Sidenote, I can't let this one go (lol), you don't EVER dismiss any of Tannehill's first season or two when discussing him.
No. 7 pt games account for 46% of all games. They are not special, unusual or unique. The QB who typically goes 12-4, i.e., wins 75% of his games, will likely have 7-8 of those games be close ones. And he'll probably win about 75% of those. A QB can "win" a "close" (7 pts here) game by being great through 55 minutes to build a 21 pt lead and then throw 2 pick-6s in the last 5 minutes to end up with a 7 pt win.
You're just making assertions there. Can you provide evidence that most "close" games are the result of the winning team actually having led through most of the game? Absent that, the default assumption here should be that close games are the result of two teams playing at similar strength.
C'mon, you know as well as I do that such a conclusion is a stretch. First, it wasn't any kind of gradual improvement at all. His first 5 playoff games he had ratings of 59.1, 89.4, 18.2, 10 and 48.4. That's one average (in today's NFL) game and 4 horrific disasters. Then, magically, the Clutchness Fairy appeared at his bedside and sprinkled him with fairy dust and he posted ratings of 115.4, 125.6, 116.2, 106.3, 124.2 and 114 in 6 of his next 9 playoff games. Yes, that's a very impressive stretch of games. But no, it is not unprecedented or limited to the playoffs. In 2010, he had a 9 game stretch in the regular season in which he posted ratings of 119.3, 111.0, 129.6, 99.5, 110.8, 103.2, 88.7, 107.5, 112.9. It happens. Flacco is a guy who has been great for certain stretches of his career, average for other stretches, an abysmal for others. That happens whether it is playoffs or regular season.
I don't think you realize how strong a statistical argument it is to say Flacco most likely improved. Take the correlation between his passer ratings as a function of year (so for each year you plot the passer ratings, and the "year" is just ordinal in nature, so: 1,2,3,4,5,6 for the 6 playoff years). The correlation is 0.7276. The sample size is 15. Do you know what the "p-value" is for those numbers? That is, the probability that occurred by chance? Try it out yourself: http://www.danielsoper.com/statcalc/calculator.aspx?id=44 We want the two-tailed test here (meaning we a priori don't know whether the correlation should be high or low), and that is 0.0021, meaning it's 0.21% likely what Flacco did occurred by chance!! The conclusion that it was NOT due to an actual improvement in skill is actually a massive stretch!
I haven't said most close games are like that. But certainly some are. I've seen them dozens, if not hundreds, of times. It's not usual that a team that is losing big starts chucking the ball and the defense goes into prevent and allows a few scores to make it a close game, but still hold on to win. If you were to say you haven't seen that at least dozens of times, I'd conclude that you are either lying or you don't watch much football at all. But I'm pretty sure you won't make that claim. So no, the default assumption should not be that both teams played at similar strength for the entirety of the game because that would contradict what everybody has seen for at least the last 40 years.
Doesn't matter that some are like that. The question is what do most close games look like? Default assumption is that on average close games occur when both teams are playing at similar strength. I'll accept a different assumption when evidence is provided against it. And if both are playing at similar strength, then the expectation should be 50% win probability in close games irrespective of those teams' records.
That's not what I've said. Joe Flacco has certainly improved since his rookie year. But his playoff "improvement" was anything but gradual. In fact, it is about as dramatic as one could possibly imagine. And that isn't a function of his own improvement as a player -- it is a function of chance, inconsistency, etc. If your correlation is comparing playoff rating to regular season rating, then the small sample size makes it useless. For some years the playoff sample is 2 games. It is never more than 4 games (once).
I only used playoff ratings. The statistical evidence he improved in the playoffs is massive! The sample size if you didn't notice is taken into account. Change the sample size you get a different result. So GIVEN the sample size of 15 his playoff performance (that is, it's not due to chance) most likely improved with a likelihood of 99.79%!!!
I am not cherry picking. What you describe for RG3 is cherry picking, and I did no such thing. Cherry picking is random selection of stats for no reason except to prop (or drop) the subject. I provided further insight by showing his really poor games were at the beginning of his career as a rookie and first year player. His performance since then has been out of this world. These are called "trend lines." if his bad years were the last two years then my argument wouldn't work. Cherry picking would be trying to single out a 10 game stretch of really good play and ignore stuff before and after that.
