|
The AFC South Division Champion of 2019 will be... This poll is closed. |
|||
|---|---|---|---|
| Houtson |
|
29 | 31.18% |
| Indianapolis |
|
25 | 26.88% |
| Jacksonville |
|
22 | 23.66% |
| Tennessee |
|
17 | 18.28% |
| Total: | 93 votes | ||
|
Metapod posted:Clowney soon plz
|
#
?
Jul 28, 2020 07:51
|
|
|
|
| # ? May 3, 2024 06:23 |
|
|
Get well soon Isaiah
|
|
#
?
Jul 28, 2020 19:01
|
|
|
The graphic seems odd for this type of announcement https://twitter.com/Jaguars/status/1288202569857355777
|
|
#
?
Jul 28, 2020 21:29
|
|
|
Intruder posted:The graphic seems odd for this type of announcement Yeah, it needs Jock Jams playing over it.
|
|
#
?
Jul 28, 2020 21:47
|
|
|
DJ Chark doo doo doo doo doo doo
|
|
#
?
Jul 31, 2020 17:53
|
|
|
Mr. Nice! posted:The titans last year were a team that won by rushing. It is very possible to have a rush attack that is strong enough you can still have a lot of high scoring dominant wins where the qb has ~150 yards and maybe 1 passing touchdown. Those numbers might be a bit deceptive. quote:There’s no doubt that the Titans were a run-heavy team. But Tannehill produced 0.25 EPA per dropback on early downs, which ranked first in the NFL. The Titans were good (10th) running the ball on early downs, but they were among the best in the league passing the ball on early downs. With respect to the playoff game against New England: https://twitter.com/benbbaldwin/status/1213899975026970625 Henry was also only 8th in EPA/rush in 2019. And don't forget, The Titans squeezed into the Playoffs at 9-7. and 5 of those wins came against the Browns, Colts, Jaguars, Bucs and Raiders. On that run of wins after switching to Tannehill, he completed 70% of his passes, threw 22 TDs to 6 Ints, and posted the best Air Yards per throw in the NFL. In the Playoffs, their defense held New England and Baltimore to 13 and 12 points respectively. Its not accurate enough to say they "won by rushing." Its more accurate to say they won by having a very efficient passing attack, holding opposing offenses to few points, and by running the ball well situationally. In every one of those winning efforts, Tannehill posted a QBR of nearly 110 and higher, threw for at least 2 touchdowns, and completed nearly 75% of his passes. The fact that he wasn't very good against New England but they won, and was very good against New Orleans but they lost are the outliers with respect to "how" Tennessee won, not the template. If you could (and you can't, but you kind of can with WPA and EPA) make a pie chart of the relative percentage of factors which contributed to the Titans victories in 2019, Ryan Tannehill's performance would consume the majority of the pie chart, and a much larger portion of the chart than Derrick Henry's portion. Metapod posted:The titans go against a lot of stacked boxes because of how good Henry is. This allowed the vertical pass attack they use to work in a more effective manner. This is because of the 1v1 matchups stacked boxes create combine this with playaction causing hesitation to the defense you have an advantage on your pass because of Henry. So statistics that show perfomance against # of Defenders in the Box is hidden behind NFLScrapR; but you can still look at situational passing on plays where defenses are expecting a pass over a run, and see how he performed. In literally every situational context, down, distance, quarter, margin of victory, etc. Tannehill had an exceptional QBR of 100+. In situations where Tennesee was losing by a substantial margin, 9 - 16 points, where the defense would be trying to stop the pass for sure, his rating was 108. When playing from behind in any situation, it was 114. And in the 4th quarter, with the game on the line, his rating was an astounding 144. In pure, and obvious passing situations, Tannehill was still excellent, without the help of the threat of the run. More importantly, however, he was generally just as good across all situations which means you can't point to clear rushing situations (and thus, the threat of Henry) or clear passing situations as something that impacted how successful he was.
|
|
#
?
Jul 31, 2020 18:48
|
|
|
Metapod posted:I dont think you understand how good henry is or how much of the offense revolves around him. The titans were able to score in 2 minute drills just by giving him the ball lol Henry only carried the ball 11 times all season in the last 2 minutes of any half (the 2 minute drill). He had 0 touchdowns.
|
|
#
?
