Expanding (also improving) upon my post from earlier in the week, and with more rosters being provided by Kyle Glaser over at Baseball America, I decided to cover the entire American League and come up with projected records based on homegrown talent.
Using Glaser’s research, I recorded my projected 2018 WAR (averaging out STEAMER and Depth Charts forecasts via FanGraphs) for each player listed in the article.
I settled on 48 wins a year for a team of replacement players; a winning percentage of precisely .294 (48/162). 48 was as close to a mean I could come to when referencing other projection sources. Glaser gives us 16 players, so we’ll scale the 16 players to the 25-man roster (40-man rosters are only available for a month).
So, 16 of 25 is 64% of a regular MLB roster. Then, I’ll take 64% of the 162 games (or 104 games) as our scale to total WAR. Not an exact science, but not a bad theoretical formula.
More on/with those figures in a moment.
Again giving credit where credit is due, Glaser put together these rosters based upon the following criteria:
“….what every team’s 2018 lineup and starting rotation would look like if it was made up solely of players drafted or signed by that team. In other words, a look at the team if it was made up entirely of homegrown players.”
“….foreign professionals signed from Japan, Cuba, South Korea or other countries are included, in addition to those drafted and signed, signed as international amateurs or signed as undrafted free agents. Players must have been active in 2017 and are scheduled to be active in 2018 to be eligible.”
Now, here’s how the theoretical analysis will work.
Say we have a roster that produces a collective WAR of 29.4. Since the replacement team in this example would win 30.5 of the 104 games (using the .294 winning percentage*104), we’ll combine that with the theoretical team’s original combined WAR of 29.4 to get a total of 60 wins (precisely 59.9) in 104 games. That winning percentage is 58% (approximately .577), so scaling that up to 162 games, the team would have 93 wins.
As you’ll see when you inspect the rosters, most teams are very close in WAR. That presented a problem in terms of record as almost every team had a winning record. Regardless of the fact that this is a pipe dream that would have occurred prior to the age of free agency, we want some level of accuracy.
So, I decided to average the WAR of each division. When I got the mean, I either added or subtracted (depending on combined WAR being above or below the mean) from the wins and reached the projected record.
Here is the result, using the prior example, of what the team record would be:
Division WAR mean (30) minus original team WAR (29.4)= 0.6 difference. Prior to this, the team was projected to win 93 games. Subtracting the 0.6 from 93 gives them a final win total of 92 (rounded down).
Got it? No? Good. As a wise man once said ‘The less you do, the more you do”.
So, using the formula I ‘created’, here is the projected final standings for the East:
Red Sox 102-60
Blue Jays 76-86
The AL Central:
The results of the AL Central:
White Sox 79-83
The AL West:
And the West pans out as so:
The playoff seeding would be: Astros (1), Red Sox (2), Indians (3), Yankees (4), Royals (5).
For the sake of reference, the average winning percentage for the American League in 2017 was 49.8%. The winning percentage with this assessment is 54.8%, which gives us a variance of 5%.
I’ll leave the rest up to your imagination. Let me know in the comments what you end up with and be on the lookout for the National League version!