Carroll said "doubting fans need to “do their homework”

[DavidSeven]

Are we allowed to be happy about how the team is performing overall in spite of numerous key injuries while conceding there are areas of possible improvement like there are with every NFL team?

Be fair. I logged onto .NET and the threads up front were :Would we have won if Locker played? Judging on how we played, can we win on Thursday? Disapointing subpar performance today.

The worst part? Some of those same threads were in the game day forum, which is where emo reactions belong. But some felt the need to have the thread in two places.

Then the be happy crowd weighed in. I don't believe in being 5-1 any more than I believe the team is in trouble even if we don't get home field advantage. Frankly, I don't believe in sweeping generalities. Beastquake pisses on your generalities.

Everyone is entitled to an opinion. But if your opinion is not rooted in fact, I expect it to be destroyed as it should be. I am ecstatic about the win. I also think with about 3 different choices on offense, the score is 31-6. Toss in the pick 6 Thurmond couldn't get his hands on, and we are talking blowout.

The ghost of 150, the fear we are not good enough to get HFA, and dummies betting on the over is stuff beyond my control. I try to stick to what I actually saw.

Agreed. I was just trying to put the reverse spin on the "are we allowed to criticize this team without getting blasted" comments. At the end of the day, everyone is entitled to their opinions. But as you pointed out, a lot of the threads that popped up on Sunday seemed to be filled with negative generalities and "sky is falling" type sentiment. That's the stuff I can't relate to. Not saying people have to put on a happy-face no matter what, but I predicted all of these last three games to be close before they happened, so nothing I saw surprised me (besides the fluky goofiness). In fact, I thought Seattle moved the ball even better than expected against TEN.

 

[HansGruber]

All it takes is one big play to throw off ToP. How many NFL games each week have at least one big play? What, all of them? Oh, gee.

Hans, some weeks also have MOST winning teams losing ToP. Look up each week for an entire year, if you want.

OK, since I find the topic interesting from a purely scientific standpoint, I will look it up and compile the stats.

However, from my experience extrapolating small tests into larger ones, generally speaking, trends tend to carry across larger data sets. Sure, it won't be 66%, but it is quite unlikely that less than 50% of the TOP winners have won the game.

BRB with the official stats...

 

[Smelly McUgly]

You'll get more emotional posting right after a game where emotions were high. I was way more pessimistic about the game right as it concluded, and I bet if I tracked my posts, I became more pleased with not only the outcome, but with how the team got there as I got out of the moment and was able to view the game with a more critical eye.

 

[kearly]

Regarding TOP, it's only an indicator and not a rock solid one. For example, teams that run hurry up, no huddle offenses wear out defenses much quicker than traditional offenses do, despite being at a disadvantage in TOP. The more meaningful indicator is total plays by each team. Burning off more clock between plays doesn't really impact performance, though hurrying to the line does.

I am also a bit skeptical that hurry up teams put their defenses at a rest disadvantage. Oregon's defense hasn't shown any signs of fatigue. Neither has Washington's. Philly's defense sucks, but I think that's just because they suck and not because of rest. The difference between a hurry up team and a traditional team is maybe 15 seconds of rest per play. If it's a typical 8 play drive, it's only a 2 minute difference in rest. You include the two TV timeouts, and it's probably the difference between 8 minutes of rest vs. 10 minutes. Nothing earth shattering.

The only real benefit of TOP is that teams that control it usually control the running game, and this allows them to shorten the game. Shorter game = fewer plays = advantage defense. So teams that are defense first like ours prefer to use up as much clock as possible.

 

[Sac]

All it takes is one big play to throw off ToP. How many NFL games each week have at least one big play? What, all of them? Oh, gee.

Hans, some weeks also have MOST winning teams losing ToP. Look up each week for an entire year, if you want.

OK, since I find the topic interesting from a purely scientific standpoint, I will look it up and compile the stats.

However, from my experience extrapolating small tests into larger ones, generally speaking, trends tend to carry across larger data sets. Sure, it won't be 66%, but it is quite unlikely that less than 50% of the TOP winners have won the game.

BRB with the official stats...

The reason TOP is meaningless is because you can't point to it and say "we should do more/less/better of that". If we possess the ball for 40 minutes, you can infer that we kicked butt. But the butt kicking isn't caused by merely possessing the ball. Possessing the ball for 40 minutes is the result of kicking butt.

 

[Largent80]

When the goal is to simply win, T.O.P. is an important part because as we have seen in years past, the defense should not be playing as much as the offense mainly because the opposing offense knows exactly what they are going to do.

If your defense is playing well, they are giving the offense the chance to control the TOP.

If the offense is 3 and out, the defense is sucking air late in the game. In Indy, we had their D gassed, and allowed a blocked FG that went for a td, which lost us the game.

