I Love It...

...when the the narrative-chasing intuition monkeys get involved.

  • Spread The Love
  • Digg This Post
  • Tweet This Post
  • Stumble This Post
  • Submit This Post To Delicious
  • Submit This Post To Reddit
  • Submit This Post To Mixx

39 Responses to “I Love It...”

  1. Anonymous says:

    "Narrative-Chasing Intuition Monkeys" gets zero hits on Google. I will be sure to credit you when I use it!

  2. Anonymous says:

    Tebow no doubt MVP. Possibly HOF material if he can carry his team to the playoffs. He just needs another year or two like this and he is first ballot.

  3. James says:

    I thought the most outrageous thing he said was "The Packers wouldn’t be perfect without [Rodgers]. But they would win their division."

    Based on WHAT? Their pitiful running game that averaged 2.4 ypc against the Giants, not including Rodgers' scrambles? Their run defense that allowed 5 ypc to a team averaging 3.6? Or is it the Packers passing defense, the one everyone defended by saying they've been playing soft coverages with a huge lead, that allowed Eli to throw for 8.7 ypa, above his season average, in a game they never led by more than 11? The defense that allowed the Giants to take the lead or tie the game 4 different times? What about that says the division crown is a done deal?

    Also, since when is clinching the division in Week 13 and gunning for a #1 seed not "valuable"?

  4. Anonymous says:

    I would never argue that Tebow is anything like an MVP candidate, Brian, but I actually have used numbers from your website to convey that Tebow can be a viable offensive asset when conversing with some of his very harsh critics.

  5. Jussi says:

    What was this Couch person saying? I didn't understand a thing. Tebow is great because the writer's son likes to bang around under the basket?

  6. Dan Whitney says:

    Another aspect of this argument most people are not talking about is the weak schedule Tebow has faced. Here are the teams he has played
    Week 7 Miami W 18-15 OT
    Week 8 Detroit L 45-10
    Week 9 Oakland W 38-34
    Week 10 Kansas City W 17-10
    Week 11 New York Jets W 17-13
    Week 12 San Diego W 16-13 OT
    Week 13 Minnesota 35-32
    I also looked up the Time of possession since many people claim the running style offense is keeping their defense off the field. In most game the Broncos have won they have very slightly out-possessed the defense (most games they had the ball for about 32 minutes Highest game was about 33:30). In fact in the past game against the Vikings the Broncos only held the ball for 23 minutes.
    Just look at those teams and scores can tell you all you need to know about Tebow and the Broncos. They are a mediocre team that has just managed to squeak out wins against other mediocre teams behind stellar defense (in four of their wins they have scored >20 and the opposing team has scored >15) We'll see how Tebow and the Broncos D perform when they play a top tier NFL team rather then other mediocre teams.

  7. Anonymous says:

    I'd have to look it up, but I'm pretty sure Tim Tebow wasn't the one who came up with that interception with three minutes left in the Vikings game. Nor was he the one who missed the field goal for the Chargers in overtime last week.

  8. Anonymous says:

    Tebow has the ability to get into other kickers heads. How would you like to try to make a kick when you know fate is on the other team.
    How can anyone not vote for Tebow for MVP? I would love to hear their rationale.

  9. sunrise089 says:

    @Dan - Your analysis will look smarter when you learn what the ">" sign means.

    Anyways, so Tebow has played a week schedule. Since you suggest his wins are therefore unimpressive I bet he was heavily favored to win all of his games, right? Actually according to Brian's game probabilities Denver has been favored in ZERO of the games Tebow has won, and had less than a 30% chance of winning several times. Total probability of Tebow winning the games he has: .2%.

    The probability of Green Bay winning the last 9 of their games in a row, by the way, was 2.6%. An order of magnitude higher. I'd say it was "greater than" Denver's, but we know you have issues there...

    I really love AdvancedNFLStats, but it's becoming a frustrating site to visit with how cultish some of the anti-Tebow sentiment can be. Yes pro-Tebow folks talk about non-mathy things like "intangibles," but as Brian himself said if those things matter they'll be reflected in stats. Wins Are A Stat.

