When you go through the table below and run the math on the Houston Astros recently becoming the first team since 1887 to win four straight games by at least 9 runs, you see the odds are 1 in 1.9 million. A team has only 2.68% chance of winning by at least 9 runs in any given game, so the chance of doing it four times in a row is calculated by 0.0268 x 0.0268 x 0.0268 x 0.0268 which equals on in 1.9 million.
For the Cleveland Indians historic 22-game winning streak, you simply multiply the 50-50 chance of winning a given game 22 times in a row for 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5, which equals 1 in 4.2 million.
One suggestion for anyone who plays video games or old board games like Statis-Pro Baseball, Accurate Baseball Made Simple or APBA, is to give the winning team credit for a three-game sweep whenever they win by at least five runs but give them a 2-1 record anytime they win by four runs or less.
Because it is hard to play enough games for accurate records, this gives a greater chance of better teams having more reflective records over time. Here are the margins for all games, with the negative indicating the home team one and the positive number indicating the home team won.
Road team margin | Occurances | % of time | running % |
---|---|---|---|
-17 | 1 | 0% | 0% |
-16 | 1 | 0% | 0% |
-14 | 2 | 0% | 0% |
-13 | 4 | 0% | 0% |
-12 | 7 | 0% | 1% |
-11 | 15 | 1% | 1% |
-10 | 19 | 1% | 2% |
-9 | 17 | 1% | 3% |
-8 | 37 | 2% | 4% |
-7 | 59 | 2% | 7% |
-6 | 73 | 3% | 10% |
-5 | 105 | 4% | 14.0% |
-4 | 135 | 6% | 20% |
-3 | 181 | 7% | 27% |
-2 | 221 | 9% | 36% |
-1 | 410 | 17% | 53% |
1 | 276 | 11% | 64% |
2 | 224 | 9% | 74% |
3 | 183 | 8% | 81% |
4 | 141 | 6% | 87.0% |
5 | 97 | 4% | 91.0% |
6 | 62 | 3% | 94% |
7 | 62 | 3% | 96% |
8 | 31 | 1% | 97% |
9 | 28 | 1% | 99% |
10 | 16 | 1% | 99% |
11 | 9 | 0% | 100% |
13 | 7 | 0% | 100% |
14 | 1 | 0% | 100% |
15 | 1 | 0% | 100% |
16 | 1 | 0% | 100% |
19 | 1 | 0% | 100% |
Total Games = | 2427 |
Thanks to MLB Sweeps for compiling the series sweep records in three game series, something I have wanted for decades. As you can see, while the chance of either a Sweep or a win in a single game by at least five runs is 14% in both cases, the home team has a better shot at a sweep than a 5-run win (17% to 14%), but that is balanced by the road team having a better chance at a 5-run win than a sweep (13% to 11%).
2016 counts of games and series | Occurances | Percentage |
---|---|---|
Total 3-game Series | 552 | |
Home Team Sweeps | 95 | 17% |
Neither team swept | 395 | 72% |
Road Team Sweeps | 62 | 11% |
As a result, the following is the suggestion for playing out a short season by counting each game as three games on the record:
Proposed process for building records (every game counts as
a 3-game series until playoffs)
1. If either team wins by at least 5 runs, they are credited
with 3 wins to none
2. If the game is decided by within 4 runs (or extra
innings) then winning team gets 2 wins and losing team 1 win
3. Home team with a lead can finish their final at bat
(bottom of 9th or extra inning) to try to get to 5-run margin
4. Teams can use best relievers to try to preserve or get to
5-run game
This process lowers the chance of bad teams getting lucky
because "luck" typically results from bad teams barely beating good teams
in some games while being blown out more often.
Going into a game each team has a little better than 13%
chance of winning by at least 5 runs.
Going into a 3-game series, each team has a little better
than 13% chance of sweeping all 3 games.