That had nothing to do with TOP, but what it did show was a lack of offense that forced a long FG try, which resulted in a lower trajectory of the ball, hence the block.

Winning the TOP usually means winning the game.

 

[Laloosh]

You'll get more emotional posting right after a game where emotions were high. I was way more pessimistic about the game right as it concluded, and I bet if I tracked my posts, I became more pleased with not only the outcome, but with how the team got there as I got out of the moment and was able to view the game with a more critical eye.

This is pretty accurate for many I'd imagine. I wasn't posting after the game but I'd probably have been pretty "concerned" if I were.

It's ironic because I've been a lot less "stressed" in our earlier games where we were down by at least a TD because I know the team doesn't freak out about being behind going into the half. This game though... just didn't know what to think because that damn ball just kept squirting out. I felt more at ease in the 49er game being up by 5 than I did being up by 7 against the Titans.

Wrong or right, it's where I was at. Having watched the game a few times now, I'm looking forward to seeing this offense running on all cylinders and I was really happy to see Bevell try to adapt to what defenses are giving him (opportunities in the short passing game and space for Russell to run).

 

[HansGruber]

Wow. Was not expecting the results I am seeing:

In Week 1, out of 16 games, 13 were won by the team with best time of possession. Which equates to a stunning 81%.

So far, the "TOP is useless as a predictor" theory is being blown out of the water. So far, I have analyzed 3 weeks of the 2013 season, and the lowest percentage of weekly TOP winners is 66%. In every other week, more than 66% of the teams with the best time of possession won the game.

Will be back with full 2013 stats in another minute.

 

[MOCHawk]

It's not that it is worthless as a predictor (strange to call it a predictor when it is observed after the game), or that it is irrelevant (which nobody said), it is that it is overrated as a statistic. Trying to dominate TOP is one way to go. Running lots of plays is another. There are many more.

Looking at the end result of a game and saying "see, the team that possessed it longer won!" Of course they did, they were ahead so they ran the ball. And the other team was behind, so they threw it.

It's like the useless statistics that the idiots always throw out: "The Cowboys (or any other team) lose the game 90% of the time when they throw for 400+ yards." They didn't lose because they threw the ball, they threw the ball because they were losing. The corollary is X team wins X% of the time when they rush for more than 100 yards. Of course they do. They rushed the ball because they were winning.

 

[kearly]

15th in RZ TDs
28th in 3rd-down pct.
29th in total passing yards
18th in total rushing yards

Just so you know, Seattle is 2nd in the NFL in rushing yards per game, behind only Philly. (Random note: Green Bay is sitting at 5th place. Wow.)

 

[Smelly McUgly]

15th in RZ TDs
28th in 3rd-down pct.
29th in total passing yards
18th in total rushing yards

Just so you know, Seattle is 2nd in the NFL in rushing yards per game, behind only Philly. (Random note: Green Bay is sitting at 5th place. Wow.)

Oh, I know. People got on me (fairly enough) for that one. I think I just clicked the categories in ESPN's virtual spreadsheet incorrectly or something.

 

[Laloosh]

It's not that it is worthless as a predictor (strange to call it a predictor when it is observed after the game), or that it is irrelevant (which nobody said), it is that it is overrated as a statistic. Trying to dominate TOP is one way to go. Running lots of plays is another. There are many more.

Looking at the end result of a game and saying "see, the team that possessed it longer won!" Of course they did, they were ahead so they ran the ball. And the other team was behind, so they threw it.

It's like the useless statistics that the idiots always throw out: "The Cowboys (or any other team) lose the game 90% of the time when they throw for 400+ yards." They didn't lose because they threw the ball, they threw the ball because they were losing. The corollary is X team wins X% of the time when they rush for more than 100 yards. Of course they do. They rushed the ball because they were winning.

That said, I'm sure there are plenty of other stats that would correlate to wins as often as TOP. First downs, 3rd down conversion, etc. If you're getting either of those more often, you likely have a higher TOP and more importantly, more points

Hans, I think a comparison of some of those stats to TOP as they relate to win % might be more interesting than simply determining how often TOP indicates a winner.

 

[HansGruber]

OK I have final stats for the 2013 season, and they are surprisingly strong.

First, here is how each week fared in wins/losses for the best TOP (wins shown on left):

Week 1 = 13/3 (81%)
Week 2 = 11/5 (69%)
Week 3 = 11/5 (69%)
Week 4 = 11/4 (73%)
Week 5 = 10/5 (66%)
Week 6 = 10/5 (66%)

2103 Season = 66/27 (71%)

So, in the 2013 season, 71% of the teams who have won the time of possession battle have won the game. In the medical/scientific fields, this would be considered an "extremely strong indicator".