    Brian's "narrative-chasing intuition monkey" line was funny, but anti-Tebow pro-stats folks shouldn't go down the same road. The Broncos won 35% of the games Kyle Orton started in '09-'11, 58% of their 2011 games (with the supposed elite defense the anti-Tebow crowd credits most victories to), but 86% of Tebow's games. Yes the sample size is small, which is why I'm referencing the .2% chance of winning above. A 2-5% chance...that's normal with 32 teams. .2% makes me at least start to think there might be something here. Remember, Belichick doesn't throw a pass or make a tackle, but most pro-stats folks still think he has an impact on wins, because his teams consistently win more games than we'd expect due to randomness. You could say it's luck, as I would to explain, say, Rex Ryan's playoff success so far. But at a certain point Occam's razor prevails.

    In the end do I expect Tebow to win the super bowl this year? Of course not, I expect him to lose 1-2 of his remaining games. But it's so consistently disappointing seeing otherwise forward thinking sports fans struggling to come up with reasons Tebow deserves to fail.

    Miami, with one of the greatest come-from-behind victories ever - "Orton [Mr. 35%] wouldn't have needed the big comeback."

    Oakland: "It was all the running game [because Tebow doesn't make the run defense have to account for another guy AT ALL]."

    Kansas City: "Bad rival [despite Orton loosing at home to common opponents Tebow would beat on the road]."

    Jets: "Credit the defense [who apparently played for a different team pre-Tebow]."

    San Diego: "The kicking game...maybe...?"

    Minnesota: "Well we did say Tebow couldn't win a shootout...but...well the defense got some lucky turnovers..."

    Come on guys!

  10. Anonymous says:

    ANS has a win probability model. It might not be a perfect win probability model (in fact it is most certainly not perfect, that would be impossible without time travel), but based on this model, Tebow has had less of a direct contribution to the Broncos' win probability than the defense has had. You certainly have some points, particularly with defenses having to account for an extra runner, but Tebow doesn't play on defense at all.

    That isn't to say he's not somehow having an effect on the rest of the team even when he's off the field. If nothing else, the fact that the Broncos have come back from a terrible start is cause for hope among their players. But psychological effects are not quantifiable and fall outside the scope of this site.

  11. James Sinclair says:

    There's anti-Tebow sentiment here? If you say so, I guess, but I haven't seen it. I think what's being labeled anti-Tebow sentiment is really an appreciation for looking at all the factors contributing to a team's success, which can include quarterback play, defense, opponent strength, luck, and, possibly, Jesus. Finding holes in the popular narrative and looking for alternate explanations isn't the same thing as "struggling to come up with reasons Tebow deserves to fail."

    I'm a long-time Advanced NFL Stats fan/commenter, and I'm all in favor of Tebow continuing to win in inexplicable ways, because it's fun (of course, I went to Florida, and I'm not fanatical about former Gators or anything, but I had a head start on the enjoying-Tebow-because-he-drives-everyone-else-crazy mentality). But I think it's always wise to be skeptical when one player starts getting this much credit.

  12. sunrise089 says:

    @Anonymous directly above me:

    Ok, that's pretty fair. I only want to suggest two things. First, it seems to me a psychological effect COULD be quantifiable, even if that was quite uncommon. For example if Tebow got hurt a lot and the defense had +.3 WPA/game (like Carolina) when he was out and -.15 WPA/game (like Baltimore) when he was playing, and that held true over dozens of games, that's quantifiable. I'm no stat expert, but I'm curious what number of wins above expected wins a team needs to have before it's more likely that a model doesn't capture something unique to that team versus the team just being lucky.

    Second, if the anti-Tebow crowd just said this discussion was "outside the scope of this site" I'd have no real issue. Many of them don't though. They do things like Dan Whitney does above with the same sort of ex-post search for meaning narratives the anti-stats sportswriters do in their columns.

  13. tunesmith says:

    I'm just curious though, what if a game plan broke the model? Completely aside from Tebow.

