You might think that average isn't very good, but when it comes to MLB players, that isn't the case. I set out to find out just how many below-average players a playoff team is normally plays during a season. I wasn't quite prepared for what I found.
The stats for this study all come from Baseball Reference, from team pages like this one. The tables are sortable by a variety of stats, including WAR - Wins Above Replacement, and WAA - Wins Above Average. We could debate the accuracy of these composite metrics, but give us a reasonable sense of thing. As someone once said, All models are wrong, but some are useful."
By the very definition of "average," there must be players worse than average who are playing a significant number of innings ever year, but perhaps winning teams don't rely on those below-average players very much. Maybe those players are all concentrated on the sub-.500 teams. How many weak links can you use when building a World Series winner?
Here are the below-average position players used by the Tribe in 2013:
*F-T Equiv (full-time equivalent): This represents what piece of a full season (set at 680 PA) each player played
Plenty of familiar names, along with one guy I definitely don't remember (Santos). Right away you can see that the Indians gave a lot of plate appearances to below-average players. Of particular note is Asdrubal Cabrera, who is credited with -0.8 WAA but still managed 1.2 Wins Above Replacement. Asdrubal had a down year, but nothing like Mark Reynolds, who went on to be three-times as below-average as Asdrubal last season, in 37 fewer games.
As it turns out, such totals are not at all unusual for playoff teams:
Here we see that every playoff team gave more than 1,000 plate appearances (and whatever accompanying defensive innings) to below-average players last year. Most teams used below-average guys at twice that rate. The average playoff team was using below-average position players for 3 spots in the lineup each game. The average value of upgrading all those player to league average would be 7.22 wins for the season. That's more value than you would get by replacing an average outfielder with Mike Trout for a year (he was 7.0 Wins Above Average in 2013). Seriously, Mike Trout!
If we expand our search to include the last five seasons of MLB playoff teams:
|Playoff teams '09-13||12.3||2290||3.4||-7.62||-0|
|World Series teams '09-13||12.1||2315||3.4||-7.52||0.23|
|World Series winners '09-13||10.8||1754||2.6||-5.42||0.22|
Even the average World Series winner gave a tremendous amount of playing to below-average players. Upgrading all of them to average is worth more than swapping an average second-baseman out for Robinson Cano. And notice the difference between the average playoff team and the average World Series winner- not only are they giving less playing time to below-average players, but the below-average guys they have are less-below-average than usual.
Turns out, you can build not only a contending team, but a championship team with below-average players. Superstars are not the only difference between winning it all and being an also-ran. A team that can avoid giving away playing time to below-average players will have a big leg up on the competition.