Super insightful as always. My favorite read, bar none.
Not to add to your "to do" list, but I would be very interested in understanding any correlation between roster churn and winning percentage over some period of time from this methodology. What churn is healthy or expected given the data, and what churn level suggests future problems?
High roster churn would seem to be a proxy for trouble because it is only a part of effective roster construction. This includes the Draft, free agent signing, and extensions of longer termed players. Does higher roster churn suggest that those other avenues are not delivering at the levels needed? Perhaps.
It's always been interesting to me that the media seems to universally applaud the prospect "haul" in exchange for established organic talent as synonymous with effective future team building. Yet, there is precious little retrospective analysis of whether that lock step applause is actually merited over time.
Just yesterday, I read where Skip Schumaker, manager of the Marlins, gave a brutal press conference after another ineffective Sixto Sanchez start - saying he is no longer a top prospect and needs to start delivering now. I immediately recalled four years ago how many in the press gave the Marlins high marks for "stealing" the best pitching prospect in the Phillies' system for only - you know - J.T. Realmuto.
Even more specifically, I would be interested in looking at the A's and the entire "Moneyball" mythos that is so prevalent today. I suspect the A's churn rate is even higher for the period 2000 to date when the whole supposed Moneyball approach started.
Does their level of churn over the past 23 years when looking at winning percentage suggest optimization or desperation with regard to this strategy? Thanks again for adding to my daily life!
Across the entire sample, there's a very small inverse relationship (-0.111 correlation) between winning percentage and weighted average WS share with a team -- which perhaps surprisingly implies that the more churn (i.e., lower avg WS% with a franchise), the more you win... But like you said, there are many possible reasons why that might be true, and scenarios under which either a lower % or a higher one could be better depending on the team's situation. So it's tough to draw a whole lot of conclusions on aggregate.
(Maybe there is an optimal churn rate that strikes the right balance... but it's also a chicken/egg thing, where bad teams churn because they need new talent and/or sell off expensive veterans, while good teams would seem to be less likely to churn because they're going with the core that won them games.)
Super insightful as always. My favorite read, bar none.
Not to add to your "to do" list, but I would be very interested in understanding any correlation between roster churn and winning percentage over some period of time from this methodology. What churn is healthy or expected given the data, and what churn level suggests future problems?
High roster churn would seem to be a proxy for trouble because it is only a part of effective roster construction. This includes the Draft, free agent signing, and extensions of longer termed players. Does higher roster churn suggest that those other avenues are not delivering at the levels needed? Perhaps.
It's always been interesting to me that the media seems to universally applaud the prospect "haul" in exchange for established organic talent as synonymous with effective future team building. Yet, there is precious little retrospective analysis of whether that lock step applause is actually merited over time.
Just yesterday, I read where Skip Schumaker, manager of the Marlins, gave a brutal press conference after another ineffective Sixto Sanchez start - saying he is no longer a top prospect and needs to start delivering now. I immediately recalled four years ago how many in the press gave the Marlins high marks for "stealing" the best pitching prospect in the Phillies' system for only - you know - J.T. Realmuto.
Even more specifically, I would be interested in looking at the A's and the entire "Moneyball" mythos that is so prevalent today. I suspect the A's churn rate is even higher for the period 2000 to date when the whole supposed Moneyball approach started.
Does their level of churn over the past 23 years when looking at winning percentage suggest optimization or desperation with regard to this strategy? Thanks again for adding to my daily life!
Thanks for the comment!
Across the entire sample, there's a very small inverse relationship (-0.111 correlation) between winning percentage and weighted average WS share with a team -- which perhaps surprisingly implies that the more churn (i.e., lower avg WS% with a franchise), the more you win... But like you said, there are many possible reasons why that might be true, and scenarios under which either a lower % or a higher one could be better depending on the team's situation. So it's tough to draw a whole lot of conclusions on aggregate.
(Maybe there is an optimal churn rate that strikes the right balance... but it's also a chicken/egg thing, where bad teams churn because they need new talent and/or sell off expensive veterans, while good teams would seem to be less likely to churn because they're going with the core that won them games.)
Wow...thank you for this! Super interesting, and I greatly appreciate you taking this on so quickly.