Your Opinions of the Team Projections (American League)

Hello! Early last week, in part out of curiosity and in part because I had things to do, I asked you to do a bunch of voting related to the site’s Steamer team projections. I put up two posts, and here’s the one about the American League. Basically, we lean on the projections a lot, but we seldom ask for feedback, and I wanted to know how the community felt about projected team records, based on where things stood last Monday. Now, some things have changed since Monday, but nothing important has changed since Monday — excepting, say, Seth Smith — so circumstances remain mostly identical.

At the time, I promised I’d eventually review the data. Here now, the data shall be reviewed, and in this post, I’ll consider the results from the AL polls. I know it’s not perfect science. Different people voted in different team polls. Different numbers of people voted in each team poll. I probably to some extent biased the voters by offering commentary before each poll was embedded. Nevertheless, information is information, and, let’s see what we generated! With which projections do people agree the most? With which projections do people agree the least?

We’ll start with a table of raw vote data. Remember, each poll had five options. Example:

Here’s a bunch of numbers:

Team	W	Low, more than 3g	Low, up to 3g	Good	High, up to 3g	High, more than 3g	TOTAL
Angels	84	524	724	471	98	35	1852
Astros	76	74	160	662	553	227	1676
Athletics	82	231	437	595	314	162	1739
Blue Jays	84	327	620	622	164	36	1769
Indians	85	80	197	601	578	128	1584
Mariners	88	104	206	695	686	232	1923
Orioles	79	677	511	438	114	45	1785
Rangers	77	234	328	560	218	94	1434
Rays	83	33	136	475	554	291	1489
Red Sox	87	253	344	665	445	234	1941
Royals	81	140	323	731	228	51	1473
Tigers	86	54	206	700	398	77	1435
Twins	76	24	95	505	440	281	1345
White Sox	78	414	566	389	59	29	1457
Yankees	83	89	229	671	414	201	1604

That’s all best understood with percentages, which we’ll get to. But before advancing, note the last column, tracking total votes (as of late Monday night). That’s the sample size for each team. Every team’s comfortably north of a thousand, which is excellent. But, no team got more votes than the Red Sox, at nearly 2,000. Maybe not a shock; the Red Sox are immensely popular and particularly polarizing. And because they’re trying to be good again next year, there’s a lot of interest there. Meanwhile, no team got fewer votes than the Twins, at 1,345 above. The Twins are in last by a good margin. Though a fan base certainly exists, said fans might’ve been rendered a little bit numb. The Twins should be exciting soon, but “soon” isn’t as soon as 2015, and with the on-field product looking both mediocre and a bit dull, there wasn’t much voting enthusiasm, relatively speaking.

Now, percentages. I’ve also included an attempted “Average” column, where I assigned each poll option a number between 1 and 5, with 5 corresponding to “low, by more than 3 games”. This is an attempt to measure the average vote, and an average above 3 would indicate the audience thinks more highly of the team than the projection. An average below 3 would indicate the opposite, in accordance with how numbers work. The overall AL average was 3.07 — indicating inflation, which we tend to observe in fan voting — but it’s no big deal.

The table ought to be sortable!

Team	Low, more than 3g%	Low, up to 3g%	Good%	High, up to 3g%	High, more than 3g%	Avg.
Angels	28%	39%	25%	5%	2%	3.87
Astros	4%	10%	39%	33%	14%	2.58
Athletics	13%	25%	34%	18%	9%	3.15
Blue Jays	18%	35%	35%	9%	2%	3.59
Indians	5%	12%	38%	36%	8%	2.70
Mariners	5%	11%	36%	36%	12%	2.62
Orioles	38%	29%	25%	6%	3%	3.93
Rangers	16%	23%	39%	15%	7%	3.27
Rays	2%	9%	32%	37%	20%	2.37
Red Sox	13%	18%	34%	23%	12%	2.97
Royals	10%	22%	50%	15%	3%	3.19
Tigers	4%	14%	49%	28%	5%	2.83
Twins	2%	7%	38%	33%	21%	2.36
White Sox	28%	39%	27%	4%	2%	3.88
Yankees	6%	14%	42%	26%	13%	2.75

Here’s all the juicy stuff. It might be a lot to consume. But, we have to start somewhere, so, let’s start with the Orioles. The Orioles had the highest average vote in the American League. At the time they were projected for 79 wins. A good 38% of voters thought that was low by more than three games, and another 29% thought it was low by up to three games. That 38% figure — it’s the highest in the league, by ten percentage points. Just 9% of voters thought the Orioles’ projection was favorable. The Orioles are trying to be active, looking to fill out the roster, but even with the roster unfilled, most voters think the O’s are better than Steamer gives them credit for. Given what the Orioles have done lately, this result isn’t a surprise.