No. Simply not true. IMO most games that end close are not like that. I don't have readily available stats over any long period fo time for it, but just anecdotally looked at the Pats' close games from last year. And by "close" I am using the 7 pt spread talked about on this thread even though I don't think that means much for the reasons I have stated. The Pats had 8 games decided by 7 pts or less. Of those 8, 3 were relatively close throughout, whereas 5 were not very close for most fo the game but appeared close because of a late gimme score. Pats 28, Pitt 21. Pats were up comfortably 28-14 but gave up a gimme TD with 7 second left to make it a 7 pt game. Pats 34, Indy 27. Pats were up 34-21 but gave up a gimme score with 1:26 left to make it a 7 pt game. Pats 30, NYJ 23. Pats were up 30-20 with 23 seconds left and gave up a meaningless FG to make it a 7 pt game. Pats 20, Bills 13. Pats were up 20-10 with 3:34 left but gave up FG to make it a 7 pt game. Philly 35, Pats 28. Pats were down 35-14 with 5:33 left but scored 2 late TDs to make it a 7 pt game. In the first 4 games listed above, Brady did nothing late in those games to win them. Those games were already won. They were "close" only because the other team scored meaningless points late, something Brady had nothing to do with. The one loss in that group -- the Philly game -- was the one game that Brady did something late. But it wasn't enough and he is "credited" with the loss in a close game. Of the 3 games that were actually pretty close throughout, Brady lost 2. Does that mean Brady isn't clutch? And again, you can't assume a 50% win probability in close games at all. Of the 5 close Pats games that they actually won, the Pats led by more than a TD in the final minutes of those games. Those are not 50% win scenarios. Even bad teams should win more than 50% of the games when they are up by more than a TD in the final minutes.
Or he just had a great run like he had in the regular season in 2010. Run your numbers for that 9 game stretch and I'm sure they'll suggest it wasn't just a random stretch. The correlation will be basically the same as what you provided for the playoff run. But we know that 2010 run was just random because he hasn't sustained it, or come anywhere close to sustaining it, in 4 seasons since then. Hot streaks and runs like that do happen. And then they end. It's very common in every sport.
You did cherry pick. You concluded that Joe Flacco is clutch as evidenced by his playoff performances. Then you saw that he had several awful playoff performances and excluded them. Then you justified their exclusion on the grounds that it was early in his career. But it could just as easily be that he had a bad streak followed by a hot streak. That happens all the time. When you look at the full set of data for both regular season and post-season you see a 4 pt higher rating in the post-season. As I said, that's about one extra 12 yd completion in a 36 pass attempt game. It's nothing. And with such a small sample size it is simply not significant.
Here's a stat that is evidence of "clutch", why are so many world records broken or set in the Olympics, the highest stage, when most events, by a very large margin are NOT held at the Olympics, since the Olymipcs are held only once every 4 years. Then after that, it's the Worlds, which are once a year, between the Olympics and the Worlds, that takes care of most records, but there are multitudes of other events, so how is it that these records are being broken or set on the highest stages so often? That's called clutch performances.
OK, since you brought up the Pats, I went through all (except apparently one!) of their 0-7 games with Brady and looked at how many times they were ahead at the end of the 3rd quarter. I must've missed one game (not going through that again!) because Brady's overall record in close games is 63-28 for a total of 91 games, and I have 90 counted. Anyway, out of the 90 I counted, he was ahead at the end of the 3rd in 44 of those games, which is just about 50%!! So yeah, the assumption works even for Brady. What you're demonstrating is just selective memory.
Here's the 2010 season: http://www.pro-football-reference.com/players/F/FlacJo00/gamelog/2010/ If you mean the first 9 games you're right the correlation is high at 0.6148, and with sample size 9 you get a 7.8% chance that was due to chance. So yes if you pick those 9 games and that's the only data you have, one could argue that was unlikely to be due to chance (though technically statisticians won't reject the hypothesis it was due to chance unless it's less than 5% in most cases), but 7.8% is VASTLY different from 0.21%, showing you just how unlikely Flacco's playoff performance really is compared to a cherry picked 9 games in the regular season. And the one problem with picking those 9 games is that it really is cherry picked because you can't independently argue the conditions were different from other regular season games. You can argue the playoffs represent a different condition.
