Roster Turnover - So What?

A look at year-to-year Cardinals roster changes with a conclusion unexpected by some.

Just like clockwork, every year at this time, "experts" write seemingly dozens of cookie cutter-like stories superficially recapping the off-seasons of all 30 teams.

These reports note the inevitable player turnover on key teams while relying upon vague observations to predict disastrous drop offs in team performance.

On the other side of the coin, they often overvalue teams that have added stars from other organizations, despite the players having not even met each other yet. It is far too easy to assume that nifty lineup cards with big names are shoo-ins to win the World Series (hello, Mets!) and also far too early for fans to remember at the end of the season how far off these predictions actually were.

While far from the most blatant example of this phenomenon, here is a quote from one of the most recent – from Sports Illustrated's John Donovan on February 15.


"The defending-champion Cardinals are favored to repeat, but they'll have to do it with a vastly different roster. The Cards lost a ton of players to trades, free agency or retirement over the winter…"


Let's see if we can define that highly technical nomenclature, and figure out what "a ton" of players mean. First, I checked my sabermetric encyclopedia, but did not identify this term, so I guess I am on my own.


In this small, but factually-based study, we'll look at overall roster turnover on the Cardinals each season, comparing opening day squads since 2000 while comparing that to the change in their won-loss record each season.


Granted, this is a simplistic view in that it looks at quantity, not quality, but at least there are some numbers involved. To look at front-line players, I also ran the numbers excluding the bench and bullpen.


In other words, I did the same comparison, looking at just the starting eight position players and the five men in the starting pitching rotation, testing the premise that turnover in the front lines might be the riskiest kind.

For the actual rosters from 2000 through 2005, refer to my story, "Balancing the Bullpen & the Bench – Part One". For 2006, I used my "Projected Roster B" from "Balancing the Bullpen & the Bench – Part Two".

No matter which projected roster is used, most agree the Cardinals will have four new starters in 2006 compared to opening day 2005. They are the fifth starting pitcher, Sidney Ponson (or Anthony Reyes), second baseman Junior Spivey and outfielders Juan Encarnacion and Larry Bigbie (or So Taguchi).


Looking at the entire projected 25-man roster for 2006, it looks like almost half, 12, will be different from the players who began the previous season on the Cardinals' team.


That does look onerous, doesn't it, Mr. Donovan?


Well, not so fast. The facts tell us something else entirely. Turns out that prior to the last two seasons, when during each the Cardinals led the majors in with 100 wins or more, they turned over even more front-line players than this off-season.


Specifically, the Cards had five new starters in 2005 and six new ones in 2004. In total, while "only" 11 players were new Cardinals in 2005, a full 3/5 of the roster, or 15 players who donned the Birds on the Bat last April were not around the previous year.


Another way to look at it is this. The 2006 changes of four starters and 12 players overall is also exactly the same as the Cardinals' average year since 2000.


For reference, I included the number of team wins each regular season and the year-to-year difference in wins. However, as far as I can tell, at least with this small sample size, there is absolutely no correlation between roster turnover and won-loss record, whether considering the entire roster or just front-line players only.


Going back a few years further, while there was slightly less roster turnover for 2001, 2002 and 2003, those teams didn't do as well in regular season wins.


Three simple points in conclusion

1)     The amount of change in the Cardinals roster between last season and this is typical.

2)     Any impact of roster turnover on what eventually happens on the field is entirely unknown.

3)     Anyone suggesting anything else without data to back them up is just plain uninformed.

Number of different opening day players  Starters* 25-man

Total wins

YTY wins

2006 projected (vs. 2005) 4 12



2005 (vs. 2004) 5 11 100 -5
2004 (vs. 2003) 6 15 105 20
2003 (vs. 2002) 3 11 85 -12
2002 (vs. 2001) 3 10 97 4
2001 (vs. 2000) 3 11 93 -2
2000 95
Six-year average



* Rotation plus eight position players

Brian Walton can be reached via email at

© 2006 All rights reserved. This material may not be published, rewritten or redistributed.

The Cardinal Nation Top Stories