    Say you have a coach that puts a premium on ball control, limiting mistakes, and time of possession - ahead of always taking the best percentage shot of scoring a touchdown. Say that extends to opening up the playbook when in danger of being behind two scores, and closing down the playbook when ahead.

    In those cases, wouldn't you expect a bunch of strange statistical outputs?

    Another thing... those "Well, Tebow wouldn't have won if not for the interceptions" statements drive me batty. After a point, you have to remember that there's a football game being played. And, Denver also wouldn't have won had Tebow not thrown a touchdown pass after stiff-arming a defender to the ground. I suspect that that sort of play is pretty rare.

  14. James Sinclair says:

    "ball control, limiting mistakes, and time of possession"

    That sounds a little like how the Falcons play, and they have indeed made a habit over the last few years of winning in defiance of the odds (yesterday's game notwithstanding—that was a strange mess, albeit a mess the Falcons could easily have won despite being huge underdogs in terms of win probability).

  15. Unknown says:

    @tunesmith and James Sinclair, limiting the number of possessions and keeping games close seems like a good high-variance strategy, you'll win closer to half the games even if you are the inferior team (maybe you'll win more than half the games if you have a skill at winning close games, though I'm not sure that exists). But ultimately, wouldn't you rather be the superior team and employ the low variance strategy? Seems like ball-control should generally be the insurgent strategy, and teams with Super Bowl aspirations should attempt to become the superior team so they can apply the strategy that demonstrates their superiority (e.g. maximizing possessions and maximizing possession efficiency).

  16. Joshua Northey says:

    Tebow has been proving a lot of his doubters (including me) wrong. Perhaps he can consistently be the 20th best QB in the league and hang on as a below average starter like Mark Sanchez. I thought there was almost no chance he could be named the starter next year, now it looks likely.

    And that is exactly the point! The upside for Tebow is he doesn't wash out in 2 years, yet from the coverage you would think his is a potential Top5 QB. There is a 0% chance of that.

    If that defense was a little worse, or the bounces a little less DEN friendly, people would be talking about how his career was doomed. Even with exactly the same performance out of Tebow a 2-4 record with his current stats would have most DEN fans out for blood.

    That is what drives us advanced stats people nuts. Not his play, we like novel play. We hate watching the outcome bias orgy around a team that has won 4 games while scoring under 20 pts. That just doesn't happen in the NFL consistently unless you have a tremendous defense.

    Year after year we watch people overvalue close wins, overvalue "intangibles" (not because they don't exist but because they have no better handle on them than anyone else, they are after all "intangibles"), overvalue offense over defense, overvalue QB play compared to the rest of the team, yada yada yada.

    Tebow is an embodiment of all of that, and a very polarizing player stylistically and personally to boot.

    Trust me if Tebow was playing exactly like this but the Broncos were 1-5 in close games the advanced stats community would be all "Tebow is much better than ESPN thinks he is" and sticking up for him left and right while the hoi polloi demanded he be cut immediately.

  17. Dan Whitney says:

    Tunesmith and James Sinclair. I think it is possible to "break the model" but I don't think that the Broncos are following that. The Broncos as a whole are below average in time of possession, and according to Football Outsiders rank 27 in drive success rate with only 63.8% of their first downs resulting in another first down or touchdown. That doesn't sound like they are controlling the ball or time of possession. If you want a team that is breaking the model in those ways I think it would be a team like San Francisco. They seem to always be controlling the ball and maintaining time of possession.

