
As you can see, in general when it is cold, there is much less scoring. I found it interesting though that the scoring average increased with temps under 40. There are some reasons this could be the case and the main one is the starting pitcher’s grip on the baseball. As their fingers are cold or numb, their command and control will be less. The only true batter performance increase is the walks whereas batting avg, slugging pct, and such all are less. Over the past 10 seasons, the OVER has gone 43-33-3 in games under 40 degrees weather. A sign that the books have adjusted. The vice versa is also true for the games in the heat. When the weather is 90-99, it is most optimal. These are scenarios to think about when wagering games from early April in Detroit or the middle of May in Texas!
Another factor of weather effects on MLB scoring is air density! A scientific fact is that a baseball will travel further in low-density air and less far in high-density air. Makes sense to me. We are going to be looking for something called the Neely Scale. It is an index of air density with a range of 40-70. The lower the number, the less dense the air is. A famous ballpark for low dense air is Coors Field in Colorado. It is at an elevation that creates this low density which allows the baseball to travel further. In recent seasons, humidors have been introduced. Gameday baseballs are stored in these humidors to offset the air density of certain ballparks. These humidors will change the composite makeup of the baseball which is reflected in the ‘bounciness” off the bat it will exhibit. The goal is to level-set all the ballparks to a standard as much as possible. However, the bottom line is that air density allows or restricts the distance a ball travels thus the denser the air, the less far it will travel, indicating fewer damaging hits and lower scoring games. Keep an eye on this one.
The long season will also allow every team to have spurts of success and failure. The best teams in the game will have days and weeks where they did not play their best. The worst teams will have their time where they are playing great too. We need to pay attention to the streaks teams are on. Due to the length of the season, MLB will have longer streaks both winning and losing ones. The best advice is to not predict when the streak will end but be riding the streak as long as possible. The better teams tend to have more winning streaks and the losing teams have more losing ones. A common streak is 3-5 games, but it is not rare to see them extend beyond.
Handicapping these games requires a set of data points unique to the sport. Anyone who delves into the numbers will tell you that a stat is only as good as the data and to have a good stat, you need an adequate supply of data. This is where a balance is required for MLB. Let’s say you are handicapping a game in the middle of July. Ask yourself this question. Should it matter to the outcome of this game what a player produced from April? The obvious answer is NO! If this is the case, then why are season-long stats used to handicap the game in July? They shouldn’t be! However, we still need enough data to get a reasonable suggestion of performance so we can predict how well the player or team will perform today. We need current performance indicators. I utilize these sets of numbers. A starting pitcher will make 32 starts if he doesn’t miss a turn in the rotation. I use his last 30 days as a fair representation of his current performance. He should make 5 or 6 starts and throw roughly 30 + innings. The bullpen pitchers are used more frequently so we need a shorter time frame for them. If 3 weeks ago they were lights out, does that mean today we can expect the same? NO. I utilize the last 14 days. I have found that if I use a shorter time frame, one bad outing can skew the results too badly to be relied on for accuracy. A reliever will normally make 6-8 appearances in those 14 days which is enough to give us an indication of what his current performance is. The batters are similar to the relievers, but they get hot and cold quicker. This means we need a shorter time frame to get the most accurate performance. I use the last 7 days for hitters. I think you can see the overall idea here is to gauge current performance. We cannot do that by using season-long data. If we do this right, we can capture the spurts and react faster than the books giving us many edges and opportunities to capitalize.
