Welcome to the first projections for the remainder of the 2022 season as we approach the half-way mark.
These are based on two things, each team’s current record as the point of departure, and runs scored and allowed, which are the best predictor of future wins. They are then put into a league framework, since with interleague out of the way, the maximum number of wins left in each league is known.
I don’t expect the Swallows to win the league by 19 games, because if they do clinch, their focus will shift as much to keeping everyone healthy and involved rather than winning as many games as humanly possible, and only one team in history, the 1990 Giants, has won by more than 18-1/2 games.
The median NPB margin for pennant winners is 5 games. That’s probably more likely.
Applying regression to the model and weighting each team’s strength based on their runs scored and allowed through their games on Sunday, and a strength of .500 for their remaining games would give Yakult an 80 percent chance of finishing with between 81 and 91 wins. But 81 wins will probably be enough in the CL this year.
Let’s throw that model against the wall and see if it sticks…
That is probably closer to reality than evaluating teams’ strength based on just 60 to 70 games. Let’s do that with the PL as well.
That seems more reasonable for a projection at this stage of the season. Perhaps to get an even better weighting, I could add extra weight with say 20 more games at .500 and include 20 games from each team’s Pythagorean win percentage from the previous season…