As we’ve been saying: Thanks to everyone who’s submitted so far to The Barking Crow’s Best of 2022 collections. If you’ve yet to make submissions and you’d like to do that, send them over to allthingsnit@gmail.com, or message us on Twitter or Instagram. Categories remaining are…
- The Barking Crow’s Best Work of 2022
- The Barking Crow’s Silliest Work of 2022
Now. Our right-est work of 2022:
Yesterday, we told you about our wrong-est work, so if you like watching us eat—ahem—crow, there’s that. These are the things we got right. The list is bigger than the wrong list, but I also got to make both lists, so you are not dealing with an impartial judiciary here. Moving on, quickly, to muddy the waters…
Movelor Was Good
We took a little risk using Movelor as the simulator for our College Football Playoff model rather than outsourcing the task to a more proven rating system, but Movelor did the job more than adequately. It was on top of USC being overrated by markets and media. It was an early Michigan believer. When Notre Dame rose back up against Clemson, Movelor was unsurprised. This is cherry-picking, to an extent, but the last I checked—mid-November—the average margin of error was in the 13’s, meaning Movelor missed final game margins by an average of roughly 13.5 points, not far off the 12 or 12.5-point error margins seen from the best models out there. Movelor is very simple relative to these models, but simplicity can, in the college football realm, evidently get you a large share of the way there.
Tom Brady Tried to Be a Dolphin
This was luck, but we did publish a tip on February 22nd saying Tom Brady was trying to not only come out of retirement, but play for the Dolphins, of all teams. Looking back, we should have dug further. We may have uncovered Brady and Stephen Ross’s grand scheme, which should have been the biggest sports scandal of 2022.
Three Baseball Takes
We buried this first one, but before Joe Girardi was even fired, we were on top of the Phillies being flawed but completely fine. Our only error was the degree to which they were still ok.
We were likewise on top of the inconsistency of relievers, praising the Brewers for dumping Josh Hader when they did and the Cubs for trading Scott Effross. This is an old theme of ours, but it continues to be correct. If you have a good relief pitcher, trade them. They almost never last.
Finally, we were ahead of the curve on acknowledging that Aaron Judge was definitively better than Shohei Ohtani in the 2022 regular season. This isn’t as subjective as MVP voting itself is—the numbers didn’t lie, Judge was a silly amount better than everyone else, Ohtani included—but it was a nice bonus to see the media come around and vote Judge in for the title.
Conference Realignment
Were we right about conference realignment? I’m not sure. We predicted about a dozen different things, and there is still time for most of them to play out, because the Pac-12 is still the Pac-10 and the Pac-10 lacks a media rights deal and those things, in conjunction, imply more change is on its way. We did, though, have a fairly clear-eyed view on the situation, and since most of the reporting on the matter came from participants like ESPN and FOX and even CBS, there was an absence of that in the market. I’d say our best reflection on any moment’s situation was this one, although I’ll concede that I very firmly stated in there that TCU was not going to bring in College Football Playoff revenue anytime soon, which probably could have made yesterday’s post.
Trust Nate Silver
For those who follow the bets, this was the big one, and it’s a lesson political media continues to ignore. Nate Silver and FiveThirtyEight are right a whole lot, and we were right about him (and them) being right. For the second straight election cycle, it made us a lot of money.
The thing too many journalists don’t get is probability, and the thing that leads too many journalists to not get, in this sphere, is that the point of the FiveThirtyEight model is not to predict winners. Any old idiot with a blog (you’re looking at one) can do that. The point of the FiveThirtyEight model is to give probabilities of certain events happening, and to make those as exact as possible. For bettors like us, that’s useful for finding value in markets (place enough bets at 2-to-1 odds on something 40% likely to happen and water will eventually find its level, a level that leaves you with a 20% return on your investment). For journalists, that’s useful for narrating the state of the race without pretending something called “vibes” exists with this stuff when you’re really just participating in blind, aimless groupthink.
So, to be right one last time, let us offer this: Journalism will improve as an industry if journalism education begins to include rigorous instruction in probability. In the meantime, Nate Silver will continue to be right, and we will continue to point his way.
