24 Dec 2013
We would like to apologize for last week's ESPN Playoff Odds Report, which contained numerous errors.
What happened? We wanted to make sure that when Danny Tuccitto wrote the article, he had the proper equation that the playoff odds report uses to figure out the odds for each game. Unfortunately, when I sent him the equation, I apparently sent the wrong one, something we had used in a different simulation in the past. Therefore, in last week's Playoff Odds Report, some of the single-game probabilities were incorrect. (They were generally slanted way too heavily towards the favorite.) The odds given for two-game scenarios (i.e. what appeared in the Playoff Odds Report here on FO) were correct.
Because the Monday game between Detroit and Baltimore was so important for playoff odds, we didn't end up writing the article until really late. Rushed for time, neither Danny nor I did our due diligence, which means neither of us noticed that the single-game probabilities did not at all match up with the playoff odds. That's on us, huge mistake, and we apologize. We're watching to make sure this does not happen in the future. This week's Playoff Odds Report at ESPN Insider was checked extra to make sure the numbers agreed with each other properly. (You can read that article here.)
We don't like to make mistakes, of course. We're human, and it happens, but we're not happy about it. In the future, if you discover that we've made a multiplication error in one of our articles, the best way to let us know is through e-mail to mailbag-at-footballoutsiders.com. Commenting in the discussion thread on Football Outsiders will sometimes get our attention, but not always. We only recommend using Twitter if you don't expect a response in return.
Speaking of Twitter, one thing I've learned these past few days while dealing with this issue is that 140 characters is simply not enough space to respond properly to criticism without seeming snarky and unpleasant. Therefore, in the future, I will not be responding to criticism over Twitter. If you just need to bring our attention to a small error, then sure, use Twitter. However, if you feel that a larger piece of analysis is subpar or want further explanation of something I have written, please ask your question over e-mail. You can also use a discussion thread, although I am less likely to respond. However, with either e-mail or a discussion thread comment, I can give a response that properly respects the reader's point of view without a character limit that forces incomplete responses that often read as obnoxious.
(OK, let's try this again... There seems to be a lot of misunderstanding about what I'm asking for here. I have no problem with being criticized on Twitter. The problem is that people criticize us on Twitter, then criticize us a second time because they feel the response is too short or too obnoxious. But that's the thing about Twitter... it's short and obnoxious. I was upset because the Twitter response to our mistake in this article was so negative, but then someone pointed out that my own response to someone's criticism of us in the past had been equally snarky and negative. I don't really want to be like that. If people have legitimate criticism that deserves a well thought out response, I want to give a well thought out response. And so please, by all means, criticize us publicly on Twitter when we make mistakes. What I'm saying here is that if you want to actually get a response to your criticism longer than just "whoops, we'll go fix that, thanks for letting us know" then Twitter is really not the way to go.)
Thanks again for understanding, everyone. As always, we're doing our best around here.
UPDATE: There is now a corrected version of the article posted at ESPN Insider.
47 comments, Last at 23 Jan 2014, 8:32am by Cheap Fitted Hats
How much do we tend to know after five weeks? Bill Connelly compares five-week data to full-season data to find out if we should be worried about TCU and Baylor.