Has W.A.R. been used in this fashion?

Question

Whether justifiably or not, in modern baseball conversations, W.A.R. (wins above replacement) has become the goto statistic to qualify who's the best in the game.

My question is, can it be applied to teams? And has it?

What I mean is: say we figured out the record of an all replacement level squad - it seems like it would vary based upon the year, but for ease of this question, let's say 62 wins, 100 losses.

Maybe it should be less, but whatever the case, what we're talking about here is the win-loss record for a team with a combined W.A.R. of 0.0.

Now - if we took at a team with 92 wins and 70 losses, how close would their record come to their combined team W.A.R.? In other words, could we expect them to have a combined team W.A.R. of 30? If not; why not?

Is there any sort of statistical interest in combined team W.A.R., and / or has any sort of analysis been done to see how closely team W.A.R. correlates to actual Wins and Losses?

And if it doesn't correlate in that manner, shouldn't the name be changed?

Answer 1

There has been analysis done to look at the correlation between individual WAR and the record of teams. Most entities that calculate WAR make the same assumption that a team full of replacement level players would win about 52 games in a season (.320 Win %).

Below are 2 links that show high correlation when studying the relationship between WAR and team records. Hardball is more in depth and (mostly) impartial. FanGraphs is more of a summary and could be less impartial.

Hardball Times WAR Study

FanGraphs WAR Study