Jul 31, 2020 18:58
|
|
|
I'm referring to the playoffs op
|
|
#
?
Jul 31, 2020 18:59
|
|
|
I was content having an elite RB, but if you’re saying we also have an elite QB, I guess I’ll allow that as well.
|
|
#
?
Jul 31, 2020 19:07
|
|
|
Metapod posted:I'm referring to the playoffs op blarzgh posted:https://twitter.com/benbbaldwin/status/1213899975026970625
|
|
#
?
Jul 31, 2020 19:21
|
|
|
Yeah you posted that before but it seems to take no consideration in the fact that one player (derrick henry) was good that game and the other player (tannehill) was awful and was actively helping the Patriots
|
|
#
?
Jul 31, 2020 19:26
|
|
|
Luckily Brady was helping the Titans so it all worked out in the end.
|
|
#
?
Jul 31, 2020 19:27
|
|
|
Mega64 posted:I was content having an elite RB, but if you’re saying we also have an elite QB, I guess I’ll allow that as well.
|
|
#
?
Jul 31, 2020 19:27
|
|
|
Mega64 posted:Luckily Brady was helping the Titans so it all worked out in the end. 🙏🙏🙏🙏
|
|
#
?
Jul 31, 2020 19:28
|
|
|
Metapod posted:Yeah you posted that before but it seems to take no consideration in the fact that one player (derrick henry) was good that game and the other player (tannehill) was awful and was actively helping the Patriots It was nice for Tom Brady to actively help the titans by throwing a pick six on his last pass as a new englahahahahahaha
|
|
#
?
Jul 31, 2020 19:28
|
|
|
Blarzgh is just saying to print the shirts.
|
|
#
?
Jul 31, 2020 19:28
|
|
|
Metapod posted:Yeah you posted that before but it seems to take no consideration in the fact that one player (derrick henry) was good that game and the other player (tannehill) was awful and was actively helping the Patriots quote:Expected Points Added was created to address the misleading nature of yards gained by putting value on down and distance as well as field position. For example, a gain of five yards on a 3rd down and 10, and a gain of five yards on 3rd down and 3, will both register as +5 in yards gained. However, the five yards gained with three yards to go is more valuable than the five yards gained with 10 yards to go, as it results in a first down and extends the drive. So, again: https://twitter.com/benbbaldwin/status/1213899975026970625
|
|
#
?
Jul 31, 2020 19:30
|
|
|
If a stat saya a player who threw an interception and fumbled twice in 21 attempts was better than the guy who ran for almost 200 yards than your stat is bad lol
|
|
#
?
Jul 31, 2020 19:33
|
|
|
gently caress I forgot he didn't even throw for a 100 that game lmao
|
|
#
?
Jul 31, 2020 19:34
|
|
|
Metapod posted:If a stat saya a player who threw an interception and fumbled twice in 21 attempts was better than the guy who ran for almost 200 yards than your stat is bad lol 2 questions: 1) Is running for 4 yards on 3rd and 3 better than running for 5 yards on 1st and 15? 2) Is passing for 8 yards inside the opponent's 40 yard line better than passing for 10 yards from your own 20 yard line?
|
|
#
?
Jul 31, 2020 19:40
|
|
|
I think any stat that's trying to compare across positions is not going to be good because you can't compare a qb to a hb or a hb to a wr they are all very distinct positions. I also believe that it doesn't take a formula to show that a hb rushing for 180 yards helps a team win more than a qb passing for 84 yards
|
|
#
?
Jul 31, 2020 19:57
|
|
|
Yeah, generic "wins above replacement" type stats like that are always gonna be questionable in football considering how interlocking the team game is
|
|
#
?
Jul 31, 2020 19:59
|
|
|
Yeah having some kind of metric of scale like max/min/mean/median/st.dev of Expected Points per position is more useful than comparing across positions, as you could probably easily find evidence to suggest that QB and RB come from different populations/distributions
|
|
#
?