    @Sunrise First of all I'm not trying to say what the Broncos have done is not impressive. They have been able to pull off a bunch of close victories. I'm just saying that we should not make them out to be an amazing team. What really annoys me about the media and comments like yours is the line "games Tebow has won". It is not Tim Tebow who is winning these games it is the Broncos. I think that part of the reason that Advanced NFL Stats has such an anti-Tebow sentiment is that the regular media is completely blowing Tebowmania out of proportion.
    I firmly belief that if any other backup quarterback besides Tebow had come in and won the games in the same manner the media would not be hyping the story. Look at a QB like Mark Sanchez, He came in his first few years and managed to sneak out a couple close victories behind a strong defense. Yet people never claimed he was a "winner" the way they do with Tebow.
    In terms of the Broncos low chances of winning these games, yes you are right what they have done is pretty impressive, and Tebow does deserve some credit for that. However as Brian's WP calculations point out the defense is much more responsible for adding wins than Tebow is. Even other sites such as ESPN QBR have Tebow ranked as the #31 QB (before this week). Win's may be an important stats but I can't stand when my friends and the media tell me well he's doing nothing but winning so he must be a winner. Wining is not necessarily an indicative stat of how the team played, a close loss to a great team is a lot better then a narrow win over a bad team, in terms of showing how strong a team really is.
    Overall I have nothing against Tim Tebow, and think that he is a real role model both on and off the field. I just can't stand the media zealots who proclaim Tebow as a "winner" and ignore the rest of the team on the field.

  18. tunesmith says:

    I'm not trying to argue Tebow as a "winner", either - I believe that clutch doesn't really exist, and that a "clutch" performance is really just more indicative of that quarterback's *general* level of performance, except more frequently displayed on highlights.

    So, what are we to make of when a quarterback, late in the game (or in "high-leverage" situations), actually outperforms his performance from the rest of the game?

    Well, the explanation that is most commonly offered is that it's just variance, and that we'd expect a regression to the mean over time.

    But the other possible explanation is that at those other "non-clutch" times, the quarterback is not being given the opportunity to play to the best of his ability.

    Also, I don't believe that the Broncos have been playing high-variance football. Tebow is a conservative passer (being asked to "beach the ball" if the pass doesn't look high-percentage), and in the case of zone-read, we shouldn't mistake uncommon for risky. While it may be "risky" for a coaching staff, it doesn't mean it's risky in terms of execution compared to other high-variance plays like laterals and reverses.

    Anyway, my point isn't that Tebow "caused" or even deserves all of the wins he has so far. But it is that Tebow may already be a better quarterback than he statistically looks - capable of better QB performances (and higher point output) than the game plans are currently allowing him.

  19. Jim Glass says:

    Since you suggest his wins are therefore unimpressive ...

    I love it -- 53 players on the team, a coaching staff of how many(?) ... and they are *his* wins.

    It's like he's Roger Federer.

  20. Anonymous says:

    Matt Flynn for MVP!

    You Flynn-haters drive me crazy, with your "numbers" and "stats." The fact is, wins are a stat, and Matt Flynn is 12-0 this year. From 2005-2007, the Packers were a mediocre 25-23, but since the team signed Flynn, they're an impressive 39-21 -- and, oh yeah, ONE RING.

    All Flynn does is find a way to win every single week. Whatever numbers and acronyms you come up with to try to explain him, make sure this is among them: MVP.

  21. Anonymous says:

    Tebow leads the league in intangibles.
    You cant vote based on passing yards. You have to make your vote based on intangibles. That is why Tebow is the MVP and easily the best all around player in the league.

  22. Brian Burke says:

    Matt Flynn comment +1

  23. KJ says:

    There is no 'intangible' in football. None. There are just things you are unable to express. Not a knock on you, it's just that expressing anything that makes Tebow more than a below-average NFL QB is very hard to express indeed.

  24. Anonymous says:

    Great article Brian. Thanks for sharing the link. FWIW I agree 100% with the author as well. Glad we are on the same page. Whoever does not think Tebow is MVP doesn't know anything about football. Broncos are about equal to the Colts without Tebow.
    Tebow is filled with intangibles. That's all that needs to be said.

  25. Anonymous says:

    Q: What's the difference between Tim Tebow and Lebron James? A: Tim Tebow comes alive in the 4th quarter!

    Q: Do you know what AT&T stands for? A: "Another Tebow Touchdown."

    Q: Why is Disneyland rebranding itself? A: Tim Tebow has made Denver the happiest place on Earth.

    Q: Why Did Tim Tebow win MVP? A: Because Brian Burke said so.