Right behind the Orioles, we find the White Sox, who have had an opposite sort of offseason in that while the Orioles have been quiet, the White Sox have stolen the AL spotlight. Just 6% of voters think the White Sox projection is too high, which is the lowest percentage in the league. For every one voter who determined the White Sox projection was high, more than 11 voters determined the projection was low. According to Steamer, the White Sox probably haven’t done enough to really challenge for the AL Central title. According to the fans, they’re right in the mix.

And then the Angels are almost even with the White Sox. The vote distribution is almost identical. A year ago the Angels won 98 games, and they haven’t lost all that much. Losing Howie Kendrick is a blow, but the audience doesn’t think the Angels have been too horribly weakened.

At the other end, people do not like the Twins. In fairness, I had to correct an error in their section I originally made in the post, and the error might’ve thrown off a few voters, but the mistake was both obvious and fixed in a matter of minutes, so I don’t think that’s much of a factor. More likely, the Twins just bum people out. Despite a 76-win projection, less than 2% of voters thought the projection was low by more than three games, and just 7% thought it was low by up to three games. More than half the voters thought the projection was too *positive*. No team generated a higher rate of voters who thought the projection was too high by more than three games.

Very close to the Twins, we get the Rays. Unlike the Twins, the Rays are projected to be all right, but people aren’t buying it. Maybe it’s the strongly positive Evan Longoria projection. Maybe it’s because last year the Rays lost 85 games and they don’t have David Price or Matt Joyce. 11% of voters thought the Rays’ projection was low; 57% thought it was high.

Then you get the Astros and the Mariners. Again, these are different teams in different situations — the Astros are projected to be perhaps the worst team in the league, while the Mariners are projected to be perhaps the best. In both cases, people figured the projections were too strong, perhaps influenced by what’s largely been a lack of recent success. Though the Mariners were good just last year, maybe here they’re paying the price for the Angels’ being voted better. Maybe it’s as simple as the fact that it’s hard to wrap your head around the idea that the Mariners might be super. Who could blame anyone?

Lastly, I want to provide a sense of vote distribution, in terms of how voters agreed with one another on the same teams. So here’s each team’s voting standard deviation, based on the scale introduced above:

Team	Standard deviation
Red Sox	1.19
Athletics	1.15
Rangers	1.11
Orioles	1.05
Yankees	1.03
Mariners	1.01
Astros	0.99
Rays	0.97
Indians	0.96
Blue Jays	0.96
Angels	0.95
Twins	0.95
White Sox	0.94
Royals	0.93
Tigers	0.87

Voters were in the least agreement with the Red Sox, and in the most agreement with the Tigers. The Red Sox were the only AL team to get at least 10% votes for the projection being low by more than three games, and for the projection being high by more than three games. The Tigers drew just 4% and 5% votes, respectively, and about half the voters decided the Tigers’ projection was right on. Close to the Red Sox, we find the A’s and the Rangers. It’s hard to know what to make of the A’s right now, because of all the turnover. It’s hard to know what to make of the Rangers right now, because of all the injured or recently-uninjured players. Close to the Tigers, we find more of the AL Central. Relatively speaking, fans are somewhat united in how they think about that division. It’s not that they do or do not agree with the projections; it’s that they agree with one another.

You could think of this, perhaps, as representing the error bars around team projections. This is a proxy of uncertainty, or maybe a direct measure of uncertainty, where people consider the Red Sox a lot more unpredictable than the Tigers. I don’t know to what extent that’s true, but I figure the fans have to be on to something. And I think that example makes rational sense. The Red Sox look a lot more volatile. The Tigers are a mostly stable team.

Up soon, the National League results. Thank you to everyone who participated in the voting, and thank you to everyone who didn’t participate in the voting, because you’re all great.