Jul 31, 2020 20:00
|
|
|
 I've been rewatching the Titans/Patriots playoff game today because Titans Bitch, and Tannehill has a great first drive (partly due to the threat of Henry but Tannehill delivered nonetheless) but then Henry is basically the threat that keeps the New England defense honest. Hell, the second Titans touchdown was 75 yards of pure Henry (with a 22-yard Tannehill screen pass albeit). Part of it was the NE defense getting to Tannehill constantly, but it's clear they had much less of an answer to Henry than for Tannehill. Of course, it also helps that the Tennessee defense stepped the hell up that game, especially in that goal line stand in the second quarter and being capped off with Tom Brady's Final Pass.
|
|
#
?
Jul 31, 2020 21:25
|
|
|
For those interested, EPA doesn't, in and of itself, separate players into positions, or even types of plays. What EPA does, is look at the starting time, field position, down and distance, and then the resulting time, field position, down and distance, and quantify the difference between those two things based on how much more, or less likely a team is to score points after the change. quote:Suppose the offense has a 1st and 10 at midfield. This situation is worth +2.0 EP. A 5-yard gain would set up a 2nd and 5 from the 45, which corresponds to a +2.1 EP. Therefore, that 5-yard gain in that particular situation represents a +0.1 gain in EP. This gain is called Expected Points Added (EPA). Likewise, a 5-yard loss on 1st down at midfield would create a 2nd and 15 from the offense’s own 45. That situation is worth +1.2 EP, representing a net difference of -0.8 EPA. So what EPA does is divorce itself entirely from the type of play, formation, everything, and just gives us the net value of a play. Then, anyone with the compunction to analyze these plays can sort them into categories and see if any useful information pops out. It only creates an objective, baseline measurement for the points value of any given play. Providing context is up to the user. One might say, "A swing pass on 1st and 10 for 5 yards has the same total value as a RB Draw for 5 yards." EPA will agree with that statement. A shrewd observer might then say, "Ah, but the rushing play will alter the way the Defense lines up and reacts to plays later on in the game, which will make passing easier later on." Which may very well be true! EPA doesn't care about that. However, the neat thing is that you can then go check the value of future plays where running the ball early would have, in theory, made passing easier later in the game, using EPA (and other analysis)! For example, it was long believed that rushing the ball early and often was necessary to set up Play Action later on in games. Turns out, this isn't true. Teams enjoy success on Play Action whether or not they run the ball successfully earlier. Here is some more research on maximizing Play Action ratios. It also turns out that teams aren't maximizing the efficiency of their third down and red zone running games, and that certain schemes are producing better results. And still there are more findings out there that teams aren't actually running the ball enough on 3rd and 1, and in the red zone. So there are lots of uses for EPA; but like any statistic, it is only a tool that does one specific thing. It doesn't have an agenda, or context, it only tells you how much more likely a team is to get points after a specific play.
|
|
#
?
Jul 31, 2020 22:00
|
|
|
yes you are trying to use a stat incorrectly i agree with this
|
|
#
?
Jul 31, 2020 22:07
|
|
|
A datum is only useful in comparing to other data in its population, or suggesting that populations are different. Football is a game with asymmetric player values in terms of what they can contribute to the goal of winning the game. If I were motivated, I could pull data and perform a test that would likely confirm that QB and RB values are different, and even specifically set it up to gauge direction. I'm not, so I'm just assuming that QBs will have higher values on average than RBs due to domain knowledge. Tannehill having a raw value higher than Henry's raw value is only significant in terms of team composition and the constraints it presents. A more enlightening question might be how was Tannehill's performance compared to average QB performance, along with how Henry's performance compares to the average RB performance. If Tannehill was say 1 St.Dev below average while Henry was 2 St.Dev above average, I would argue that Henry had the better performance.
|
|
#
?
Jul 31, 2020 22:39
|
|
|
Shadow225 posted:A datum is only useful in comparing to other data in its population, or suggesting that populations are different. The only question EPA answers is, "What changes in field position make scoring points more or less likely?" The question of, "Who had a better performance?" is, I think, too subjective. If you set the parameters as you did, where the question of who had a better performance is determined by how far above/below average did the player perform, then you get weird results. The 1:1 comparison of this line of thinking is, "How was the Punter's performance compared to average Punter performance, along with how Tannehill's performance compares to the average QB performance. If the Punter was say 1 St.Dev below average while Tannehill was 2 St.Dev above average, I would argue that Tannehill had the better performance." Its a valid question, and a scientific way to get an answer. But the result there could inversely be that the Punter might have had a "better performance" but everyone would probably still agree that Tannehill's performance had more to do with the team winning the game. When Ben Baldwin says, "Tannehill had a higher total EPA than Henry." the only thing he is saying is that, when combined, all of Tannehills plays increased the Titan's potential to score points more than all of Henry's plays combined did.
|
|
#
?
