Articles from around the Web
PDF VERSION NOW DISCOUNTED OVER 30%
Click here to buy PDF version.
Click here to buy PDF version
Official Account: @fboutsiders
Scott Kacsmar: @FO_ScottKacsmar
Ben Muth: @FO_WordofMuth
Aaron Schatz: @FO_ASchatz
Vince Verhei: @FO_VVerhei
-- plus --
Bill Connelly: @SBN_BillC
J.J. Cooper: @jjcoop36
Cian Fahey: @Cianaf
Brian Fremeau: @bcfremeau
Tom Gower: @ThomasGower
Andrew Healy: @AndHealy
Rivers McCown: @RiversMcCown
Chad Peltier: @CGPeltier
Matt Waldman: @MattWaldman
Rob Weintraub: @robwein
15 Feb 2011
TMQ finishes the season with his annual Bad Predictions Review article.
Posted by: Bill Barnwell on 15 Feb 2011
36 comments, Last at
20 Feb 2011, 11:52am by
I don't think Gregg Easterbrook should be the one criticizing anyone for bad predictions, guy is a hack.
That doesn't mean his criticisms aren't apt. "Ad hominems" are fallacies and not examples of sound reasoning.
Yes, but Easterbrook is just as guilty of that. The tone of his criticisms of others' predictions carries a strong undercurrent of judgement: namely, "your predictions were wrong and this reflects poorly upon you as a person."
What the poor success rate in predictions proves is only that predicting things is hard, and that talking heads are obligated to make predictions as part of their jobs. Sadly, Easterbrook thinks that it is perfectly fair and acceptable to jump on people for saying things that they are obligated to say, for example, coaches claiming to have confidence in their teams and players.
The criticisms of TMQ are tired. Much more tired than the column itself, which is still entertaining to read for many people.
It's interesting the difference between King's criticizers and Easterbrook's, both of whom are repetitive. Easterbrook's rely almost entirely on name-calling, with rare refutations of some particular point. King's use more cogent derision of his personality traits and choices, with name-calling sprinkled in, usually including remarks about his obesity.
The two don't seem to have much overlap. It's an odd bird that feels moved to complain about both of them. I wonder if we got them all in a room, how hard it would be to sort them out?
Funny, because I have a completely different perception. First off, I see a lot of overlap between people complaining about them. (Not surprising give they both have huge egos which leads to probably 90% of the criticism leveled against them.)
Also, there are very well-defined gripes about both them--though most come down to their being rather pompous in how they present themselves. Some of the more popular gripes about each:
PK: Egomaniac (evidenced in everything from his name-dropping to his thinking people actually care how his latest Starbucks visit or business trip went); Overly influenced by his personal friendships(see entry under Favre, Brett); Great at getting comments and scoops but horrible at analysis and predictions.
GE: Not nearly as smart as he thinks he is; format of column has grown tired (repetitive but also tamed down to serve his masters--note the Lord Voldermort and Skeletor references are long gone); His "facts" about football tend to be highly selective and/or distorted to fit into whatever point he's attempting to make; Tends to be moralistic.
I like this trick with people whose main arguing technique is name calling. Just replace all their names with "derogatory noun" and all their adjectives with "adjective." It quickly clears the chaff out of their argument so you can see if there's anything there.
"That adjective guy is the biggest derogatory noun in all history. Anyone who would read that derogatory noun is adjective adjective in the brain."
The same flaws that always plague TMQ are as present as ever, but this one is more entertaining than usual.
'Cris Carter predicted Dwayne Bowe would be the "breakout" player of the year. Perhaps he meant to say "shutout," as Bowe had no receptions in Kansas City's home playoff loss.'
This is just being intellectually dishonest.
Then if you want to take the postseason into account, wouldn't you have to include the fact that nearly every writer for SI picked Aaron Rodgers to win the MVP award? Of course, Tom Brady won it unanimously, but if you take the playoffs into account you'd almost have to give the award to Rodgers (unless you want to give it to the big time QB who can't win a playoff game with home field advantage), which Easterbrook is using when evaluating Carter's pick of Bowe.
