I had a bad beat the other day where a pitcher had great stats the last 3 games and then got blown out by the 2nd inning. So, I started looking more closely into the pitcher's record and he had very few starts that left his team in a position to win the game, so the record over the last 3 games was an anomaly.
I added another dimension to my capping to keep that from happening again. I list the # of Games Started, the number of Good Starts, and the number of Bad Starts. While researching what might be a "Good" start, I looked at the Quality Start stat that some like to use. They use a game ERA of 4.5. I don't like the way they do that stat, but I wanted to look into this ERA of 4.5 because I thought "why 4.5? why not 2 or 8 or 3 or 6"?
I started running queries that show what a team does when a starting pitcher has a game ERA above or below 4.5. The results are pretty astonishing. The records aren't even close.
If the game ERA is <= 4.5 pretty much the team wins the game. The % varies by pitcher and team but it is between 70% and 90%.
If the game ERA is > 4.5 then pretty much the team loses the game. The % again varies by team and pitcher but it is again between 70% and 90%.
I find this extremely intriguing that there is a number at which above and below the Win/Loss result is so drastically opposite.