Adrenaline is strength based though. That can be useful on the field and it can be detrimental. You can't do more push ups than you're capable of doing though. Its just all the other times you didn't do all you can do.
Adrenaline can absolutely allow you to do more push ups than you can normally do, it can also raise awareness, vision, and ability to deal with pain.
So looking at my list of QBs who have 100+ starts in theNFL here is the list of the top 11, which is everyone with a 60%+ win%. Tom Brady 69.2% Peyton Manning 65.8% Y A Tittle 65.0% Jim Kelly 64.3% Jack Kemp 63.8% Johnny Unitas 63.8% Neil O'Donnel 63.5% George Blanda 62.0% Roger Staubach 61.2% Ken Stabler 61.1% Earl Morral 61.0% The only surprising name there is Neil O'Donnell. And even then he had the advantage of (a) a HoF cailber coach in Bill Cowher and (b) elite defenses, both of which were shown in my initial post to help a team's win% in close games. The bottom 11 Steve De Berg 44.8% Jim Plunkett 44.1% Drew Bledsoe 43.8% Lynn Dickey 42.9% Ryan Fitzpatrick 42.5% Chris Chandler 42.4% Steve Bartowski 41.3% Jim Everett 40.9% Steve Buerlein 40.0% Norm Snead 39.3% Archie Manning 33.6% Ryan Fitzpatrick's number probably explain why the Jets haven't been knocking his door down this off season. Plunkett only won a SB after being linked to one of the great coaches with a great team. The names though are a group that have had their fans tearing their hair out. If winning 0-7 point games was purely random distribution you wouldn't see such a strong split between QBs defined as 'great' and 'disappointing'. An amusing sidenote. The Manning family average 50% in 0-7 point games.
So you're not going to take into account experience one gathers as they progress in their career? Come on you can't be serious. You're trying to provide equal weight to a guy's first and second year in the league vs. his most recent years. If you have three QBs, both with 4 years of experience. QB one has been a steady 85 rating QB each year. QB two was 75, 75, then 95, 95. QB three was 95, 95, 75, 75. Lets assume the stats were comparable enough so both QBs ended up with similar career ratings of 85 (because you can't just average out ratings and need to look at underlying stats). It's not cherry picking to point out QB two's latest two years has been the best and is the best QB right now. You place the most weight in the most recent years because A) it's most recent and B) they are now more experienced. The hot streak argument doesn't hold because these stats were accumulated over several playoffs spread over several seasons with a bunch of time between each season (not to mention regular season games). In fact, your argument of a hot streak argues for clutch because his hot streak takes a break during the regular season and kicks in during the playoffs!
Talking about in general, not specifically cruch time football. But everyone has a baseline max capability. At times of extreme stress, our bodies are capable of unlocking extra strength/capability otherwise impossible. So in a sense, yes, if our body can do it, it can do it. But this capability isn't always available to us, no matter how hard you try. Sometimes takes a life or death situation to do so. Again, not talking about football but in general. It's the mom who fights of a 700lb polar bear, or lifts a 3,500 car off their son.
Because these athletes train specifically for those two things? I don't really follow track and field, or swimming, but in the Olympics the best in the world are competing, and they've been training for years for this stage. Yes there are other events during the years, but they're training. They're working on technique. I'm not sure your stat is really all that meaningful.
Practice is practice, events are events, those same best in the world participate in many events, but their best efforts are often on the biggest stage.
Is there a statistical analysis that shows records are broken at a greater rate in the Olympics due it being the big stage?
Well, at least for many swimming events it seems not to be true that most world records are set at the Olympics or World Championships. Start with this link and click on any of the races at the top to give you the progression of world records for the event: https://en.wikipedia.org/wiki/List_of_world_records_in_swimming For example, for the 100 m: https://en.wikipedia.org/wiki/World_record_progression_100_metres_freestyle For most of the individual events, it seems the majority of records are set outside the Olympics or World Championships/World Cup. Where they are mostly set at the World Cup or World Championship are the team events.