I don't think he's being intellectually dishonest. I suspect he's just dumb.
I'll agree with you on this one -- Dwayne Bowe was actually a fairly good prediction.
I don't mind the thesis of this column, but I'm going to give this one to Cris Carter on points:
Cris Carter predicted Dwayne Bowe would be the "breakout" player of the year. Perhaps he meant to say "shutout," as Bowe had no receptions in Kansas City's home playoff loss.
I think that Easterbrook misses the mark in his current cause: that rosters should list the highest school from which a player graduated as opposed to the college at which he played. I look at the player's school as giving information about the kind of coaching he recieved and the kind of competition he faced. Along with height, weight, and years of experience, it gives basic football-related information on the player. Perhaps the player's school can be listed with an asterisk to indicate graduation status.
I loved how he blasted Peter King but didn't mention the fact that he at least predicted the two Super Bowl participants correctly before the season started, then held it against him for thinking the Packers may be the best team in the NFL when no one had them number 1 (who of course went on to win the game that mattered).
The bit about the CBO was dishonest, too. The predict deficits based on current law.
I predicted I'd quit reading him this year but I was wrong, so there you go.
In January 2000, the Congressional Budget Office projected a federal surplus of $4.3 trillion for the decade ending in 2010. Actual was a deficit of about $6 trillion. Hey, they were off by only $10 trillion!
Yes, why didn't the CBO take into account the Bush tax cuts, the 9/11 attack, the wars in the Middle East, and the expansion of Medicare benefits! Not to mention the collapse of the financial markets in 2008!! How dare they!!
This is a real problem with predictions: they assume the future will be predictable in ways that are, um, predictable. Really, there's no point in estimating government expenditure over the next 10 years without including a huge asterisk for "unforeseen extras" — just as, to bring the discussion back to football, at the start of the season every team's fans expect their young players to improve, their stars to decline gently if at all, and their roster to stay healthy. It's almost impossible to imagine how your team will perform with 'average' injuries, since you have no idea who will get hurt, nor when, nor how schemes will adjust and backups perform.
Long-termists would argue, for instance, that global wars and financial collapse were predictable, just as FO kept insisting the Cowboys would eventually suffer some injuries. But nobody can predict when these things will happen, nor what their effect would be.
I see this column as showing the futility/absurdity of making predictions (and the silliness needed to make such predictions in the first place), not as an indictment of the predictors themselves.
off on slight tangent because this whole theme of budgets and overruns is so appropriate given my situation at work these days...
when I make a commitment to a budget at work, i wish I got the benefit of being able to say "well, things happened that I didn't predict, so I didn't make my budget". Everyone in the business world knows that's not how it works. You commit to a budget (or one is committed for you :p), you stick to it, and if something unexpected happens to have to find a way to mitigate it, or drop something else off the planned list/do something cheaper to hit the number in the end. That's why it's called a "commitment"
unplanned things happen, i know, but that's what management reserve is for. Don't commit to the budget number unless you're willing to take the steps necessary to accomplish the job on-target, regardless of the hard choices that need to be made along the way when unexpected things happen.
it's how you handle those hard choices that determine whether you'll hit the budget in the end or not. If we continue to overrun our US fiscal budgets year after year, what does that say about how we handle the hard choices?
my 0.02 at 9am
The U.S. government is nothing at all like a corporation.
Indeed. Extending credit to corporations is much more risky than extending credit to the US government; and many corporations make things people do not want or need.
I'm pretty sure the federal government stays within it's budget most of the time.
It's just that the budget is larger than their revenues.
"The Journal predicts sports results with fractional scores, such as Mavs over Thunder by 2.7 points or Devils 3.3, Penguins 2.8. The paper says such predictions are 'based on the average of 10,000 game simulations.' It doesn't matter if there were 10 billion simulations, an average of estimates cannot be precise, let alone generate more accuracy than is possible in the actual result."
