Posted by Jeff Clark of CelticsBlog.com
We are fast approaching the prediction season. That means beat writers, columnists, and TV personalities around the country are shaking the sand out their loafers and hanging up the Hawaiian shirts (except in Chad Ford's case where presumably he wears those all year long) and starting off the year with some good old-fashioned prognostication.
I tend to look at these predictions with a cynical attitude. They never seem to "know my team as well as I think I do" (translation: I disagree, so obviously they know nothing). The columns are churned out, people read them, agree or disagree, then move on to the next set of previews and eventually forget all about them. Occasionally you'll see someone say "nobody thought we had a chance, but look at us now."
So I wanted to take a look back to last year's predictions. This is ESPN (the Worldwide Leader), so I figured I'd pick on them. Just how well did the experts predict the season?
For this exercise I went with the easiest numerical indicators I could find. Each ESPN expert was asked to rank where they thought each team was going to end up in the standings. Their rankings were averaged together and posted at the bottom of their individual comments. This is not unlike technology forecasting methods.
For example, my Celtics were not expected to be that great. The average ranking for them was 9.7 out of 15 teams in the East. For the record, I think I would have put them at 7 or 8 if you had asked me a year ago. They ended up getting off to a slow start before getting hammered with injuries to half the roster and the team just imploded for the rest of the year. Few could have predicted just how badly they would do. They ended up 15th in the East so the differential between predicted rank and actual rank was -5.3 spots.
Here's the thing, that example is one of the few outliers. Here is a chart of all the teams. The first number (before the city) is their actual final ranking (keep in mind this is regular season finish, not playoff seedings). The number after the city it is the differential.
Eastern Conference
1 Detroit 0.8
2 Cleveland 0.7
3 Chicago 1.1
4 Toronto 6.9
5 Miami -2.5
6 New Jersey -1.7
7 Washington -0.1
8 Orlando 0.3
9 Indiana -2.1
10 Philadelphia 1.9
11 Charlotte 2.7
12 New York 0.2
13 Atlanta 1.1
14 Milwaukee -4.3
15 Boston -5.3
Western Conference
1 Dallas 0.9
2 Phoenix 0.5
3 San Antonio -0.9
4 Houston 2.1
5 Utah 3.4
6 Denver -1.4
7 LA Lakers 0.2
8 Golden State 3.3
9 LA Clippers -3.4
10 New Orleans -1
11 Sacramento -1.5
12 Portland 2.9
13 Minnesota -1.5
14 Seattle -0.6
15 Memphis -3.4
Right away you can give a mulligan to the experts on Boston, Milwaukee, and Memphis. Who could have guessed that those teams would be THAT bad? Who could have guessed that both Durant and Oden would be THAT good (creating an environment where there was very little incentive to win)?
Looking down the list, there are a lot of teams that the experts nailed right on the money. In my book, anything under 2 spots differential is some pretty good predicting. The ESPN experts got 18 out of the 30 teams within that range. Not bad. Say it with me Dennis Green. "They are who we thought they were!"
Of course there were some exceptions to the rule. Toronto and Utah did much better than expected. The Clippers were expected to make the playoffs pretty easily and didn't. The Warriors were not expected to make the playoffs and did (and boy did they make some noise when they got in).
Nobody's perfect, but the experts didn't do too bad all in all. So when the previews come out this year, I'll be more likely to pay pretty close attention to them.
Incidentally, I did an exercise last year where I asked bloggers to preview their teams and predict the number of wins they'd get. As you might expect, bloggers overestimated their teams chances by a pretty good clip. See the analysis here. Of course that's part of the fun and yes, we'll be doing it again this year. Stay tuned.
