Softball America Magazine-Year In Review / 2018

General League Announcements and Discussion
Post Reply
User avatar
Posts: 1794
Joined: Fri Nov 17, 2017 7:56 am

Softball America Magazine-Year In Review / 2018

Post by sixofdiamonds » Thu Jan 17, 2019 6:55 am

The Softball America - Year in Review magazine is now available online. The magazine was a team effort consisting of myself, Bill Shipe, Curtis Hensman and Keith Keppley. Collectively, we really hope you enjoy reading it.

You can access the magazine here... ... e.2018.pdf

Thank you,
Les Greenstein

Double A Poster
Posts: 37
Joined: Wed Nov 29, 2017 7:47 am

New All-Star Formula

Post by wdshipe » Fri Jan 18, 2019 2:36 pm

Congratulations to all of the award winners!

Les –
Thank you as always for compiling the magazine – I look forward to its publication every year. There are many points of interest in there, but for now I just wanted to comment on this one:

“All-star status is determined by statistical formula based on offensive production only. The formula has always taken into account R%, RBI%, Avg, Slg, BB%. This year, however, we will be trying out a new all-star formula. The new formula will be based off of RUNC (runs created) divided by G (games played). A minimum of 36 at-bats (including walks) during any softball year is needed to qualify for all-star list consideration.”

I’ll be very interested to see to what extent the change to the all-star formula affects the all star lists. I’ve loved the RUNC statistic ever since you rolled it out several years ago, and given its ability to pretty well reflect a team’s aggregate run production, I’m a strong believer that it meaningfully estimates the number of runs that a hitter individually contributes to their team as well.

We know that RUNC = (TB + 0.26BB + 0.52SACF)(OBP). Of course, nowhere in this formula are any terms having to do with R or RBI, and I view this as a positive for getting a clearer picture of a player’s individual performance. R and RBI are dependent on the strength of the lineup around a given player and/or the player’s position in the lineup (R is influenced by who bats after the player; RBI is influenced by who bats before the player). Employing a formula without these terms has the effect of increasing the likelihood of all-star status assignment to players who individually excel on weak teams, and decreasing the likelihood for solid players who bat in proximity to “true all-stars”. Another wrinkle to this is that there may have been situations where a team had only 1 or 2 “true all-stars”, but the team as a whole excelled (scored lots of runs) owing to team chemistry and/or everyone knowing their role and performing it. You wouldn’t want to penalize this by tagging 8 or 9 of the group as all-stars. I understand that the change for the league here was largely motivated by those situations where players were becoming all-stars as a result of playing on a high-scoring team. I’m all for trying it out – let’s see how it works!

As I’ve played around with league statistics over the last few years (e.g., benchmarking my performance from season to season, or comparing one player to another), I’ve usually looked at RUNC/APP. This is quite similar to the new formula RUNC/G. The difference between the two approaches is that those players on strong teams will have more plate appearances (APP) per game (G). Accordingly, using RUNC/G (rather than RUNC/APP) will temper the desired effect for those players on strong teams (i.e., who come to the plate more often). I think that the more moderate RUNC/G approach may be the right place to start, but if you find that it doesn’t go far enough, perhaps you could consider looking at RUNC/APP instead.

Post Reply

Return to “Administrative”