Uhhh. Are you serious?
Gregg seems to have a hard time understanding numbers less than 1. He also thinks that humans can't observe tenths of seconds.
I think his point was just that just because 10,000 games are played between two teams doesn't mean that the one game we see is in any way likely to actually have that final score. Sure, it might have the highest probability to have that score, but is it relevant if that probability is less than 10% of being within a couple points of that score? And if you only have, say, a 10% chance of being within a couple points to that score, what's the point of predicting to the nearest .1 of a point? Let's look at the Jets and Patriots this year as an example. Maybe the Patriots win by an average score of a TD based on 10,000 simulations based on DVOA, but the actual results showed such a large variance (win by 42 and losses by 7 and 14) is quite silly.
Goes back to precision vs. accuracy, just because you fire three shots off a bullseye, each one foot off at an 120 angle about the target thus meaning an average of right on the center, in no way means that you were particularly accurate.
Exactly. His overall theme that predictions are rather pointless is sound. What is odd is he picks such poor examples in some cases. The Bowe thing was outright absurd--there are countless other examples he could have used (and his other examples were pretty sound). Part of his problem appears to be that his column runs without any outside editing--it looks like he just emails it in and it gets posted as soon as the photos and cutlines are placed. A good editor would have caught the Bowe example and had him change it...and anybody would have caught the fact he left "Wyatt Earp" out of his list of Wyatt Earp movies a few weeks back.
People keep arguing that he means that predictions are rather pointless, but then he keeps including these "examples" where he criticizes coaches and front office people for saying in press conferences that they have confidence in their teams and players. Doesn't he understand that they HAVE to say that? He's lambasting people for doing their jobs. That's not pointing out how meaningless predictions are, that's taking cheap shots at people who often don't have better options.
Enough with the g-ddamn politics already.
There's always been an exception made for Easterbrook's columns, which include a lot of political comments.
I marvel at many of the topics that appear in TMQ, as the peak of my lack of interest in GE's writing coincides with where it becomes most passionate. And then the following stunned me:
"I recommend you employ the offseason to engage in spiritual growth. Take long walks. Attend worship services of any faith. Exercise more and eat less. Perform volunteer work. Appreciate the beauty of nature. Read, meditate, serve others. Do these things, and you will feel justified in racing back to the remote, the swimsuit calendars and the microbrews when the football artificial universe resumes anew in the autumn."
GE, how did you get inside my head?
At least there will be college football come autumn even if there are no NFL games.
I don't believe G.E.'s approach that all predictions are wrong or useless is correct.
Having been a sales manager for many years in my career, I know forecasts/predictions are necessary. I look at a prediction as correct if it accurately reflects what is believed at the time it is made. The world will change and more will be known tomorrow than is known today and the forecast/prediction should change to reflect the changes.
If anyone expects reality to coincide with predictions or forecasts, they are a bit naive.
"If anyone expects reality to coincide with predictions or forecasts, they are a bit naive."
Then what's the point of making the prediction? To make you feel good about the choices you're making even when you know the prediction is worthless?
FTR, I don't think ALL predictions are worthless, and I'm not sure the Easterbrook does either. He does enjoy pointing out predictions that are overly specific when the known error margin is large.
The point, from a business perspective (and football teams), is to be able to plan. Plan for the allocation of time, materials, personnel, focus and individual responsibilities. Not every resource a business has at its disposal can be deployed immediately.
Sales Depts. need budget and cost numbers (for pricing), sales and market forecasts to know where to mine for expanded or new business, to know what questions and concerns their clients are going to have and to know whether to focus on expanding a client base or work feverishly to keep one.
Predictions made by people who have no horse in the race? For entertainment purposes only.
The Seahawks' defensive back will tell you he's the best corner in the game. Is he right?
See All XP | NFL XP | College XP
© Football Outsiders, Inc. // Site powered by Stein-Wein // Partner of USA TODAY Sports Digital Properties