Yes all of that is true. But, I just don't know if it actually applies to the discussion. I mean is playing QB to get into the playoffs the same thing as the mom lifting a car off her kid? And even if it is, are those things that adrenaline heightens really everything a QB needs to succeed? And then even if you convince me of all that.......we're still left with the simple fact that a QB can do everything he's supposed to from call the right play, evade the rush, find the open man, deliver the ball accurately with the perfect velocity, and the WR drops it. Or the WR catches it, they score and then the defense al;lows the other team to march down the field and win it.
Middle distance runners are notorious for running slow times at the Olympics. Everyone tries to run a tactical race to prevent themselves from being slipstreamed or having another runner make a move they don't t see happening. I think the last men's 1500m world record set at the Olympics was in 1960 or something,
Again, you've concluded that the poor play early in his career was due to lack of experience. That isn't necessarily true and certainly hasn't been proven. It may just have been a set of bad games. And that's why you can't exclude them. It's not just about "hot streaks" in a sequential way. With stats and small sample sizes you see all kinds of performance anomalies that don't correlate with anything meaningful. For example, using Flacco, his passer rating against the NFC South is 117.9. That's almost 30 points higher than against any other division and in games spread out across his career. So what is your explanation for that? He has superpowers south of the Mason-Dixon line, but only in NFC cities? It's an anomaly with no correlation to anything meaningful and no logical causation. It happens.
From eyeballing those numbers, the guys with the high win% in close games also have similarly high win% in all games and non-close games. And the converse is true of the guys with low win%. Tom Brady wins 69.2% of his close games and 77.1% of his overall games. So his win% in non-close games is better than in close games. By the logic of this thread, he is not clutch because he loses close games at a higher rate than he does "regular" games. Peyton Manning won 65.8% of his close games and 70.2% of his overall games. So his win% in non-close games is better than in close games. By the logic of this thread, he is not clutch because he loses close games at a higher rate than he does "regular" games. Etc. Archie Manning won 33.6% of his close games and 25.1% of his non-close games. So he was "better" in close games than non-close games, according to the logic of this thread. So he is apparently clutch. Etc. It looks like QBs start to move toward the mean (50%) in close game win%. That makes perfect sense because those games are close and so as they get late in such games, the better team has less time show and exert its superiority, and vice-versa. I'd bet that if one ran a correlation of overall win% and close game win% they would correlate very closely. For guys with good win% overall the close game win% would be 8-15% (not percentage points) lower and for guys with low overall win% the close game win% would be 8-15% higher.
Marino was clutch, you just knew no matter what the score somehow someway he was going to find a way to win in the 4th quarter.
I dunno but I know when I have adrenaline pumping through my body (car accident, pulled over by a cop when I have something illegal in the car ... ooops) my body shakes uncontrollably even when I'm not cold. Some QBs can handle this more than others I guess is the point. On top of the adrenaline would be nerves. I mean, Nate Kaeding was like the 2nd most accurate kicker of all time at one point but is 3 of 8, in the playoffs, at home. Just playing to his normal level is clutch (playoffs being extra stress). Yin doesn't exist without yang. Choking doesn't exist without clutch. We can debate to what degree clutch helps (say, 1/2 the league is clutch, and only a few are chokers) but it exists. It has to. Nate Kaeding.
Because it's such a brand new and untested hypothesis that rookies generally play poor, and players generally get better with more experience. First two years, played like crap. Last FOUR years, out of this world. Draw whatever conclusion you want to. Those are the facts. You go ahead and hang onto those first two years to push your argument, I'll leave with this. His stats are amazing considering he's only started two or so playoff games at home. https://www.washingtonpost.com/news...playoff-performance-has-him-among-nfls-elite/
My conclusion is that it was a run of good playoff games. Much like his run of 9 great games in 2010. Or his run of great games against the NFC South. Those runs sometimes just happen. You've given no explanation for those other great runs because there is none. And there may be none for the run of great playoff games. You want to believe it's clutch, but that doesn't make it so. And again, it's not like he gradually improved in the playoffs -- his "improvement" seemed to have happened over night. His first 4-5 playoff games were a total sh!tshow. Then suddenly he became clutch? I joked about the Clutchness Fairy sprinkling him with fairy dust, but is that your explanation? Do you have some other explanation for how he went from awful to awesome almost instantaneously?
I still play baseball, and I know there is a totally different feeling batting with no one on, and batting with someone on 3rd and 2 outs. those guys that can erase that different feeling are the clutch players.
