Mark Bradley

Mark Bradley is a sports columnist and blogger for The Atlanta Journal-Constitution

A good young team (like the Braves) is hard to project. Here’s why


Just when we were growing accustomed to the Braves working a wonder every day, they came home and got swept by San Francisco, which as we know invalidates everything. 

Wait. Sorry. What I meant to say: It invalidates nothing.

This is baseball. The worst team in the majors, which the Giants aren’t, can take three in a row from the best, which the Braves aren’t yet. At last check, the Braves are still in first place, though the National League East has gotten bunched. The Nationals are in fourth place. Two days from now, they could be in first. It’s a bit early to be figuring magic numbers. 

Speaking of numbers: Neil Paine of FiveThirtyEight affirmed last week what had been mentioned in this space – that statistical projections for young teams can lag reality. Example: FanGraphs has the Braves going 61-68 from here on, which would put them at 80-82, which is what I predicted before the season began. I wouldn’t, however, pick them to finish below .500 now. I thought they’d be better than they’ve been; they’ve been better than I thought. 

Writes Paine: “It takes about 70 games before observed results from a season in progress reach even a 50-50 balance with preseason expectations, in terms of how much weight each deserves when assessing a team. The Braves have played less than half that many games, which probably goes a long way toward explaining why the statistical projections haven’t budged much off of Atlanta’s relatively bearish spring-training predictions. The past data says you can’t read too much into a month’s worth of results.” 

Then: “That premise was designed to hold true for all teams as a group. What happens when we look at a smaller group of teams, especially just the ones that have as much breakout potential as the Braves?” 

Dating to 1984, Paine found 25 teams that had played under .450 ball the year before but had gone .600 or better through April – and had entered the season with one of the 10 youngest teams in MLB and were possessed of one of the 10 highest-ranked farm systems by Baseball America. Those teams averaged 2.7 more wins than FiveThirtyEight’s Elo would have projected. 

Paine’s conclusion: “If it holds true for the Braves, it would imply that they’re due to win more than expected based on their pessimistic win projections at FanGraphs and in our Elo interactive — and those extra wins could be enough to elevate them from a mid-80s win tally (sketchy territory, playoffs-wise) to a number closer to 90 wins.”

In a post for ESPN, Bradford Doolittle notes that his preseason projections had the Braves winning 73.2 games. Updated projections have them winning 83. As Paine mentioned, that’s not quite playoff-caliber – but it’s not far away. The Rockies took the second wild card last year at 87-75. 

Baseball Prospectus gives the Braves a 23.7 percent chance of making the playoffs. That’s up from the 14.6 percent of Opening Day, but it still sounds low for a division leader. That’s because, as much as projection systems try to account for everything, there are some things they can’t see coming. A young team getting good before our eyes is such a thing.


Reader Comments ...


About the Author

Mark Bradley is a sports columnist and blogger for The Atlanta Journal-Constitution. He has been with the AJC since 1984.