Browsing the internet i found this interesting article.
https://www.baseballprospectus.com/news/article/37031/flu-like-symptoms-many-runs-win-game/
It was written by Rob Mains on Baseball Prospectus and he discussed a long known sabermetric principle.
10 runs roughly equals one win.
Such notion can be very disputed and it has been, because of one simple reason.
It's illogic.
Anyone who has ever seen a baseball game knows you don't need 10 runs to win a game, so why do the numbers back up that statement.
Before i assess that, here is a quote from his article with a formula based on Run Differential that proves such principle is true.
" I went through every team for every season since 1901, when the American League was formed. That gave me 2,468 team seasons. For each, I calculated the run differential (my x’s) and the team’s wins (my y’s). Then I used Excel’s linear regression software (you didn’t expect me to do this by hand, did you?) to generate a regression formula. It is:
10 runs roughly equals one win and you need 10 runs to win a game are two completely different things.
When someone says 10 runs equal a win it doesn't mean you need 10 runs to win the game, it means over the course of a 162 season with multiple inevitable losses, every 10 runs more you get will net you an extra win.
The important thing to realize is this, a significant number of runs a player's responsible for, through the course of his career will come in losses.
How many runs has Mike Trout given the Angels in the games they lost?
Hundreds.
What happens to them when calculating how many runs equal a win, they are a part of it.
A win may come with 5 runs, but other 5 runs may become useless with a loss.
This is actually very simple, the controversy comes from an assumption that 2 statements mean the same thing when they don't.
As always feel free to comment, suggest and subscribe for more content, if you'd like to support this blog, get ad-free content along with exclusive posts head over to my Patreon homepage, the link is on the sidebar, any help is appreciated.
See you tomorrow.
https://www.baseballprospectus.com/news/article/37031/flu-like-symptoms-many-runs-win-game/
It was written by Rob Mains on Baseball Prospectus and he discussed a long known sabermetric principle.
10 runs roughly equals one win.
Such notion can be very disputed and it has been, because of one simple reason.
It's illogic.
Anyone who has ever seen a baseball game knows you don't need 10 runs to win a game, so why do the numbers back up that statement.
Before i assess that, here is a quote from his article with a formula based on Run Differential that proves such principle is true.
" I went through every team for every season since 1901, when the American League was formed. That gave me 2,468 team seasons. For each, I calculated the run differential (my x’s) and the team’s wins (my y’s). Then I used Excel’s linear regression software (you didn’t expect me to do this by hand, did you?) to generate a regression formula. It is:
Wins = 0.103 x runs differential + 78.2
Looking just at the first figure there, each run gives you 0.103 more wins. Flip that over, and you get a win equal to 1 / .103 = 9.74 runs. Or about 10 runs. Each additional 10 runs scored, compared to runs allowed, equals a win."
Considering this, how can something baseball related be at the same time, true and illogic.
The answer to that question is very simple.10 runs roughly equals one win and you need 10 runs to win a game are two completely different things.
When someone says 10 runs equal a win it doesn't mean you need 10 runs to win the game, it means over the course of a 162 season with multiple inevitable losses, every 10 runs more you get will net you an extra win.
The important thing to realize is this, a significant number of runs a player's responsible for, through the course of his career will come in losses.
How many runs has Mike Trout given the Angels in the games they lost?
Hundreds.
What happens to them when calculating how many runs equal a win, they are a part of it.
A win may come with 5 runs, but other 5 runs may become useless with a loss.
This is actually very simple, the controversy comes from an assumption that 2 statements mean the same thing when they don't.
As always feel free to comment, suggest and subscribe for more content, if you'd like to support this blog, get ad-free content along with exclusive posts head over to my Patreon homepage, the link is on the sidebar, any help is appreciated.
See you tomorrow.
No comments:
Post a Comment