Jul 31, 2020 22:55
|
|
|
No Brett kern absolutely had more to do with the win than tannehill in that game
|
|
#
?
Jul 31, 2020 22:58
|
|
|
The Colts once lost a playoff game to the Chargers almost entirely due to Darren Sproles being Darren Sproles and Mike Scifres having one of the greatest punting games ever. Punters can be cool and good.
|
|
#
?
Jul 31, 2020 23:31
|
|
|
JT Jag posted:Yeah, generic "wins above replacement" type stats like that are always gonna be questionable in football considering how interlocking the team game is ^^^It's this^^^ Thank you. I enjoy blarzgh's posts and the fun he has with numbers, which are often relevant. But people have been trying to to the sabermetric approach to football for decades now it just doesn't work that way due to the 22 moving parts on every loving play. In baseball, it's easier to isolate individual contributions, primarily due to the one on one pitcher versus batter match up and the fact that a batted ball heads towards, at most, TWO fielders, assuming it's not knocked over the fence. Football is more the sum of it's parts. The ultimate team game. One player can gently caress up a perfectly called play and it's almost impossible to figure out whose fault that was. A missed block by the RG, a wrong route run by the slot receiver, a HB late releasing on a swing pass or who missed a blitz pickup, a QB that read the weak side blitz wrong and slid protection the wrong way. As much as I'm a stats nerd and get into that sort of poo poo (I like dice and card football games), blarzgh's approach to football analysis just doesn't really apply the way he thinks it does, no matter how good it makes Dak Prescott look, nor do any of those board games quite entirely work the way the baseball ones do, for the reasons I cited. For instance, how great was Emmit Smith versus how great was his offensive line? How great was Eric Dickerson or Barry Sanders independent of their blocking and the stacked boxes they faced and the terrible QB's they played with? Not to say we shouldn't TRY mind you, but measuring this stuff like we do baseball is WAY behind what the dudes who calculate that poo poo do and might even be a fool's game. Even video games teach us this. What would Dak Prescott's career look like if he'd played for GB or ATL or SEA? NE under Belichick? You can't know. Some people (rightly) compare football to chess, except in chess, all the pieces aren't moving at the same time and also the individual pieces are beholden to the skill of the person moving them.
|
|
#
?
Aug 1, 2020 00:04
|
|
|
If data shows players I like are good or players I don’t like are bad then it’s good data.
|
|
#
?
Aug 1, 2020 00:40
|
|
|
Huh, I saw a bunch of new posts and thought something happened, not that there was a big debate on who is more valuable between Derrick Henry and Ryan Tannehill DJ Chark doo doo doo doo
|
|
#
?
Aug 1, 2020 00:41
|
|
|
Allow me to write a few essays that explain how this stat is correct and you're all blind fools for not seeing that the QB who passed for less than 100 yards was more instrumental to a team's win than the RB who ran for 200 yards.
|
|
#
?
Aug 1, 2020 01:05
|
|
|
Ryan Tannehill is a shirt worthy QB.
|
|
#
?
Aug 1, 2020 04:35
|
|
|
What if Dak Prescott was made of delicious bbq ribs?
|
|
#
?
Aug 1, 2020 22:49
|
|
|
Quiet Feet posted:What if Dak Prescott was made of delicious bbq ribs? Not sure how this would work but can his team eat him during the game?
|
|
#
?
Aug 2, 2020 01:53
|
|
|
wrong thread
Raku fucked around with this message at 21:27 on Aug 2, 2020 |
|
#
?
Aug 2, 2020 21:18
|
|
|
lol Florida. now you're trying to kill Minshew.
|
|
#
?
Aug 2, 2020 22:02
|
|
|
|
| # ? May 3, 2024 06:23 |
|
|
That’s not fair Florida is trying to kill a lot of people
|
|
#
?
Aug 2, 2020 22:28
|
|
