My sermon for today takes its text from the Book of Braves14, Comment 2384135:
I wasn’t impressed at all with the .500 Cardinals in the Central and the Braves seemed to be a vastly superior team in both series we played them. And the Braves seemed pretty even with the Brewers in that series. I wonder if the NL East records are partially because the teams are beating up on each other in division play.
The NL West is obviously the best division by far in the NL.
So the season is now far enough along to pull out my favorite team ranking toy: the Bradley-Terry Ratings. I first did Bradley-Terry rankings here almost exactly eight years ago. If you’re interested in early August 2013 rankings, or the math behind the rankings, you can go to that link.
A few caveats. These rankings are the simplest ones: (a) unadjusted for home field advantage; (b) with one blended rating for offense and defense rather than separately estimated offenses and defenses; (3) treating the team as if it is the same team the whole season, unadjusted for injuries or trades; and (4) renormalized for no particular reason from the old method. (Instead of the best team getting a ranking of 100, the sum of all the rankings is 100. This makes no difference at all. I just forgot how I used to do it.)
Given the caveats, these are not intended to be predictive. Instead, their main purpose is to allow you to easily calculate wins and losses when the strength of schedule is equalized. So it helps answer questions like:
- Are the Padres, Giants and Dodgers really that good? How good would they be if they didn’t get to play Arizona a lot? Conversely, how well would Arizona do in some other division? Are they as bad as their record, or are they just unlucky to be playing the Padres, Giants and Dodgers almost 60 times a year?
- What’s the best division? What’s the crappiest? You would know the answer to that question if everyone played the same balanced schedule. It’s much harder to tell when the schedules are unbalanced.
- If you want to be really expansive, you can swap teams across divisions and rejigger their expected wins just from the schedule differences. Does it make a difference that the Mets play extra games with the Yankees (“natural rivals”) while the gNats get to Baltimore?
The new column is BSP: Balance Schedule Percentage. By comparing the winning percentage of a team, or a division, with what their percentage would be under a balanced schedule (in which they played everyone other than themselves a fixed number of times) you get an idea of how much the unbalanced schedule helps them or hurts them, and it gives you a fair way to judge divisions based on their performance to date.
So here are the numbers:
Armed with these values, we can answer Braves14 question, and a few others besides.
- When the Braves play the Cardinals, the Braves ought to win about 53% of the time. Against Milwaukee, 44% of the time.
- The NL East and NL Central are essentially tied for crappiest division.
- Under a balanced schedule, only the AL East and AL West would be over 0.500. The NL West would be essentially a 0.500 Division.
- The Braves, in a completely balanced league would be an 80 win team. That is not the most mediocre team, but it’s close. The Guardians are slightly more mediocre.
- Unbalanced schedules make differences at the margins, enough to tip close Wild Card or even pennant races. The Brewers have the largest assist from an unbalanced schedule: about 7 games when projected to 162. But since they’re overwhelmingly likely to win the division, it doesn’t matter much. If they were a Wild Card contender, though, it would be borderline unfair. As you’d expect, the Rangers and Orioles are badly hurt by playing unbalanced schedules, but the differences wouldn’t be enough to get them anywhere a Wild Card.
In the attempt to complete the sweep, Kyle Muller forgot to bring his broom and his control. The big blow was a Large Salami from Jesse Winker. Chip didn’t mention it, so I will: when you come out with two on, one out in the third trailing by three you rarely qualify for the win. Furthermore, if you are replaced by Josh Tomlin, it would probably behoove you not to leave men on base. The two men he left on scored, and Tomlin gave up a few of his own and it was 9-1 with one out in the top of the third. At this point, if I were Snit, I would simply pitch Tomlin until his arm fell off. That would save the important parts of the bullpen as well as getting rid of Tomlin. So far this year the Braves have DFA’d (or effectively DFA’d) La Gran Calabaza, Tyler Flowers, and Jonathan Lucroy (all 35 years old). Josh Tomlin is 36 and not producing. The lightning has all leaked from the bottle. Letting guys hang on for one last hurrah is something done by terrible teams and juggernauts. Teams in pennant races have no room for such things. If all goes well, Ian and Huascar can be slotted into Kyle and Josh’s spots. Something else might happen of course, but this seems pretty obvious.
They took Tomlin out after 67 pitches and he left down 11-1. I was watching when he left, and the arm was apparently still attached to his shoulder. One can dream.
I suppose that some miracle was still possible, but the one-miracle-per-homestand rule was invoked by the Reds and it was over. But look at it this way: After the first day of the season the Braves were one game behind Philadelphia for first place. We are still one game behind Philadelphia for first place. If all you care about is the playoffs, you could have gone into a 4 1/2 month coma and not missed a thing. Me? I like watching baseball.
[Both of these were from Wednesday night… I didn’t pay that much attention this afternoon, and most of what I did hear were repeats of previous pet peeves, e.g. “Huge Reds fans after today”]
“Were it not for Pablo Sandoval and Adrianza hitting those pinch hit home runs in the first five [or] six weeks one can only imagine where the Braves would have been recordwise this year.”
Pablo Sandoval hit pinch hit homers in four games. The Braves were 3-1 in those games. We’ll never know what the person who would’ve batted had we not had Pablo Sandoval would have done, but the Braves’ record is surely no more than two games better for those homers. Meanwhile, his aggregate WPA for the season was 0.796 suggesting that he made the Braves’ record no better than 1 game better, which seems about right.
Adrianza hit home runs in games the Braves won 8-1, 13-4, 20-1 and 20-2. That pretty much defines useless in the HR category. His WPA is -0.187, which includes his bases-loaded walk against the Cardinals in Walkapalooza worth 0.238 and his -.142 strikeout last night with the winning run on 3rd.
So yeah, Chip, I can imagine. Maybe a game or two back?
“That walk in the third inning tied [Votto] with Ty Cobb; imagine being tied with Ty Cobb in anything!”
First off, I’m confused. According to Stathead, going into tonight Votto had 1260 walks and Ty Cobb had 1249. (Maybe they’re including playoff appearances? Nope. Votto has 5 postseason walks and Cobb had 3.) 1260 is a lot of walks, good enough for 53rd alltime. But it’s still less than say, Adam Dunn, who will never be in the HOF. Votto trails Eddie Yost by over 300 walks. Eddie ain’t in either.
But more importantly, being better than Ty Cobb at any one thing is no challenge at all. Heck, I know a lot more about statistical analysis than Ty Cobb did. Even turning to baseball, Ty Cobb hit 117 homers. 792 players have hit 117 homers or more. Nobody is known for everything they did. His walk total is not that outstanding. Carl Yastrzemski, Rickey Henderson, and Hank Aaron all had many more walks in about the same number of plate appearances. Ty Cobb generated 151.5 WAR, but his skill at drawing walks just wasn’t a big part of that. Joey Votto has generated 62.6 WAR. He may possibly go into the Hall of Fame. But it won’t be because someone marvelled at his ability to get more walks than Ty Cobb…. unless they give Chip a vote.