  26. Mike B says:

    If ever a thread called for the "not sure if serious" joker pic...

  27. Dave says:

    Another +1 on the Matt Flynn comment. In the words of Chang from Community, "Using it".

  28. Anonymous says:

    Agree that Tebow may not be MVP. Rogers is too good. But no doubt, runner up.

  29. Anonymous says:

    +1 on the Rogers comment. But thats because Tebow has not played the entire year and Rogers has.

  30. Anonymous says:

    "+1 on the Rogers comment. But thats because Tebow has not played the entire year and Rogers has."

    oh yeah, if Tebow played from week 1, he is the MVP. I don't think its much of a discussion.

  31. Anonymous says:

    I have been watching football for 3 years now and have won 2 fantasy football leagues (finished 3rd in another) so take it from me guys, this is the best player I have EVER seen. MVP or not, this guy is a once in a generation phenomena.

  32. Anonymous says:

    Meanwhile, another team sporting a rookie QB has gone from 32nd to 8th in points scored, and he's less than 450 yards from the record for yards in a rookie season with 4 games to go. *crickets*

    Incidentally, the 1998 Giants started 3-7. Then this QB comes out of nowhere and, despite putting up relatively poor numbers, rattles off 5 wins in the next 6 games. Yeah, Kent Graham was totally more valuable than Terrell Davis' 2000 yard season, right? right?

  33. Jonathan says:

    This comments section is amazing.

  34. Anonymous says:

    Tebowmania (which i become to like it since two weeks :-))reminds me of BigMac. A punky QB who once won 25 starts in a row (NOT including the famous come in relief throw 3 TD´s in 8 minutes comeback vs MIN in a 1985 TNF-Game)and thus earning him the "unbeatable" QB-Status (which earned him Mio´s in endorsements) although posting only mediocre numbers (+ missing half of his games b/c of injuries).....

    OTOH he wasn´t truly like Tebow since his efficiency numbers were above NFL-Avg, while his replacements performed under NFL-Avg.(Fuller, Flutie at this time, Avellini etc.).

    Karl, Germany

  35. Anonymous says:

    Being a diehard Seminole and Raiders fan, NO ONE hates Tebow more than I do. But even I have to admit, he is the best player in football by a large margin.

  36. Anonymous says:

    I am not smart enough to know if he is the best player in football or not. That said, my son said at school today his PE teacher said Tebow is playing better football than any player he has seen in the last 10 years. The guy used to play semi-pro ball, so I value his opinion.
    I guess he swayed my vote.
    Tebow for MVP!

  37. Anonymous says:

    If the commentariat here were an actual majority opinion of Denver's fanbase, then I'd say: Let Tebow win the MVP. It will serve them right.

    These posts will be replaced with venom towards Tebow eventually. Bank on it.

  38. Anonymous says:

    True and disturbing story. My 5th grade sons homework assignment last week was to write 2 pages on their hero. The class is comprised of 10 males and 12 females. 2 males and 3 females wrote about 1 of their parents. 6 males and 3 females wrote it on Tim Tebow.

  39. Stringer says:

    This a lot more complicated than people make it out to be. I think ultimately a lot of the credit for the Broncos' success has to go to the coaching staff. Its clear they have plays they know will work for Tebow, but there aren't many so they do a good job of using those when necessary. If anyone recalls the Jet game, the final scoring drive was essentially all designed QB runs. They didn't run any of those plays prior to that final drive.

    The fact is that statistics can give you an accurate reflection 99% of the time. This may be the 1% of the time that they do not. Surely nobody keeps statistics on fatigue or stress. Is performing well under stressful situations, or in situations where you are fatigued a skill?

    Does Tim Tebow get statistical credit when Willis McGahee has a big play, if he forced LBs or DEs to miss their run-fits?

    The bottom line is that Tim Tebow does two things very, very well. He is great in short yardage, and he is great at minimizing turnovers. Normally turnovers will negate the value of great short yardage play.

Leave a Reply

Note: Only a member of this blog may post a comment.