I'm making a change on model leans. They have been so damn bad and even though I don't play or fade them...it's just a ref point. The way my model was set to calculate was on 21% difference for PLAYS and 11% for LEANS. Or in other words, when my model showed the true percentage to win as over 21% more than the moneyline converted implied percentage chance of winning, it is a play.
Example today: REDS +139 today implies a 41.8% chance of winning. My model gives them a true 51.7% chance of winning, including a near 8% add for the difference is starting pitching today. Therefore, if you take 41.8% plus 21% of that equaling 50.58%, anything above that constitutes a PLAY.
Same thing on leans but with 11% difference and under the 21% threshold which would be a play. Anyway, I'm changing the 11% to 16% and when I do it starting today it also backdates the whole season to only include the LEAN being under 21% différance but at least 16%. Haha, all that horse-shit for what amounts to basically a nothingburger. I just try to explain my actions and be a transparent as possible.
So the updated plays for today include no LEANS and far fewer LEANS this far in the season. Obviously, there is no change in the PLAYS for today or the season. Later in the season, I generally lower the percentages a bit as lines get tighter.
Here's all teams record when facing teams with a winning record(above 50% wins)
