Here Is What You Think of Our Team Projections
Now to analyze my favorite polling project every year. We’re a site known for its math and projections. Most of the time, we just do the math and supply the projections, and that’s the end of it. Less attention is given to community feedback. In this project, you, the community, fed back!
This is a link to our projected-standings page, which tries to forecast the 2017 regular season based on Steamer, ZiPS, and our manually-maintained team depth charts. You see projected records there, but that doesn’t mean you have to like them or agree with them. On Tuesday, I asked you all to weigh in on the American League projections. On Wednesday, I asked you all to weigh in on the National League projections. I don’t know why I just linked to those posts, because there’s no point in voting anymore, because I’m already analyzing the results right here. Two things I can tell you right off the bat: The community is high on the Rockies, and down on the A’s.
Every team had a poll, and every poll had five options. I asked whether you thought each team’s projection was way high, high, good, low, or way low. I put all the voting results into a spreadsheet and I converted the answers into numbers in order to calculate an overall average. That is, an average win adjustment to the team projection, as determined by the broader FanGraphs community. In this plot, every baseball team, with the teams on the left being considered underrated by the projections, and with the teams on the right not being considered that.
When I examined last year’s results, the community had its strongest response to the Royals projection. You all thought the Royals were being heavily underrated, which wasn’t too much of a surprise, given what had been the Royals’ recent history. Here, we see the Royals again to the left — collectively, you think their projection is low by about a win and a half. You still don’t see them as being good. And the magnitude of their community response has nothing on the Rockies, who you think are being underrated by about a win more than anyone else.
Our Rockies projection puts them at 78-84. Our community-adjusted Rockies projection would put them at 81-81. That’s still obviously not great, but it hints at the community seeing them as a wild-card contender. On the one hand, we have recently featured some pro-Rockies content. Dave talked about them as a potential contender at the end of January. I’ve discussed their chance at having their best-ever pitch-framing catchers. But then, we’ve also been critical of, say, the Ian Desmond contract. They’ve been fairly inefficient with their resources, which has to some degree held them back. But the Rockies think they can win in 2017. You are at least kind of on board.
After the Rockies, you believe the Brewers are the next-most underrated team. I wish I could say it mattered, but you still don’t have them sniffing the race. That being said, the Brewers don’t expect to be near the race, and what’s suggested here is you see a bit more realistic upside in the present roster. (Keon Broxton?) (Keon Broxton) (KEONNNNN BROXTON)
Flipping all the way to the other side, a pair of AL West rivals brings up the rear. You don’t really see the A’s as being that close to a .500 team, maybe because you aren’t so interested in their pitching staff. The group of position players is clearly limited and fairly light on defensive capability. And then, you think the Angels are more like a .500 team than anything better. I think you’ve been expressing this for some time, ever since Steamer first displayed its modest fondness for the club. There’s a reluctance to believe this year’s Angels are legitimate contenders, and given the number of question marks surrounding the pitching staff, I do get it. Health is a major concern, and there’s some chance the bullpen could be among the league’s worst. This at a time when other contenders are stuffing their bullpens five or six deep with unhittable heat.
To show this all a different way, here are standings tables, beginning with the American League. On the left, our present projections. On the right, the community-adjusted projections.
AL EAST | Projection | AL EAST | Community |
---|---|---|---|
Red Sox | 93 | Red Sox | 93 |
Blue Jays | 86 | Blue Jays | 86 |
Rays | 82 | Yankees | 82 |
Orioles | 81 | Orioles | 81 |
Yankees | 81 | Rays | 80 |
AL CENTRAL | Projection | AL CENTRAL | Community |
Indians | 92 | Indians | 93 |
Tigers | 81 | Tigers | 82 |
Royals | 75 | Royals | 77 |
Twins | 74 | Twins | 73 |
White Sox | 69 | White Sox | 69 |
AL WEST | Projection | AL WEST | Community |
Astros | 91 | Astros | 90 |
Angels | 83 | Mariners | 84 |
Mariners | 83 | Rangers | 84 |
Rangers | 83 | Angels | 81 |
Athletics | 79 | Athletics | 76 |
There’s no disagreement about any of the division winners. The Red Sox are still projected to finish well ahead of the Blue Jays. The Indians are still projected to run away with the Central. And although the West looks a little tighter, it’s still the Astros by a healthy margin. Yet there’s a change to the crowded wild-card picture. Just based on the regular projections, the Angels would sneak into the second wild card by a hair. Fold in the community response, and the Angels are replaced by the Mariners, who just barely edge out the Rangers. No matter what, that wild-card spot looks to be hotly contested. After the Blue Jays, there are six teams all between 81 and 84 wins. This is how the league can compensate for what could turn out to be some pretty lame division races.
