Atlanta Newspaper Misses Reason for Braves' 2007 Struggles

Teams and fans love to focus on records in one-run games. The conventional wisdom is that a good team "knows how to win the close ones." Sounds great in theory but Bill James proved two decades ago that this is simply not true. A team's record in one-run games is mostly (not entirely) a matter of luck. But it's hard to get people to give up on deeply-ingrained thoughts, no matter how wrong they may be.

In today's issue of The Atlanta Journal-Constitution there was an article published lamenting the Braves' season. The headline was: "The difference for Braves: close games" while the sub-head read: "Record in 1-run games a sad 16-23". But, it's worse than that - their record is actually 16-24

Now, the Braves overall record this season is 74-72 and they are 9.5 games behind the division-leading Mets. Even if you reversed their record and gave the Braves an impressive 24-16 record in one-run ballgames, they still would not be in first place. No, Braves fans - I'm afraid the problem runs deeper than bad luck in close games.

A much better indicator of a team's strength is its record in blowouts. Simply put, a good team should throttle lesser squads much more often than the rare day when it plays bad. Let's take a look at the Braves and the teams with the eight best records in the Majors to see how they've played in both one-run games and in games decided by five runs or more, respectively.

ATL - 16-24, 22-20
BOS - 22-22, 33-15
ANA - 21-16, 25-17
CLV - 25-21, 24-14
NYM - 27-15, 21-11
NYY - 13-17, 43-18
ARI - 32-18, 16-24
DET - 25-17, 22-21
SDP - 22-23, 25-21

The team with the best record in baseball is .500 in one-run games while two of the remaining seven-best teams have a record below .500 in those contests. There's only one team, the Arizona Diamondbacks, who owe their good overall record to their mark in one-run games.

So, the Braves have been a bit unlucky in one-run games this season, but not to any terrible degree. More to the point, the Braves' record in blowout games tracks very nicely with their record overall. The 2007 Braves have been an unlucky squad, but the bottom line is they are simply not a playoff-caliber team.

Braves fans reading this will immediately point to the Mets' one-run record. And the Mets record is quite good. But also check out the Mets' record in blowout games. Their .643 winning percentage in blowouts dwarfs the .524 mark put up by the Braves and gives an indication of the talent gap between the two franchises.

Atlanta fans also like to point out how similar is the run differential between the Mets and Braves. The Mets have scored 706 runs while they've allowed 635, for a difference of 71 runs. Meanwhile, the Braves have scored 735 runs while allowing 674 for a difference of 61 runs.

Many years ago, Bill James (there's that name again!) came up with a formula to predict a team's record by how many runs it scored and gave up. He named it the Pythagorean Theorem, in honor of the Greek mathematician who allegedly came up with the Euclidian geometry relationship among the three sides in a right triangle. James' formula for predicting winning percentage was:

Winning % = runs scored (squared) divided by (runs scored squared + runs allowed squared)

The Braves' "Pythagoras" record this year is 79-67 or 4.5 games better than their actual record. Meanwhile, the Mets' "Pythagoras" record this year is 80-65 or three games worse than their actual record.

James showed that most teams played within two games of their "Pythagoras" record. The reason for this is that over a large sample size like 162 games, the distribution of runs scored should be relatively "normal", which allows for predictable relationships between runs and runs allowed. If we tried using "Pythagoras" after just a few games, the records would not be so accurate due to random fluctuation in run scoring patterns.

So, "Pythagoras" shows the Mets and Braves being almost equal teams this season. But if we look further inside the numbers, we see that the Braves had two standout pitchers, Tim Hudson and John Smoltz, who had run support at the wrong times. The two combined for 18 blowout wins in which the Braves outscored their opponents, 157-32. Hudson and Smoltz (and relievers) gave up two runs or less in 13 of those 18 games. Meanwhile, the aces are just 5-12 in one-run games.

Ten times, the Braves scored eight or more runs in a start by one of their aces. In games when they scored less than that, Hudson and Smoltz were still 18-15 overall. Contrast that with Kyle Davies. He had three starts with eight or more runs (all wins). In his other 14 starts he was 1-8.

It wasn't just Davies. Check out Chuck James, who was 8-0 when the Braves scored eight or more runs but just 2-10 otherwise. Buddy Carlyle came out of nowhere to give the Braves a boost when recalled in late May. But things have been different once the league got a look at him. In his last six starts he's received 38 runs and has gone just 1-3. The Braves' hitters came alive for the pitchers who needed this support the least.

Undoubtedly, the Braves suffered bad luck this season, both in their one-run record and their distribution of runs. But the larger problem was having three pitchers in the starting rotation that needed eight or more runs on a regular basis to record a win. That should be what the Atlanta sportswriters focus on when writing the 2007 epitaph for the Braves.