Switching sides:
NL EAST | Projection | NL EAST | Community |
---|---|---|---|
Nationals | 91 | Nationals | 91 |
Mets | 85 | Mets | 86 |
Marlins | 78 | Marlins | 76 |
Braves | 73 | Braves | 74 |
Phillies | 71 | Phillies | 72 |
NL CENTRAL | Projection | NL CENTRAL | Community |
Cubs | 95 | Cubs | 96 |
Cardinals | 83 | Cardinals | 84 |
Pirates | 82 | Pirates | 83 |
Reds | 70 | Brewers | 71 |
Brewers | 69 | Reds | 69 |
NL WEST | Projection | NL WEST | Community |
Dodgers | 94 | Dodgers | 94 |
Giants | 87 | Giants | 87 |
Rockies | 78 | Rockies | 81 |
Diamondbacks | 76 | Diamondbacks | 76 |
Padres | 65 | Padres | 64 |
All the same division winners. And, actually, all the same playoff teams, who would be the same playoff teams as a year ago. That’s a fairly sensible outcome to expect, especially with the Cardinals already having lost Alex Reyes. If you’re looking for intrigue, the community has the Mets at 86 wins, but the Cardinals, Pirates, and Rockies all within five of that. One can observe that the community isn’t so convinced by the idea of the Marlins finding upside in unconventional pitching-staff usage. Going to need some proof, I suppose. I’d put myself in the same boat.
Fun fact: The best projected team — the Cubs — is considered a little underrated. Similar fun fact: The worst projected team — the Padres — is considered a little overrated. I don’t have anything to add to that. The Cubs team poll got far more votes than any other team in the National League. The Reds team poll got the fewest. The Angels poll was the first one in the American League post, so that participation led the way, with the White Sox in last.
To close out, here’s a giant and sortable table, to sate your own potential curiosity. I’ve highlighted for each team the response that got the highest number of votes.
Team | High by more than 4 | High by 2 – 4 | Good, +/- 1 | Low by 2-4 | Low by more than 4 |
---|---|---|---|---|---|
Angels | 25% | 45% | 20% | 7% | 3% |
Astros | 4% | 28% | 50% | 15% | 2% |
Athletics | 27% | 46% | 18% | 7% | 2% |
Blue Jays | 4% | 27% | 50% | 16% | 3% |
Braves | 3% | 17% | 38% | 34% | 8% |
Brewers | 1% | 7% | 32% | 42% | 18% |
Cardinals | 4% | 19% | 37% | 33% | 8% |
Cubs | 3% | 15% | 46% | 27% | 10% |
Diamondbacks | 5% | 23% | 35% | 29% | 8% |
Dodgers | 3% | 24% | 53% | 17% | 2% |
Giants | 4% | 23% | 46% | 24% | 4% |
Indians | 2% | 14% | 47% | 33% | 4% |
Mariners | 3% | 13% | 34% | 39% | 12% |
Marlins | 11% | 47% | 29% | 10% | 2% |
Mets | 2% | 13% | 33% | 37% | 14% |
Nationals | 4% | 27% | 47% | 18% | 3% |
Orioles | 7% | 24% | 38% | 23% | 8% |
Padres | 9% | 25% | 46% | 15% | 4% |
Phillies | 2% | 11% | 39% | 39% | 9% |
Pirates | 2% | 13% | 33% | 43% | 9% |
Rangers | 4% | 16% | 37% | 32% | 10% |
Rays | 15% | 40% | 31% | 12% | 2% |
Red Sox | 5% | 16% | 51% | 24% | 4% |
Reds | 9% | 36% | 40% | 10% | 4% |
Rockies | 1% | 5% | 21% | 43% | 30% |
Royals | 3% | 8% | 39% | 35% | 16% |
Tigers | 3% | 14% | 31% | 40% | 13% |
Twins | 9% | 25% | 45% | 17% | 4% |
White Sox | 7% | 21% | 50% | 18% | 4% |
Yankees | 4% | 12% | 29% | 43% | 11% |
For 19 of the 30 teams, the most popular response was the middle one, expressing general agreement with the projections. For seven teams, the most popular response was that the projection is a little low. For the remaining four teams, the most popular response was that the projection is a little high. You can detect the faint scent of optimism — on average, 34% of people found the projections low, while 28% of people found the projections high. The mean community adjustment for each team is +0.2 wins, which of course couldn’t happen. Overall, though, the effect is minor. The smallest amount of people found the A’s projection to be low (9%). The Rockies, naturally, were on the other side of things (73%). I’ll note that the lowest response in any category belongs to the Brewers — 1.1% of people found their projection to be way too high. The Rockies were next, at 1.3% saying the same.
And, at last, the single most popular response: 53% of Dodgers voters expressed agreement with the projection. Okay. Thank you one more time for all of your participation. I love running these projections, and they wouldn’t go anywhere without you.
Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.
Did the fan adjusted projections do better or worse than the non-adjusted projections last year?
Very slightly better. Virtually identical R^2 figures. Projections had an average error of 5.6 wins, and a median error of 4.6. Community-adjusted projections had an average error of 5.5 wins, and a median error of 4.2. The strongest community response involved the Royals, and the community adjustment was better. Next was the Blue Jays, and the community adjustment was better. But then the community was further off for the Pirates, Red Sox, and Indians. So, ultimately, a small sample of one year, but nothing wrong with analyzing the analysis!