Arguments to the contrary must rely on outliers and statistical anomalies. This is supported by the nature of those games in which the team with the lowest time of possession won. In almost every case, wins were either the result of:
1) a major blowout
2) multiple scores on defense or special teams

Extremely few games have been won where the team lost the time of possession battle, did not blowout the opponent (by more than 14 points), and did not score on defense/special-teams at least twice. In every case where one of these factors was not present, the time of possession was extremely close.

Therefore, it is factually true that teams who have the highest time of possession in a game will indeed win the game in a great majority of instances (71% of all games played).

 

[themunn]

ToP on it's own is a totally useless stat, just like any other stat.

Pete talks about winning the ToP battle and the turnover battle.

In 2012, 11 teams had more than 30 minutes ToP and +ve turnover differential. 8 of those teams made the playoffs. Seattle and Cincinatti the only two of the 8 that didn't win their division.

The three that didn't? Chicago, Tampa and San Diego.

Chicago went 7-1 and had no right self-destructing the way they did, and Tampa and San Diego barely won the TO battle (+3 and +2 respectively).

The 4 teams that made the playoffs without winning the battle? Denver (-1 TO), Minnesota (-1 TO) Indianapolis (-12 TO, but also without doubt the worst team to make the playoffs last year), and Baltimore (28.40 ToP).
Now you might look at that and think "hang on though, Baltimore won the Superbowl", but Baltimore were dodgy throughout the year, and when it came to the playoffs, sorted it out - winning the ToP battle and finishing with a TO ratio of +6 - which was more than every other team with a +ve turnover ratio combined.

 

[Largent80]

Turnovers have a dramatic affect on TOP. So, even if a team is dominating another, a pair of turnovers skews all of it.

 

[HansGruber]

ToP on it's own is a totally useless stat, just like any other stat.

I have just proven factually that this is a false myth. In fact, the proof states quite clearly that time of possession is an extremely strong indicator of success, and success without winning time of possession would be a statistical anomaly involving one of two factors:

1) a major blowout
2) multiple scores on defense and/or special teams

Your argument to the contrary is subjective/philosophical (not based on fact). Therefore, nobody can reply with any legitimate counterpoint, because there are none.

 

[HansGruber]

Turnovers have a dramatic affect on TOP. So, even if a team is dominating another, a pair of turnovers skews all of it.

Your theory is accurate.

Going over the games just now, there is a clear pattern that winners with lower TOP were significantly caused by multiple scores on defense and/or special teams. Casually, I noticed that many were caused by turnovers leading to scores.

 

[Smelly McUgly]

ToP on it's own is a totally useless stat, just like any other stat.

I have just proven factually that this is a false myth.

No, you haven't. You have looked at six weeks' worth of data. SSS.

On another note, so to win, all we have to do is get better at TOP. How do we do that?

 

[HansGruber]

ToP on it's own is a totally useless stat, just like any other stat.

Pete talks about winning the ToP battle and the turnover battle.

In 2012, 11 teams had more than 30 minutes ToP and +ve turnover differential. 8 of those teams made the playoffs. Seattle and Cincinatti the only two of the 8 that didn't win their division.

The three that didn't? Chicago, Tampa and San Diego.

Chicago went 7-1 and had no right self-destructing the way they did, and Tampa and San Diego barely won the TO battle (+3 and +2 respectively).

The 4 teams that made the playoffs without winning the battle? Denver (-1 TO), Minnesota (-1 TO) Indianapolis (-12 TO, but also without doubt the worst team to make the playoffs last year), and Baltimore (28.40 ToP).
Now you might look at that and think "hang on though, Baltimore won the Superbowl", but Baltimore were dodgy throughout the year, and when it came to the playoffs, sorted it out - winning the ToP battle and finishing with a TO ratio of +6 - which was more than every other team with a +ve turnover ratio combined.

You are analyzing the data set improperly. We are talking about indicators of individual game success, not full season success. If you wish to analyze a unit-specific dynamic, you must isolate your data to the scope of the unit. In this case, per-game.

Comparing per-season mean averages will never give you an accurate picture of per-game success. Just as measuring the average temperature of a planet's surface will never give you an accurate temperature for a single region.

 

[RolandDeschain]

There's a correlation nowhere near as easy to see between large positive plays on offense as well, Hans.

However, I think you're drastically overstating the case. Flipping a coin will predict the correct winner 50% of the time. ToP does it 71% of the time you charted it, that's hardly a "great majority". Majority, certainly, but "great"? Not even close. It's also not a stat that has any predictive power at all. You can win the ToP battle 10 games in a row while having gone 5-5 over that stretch and still not make any accurate predictive correlation based on it for the next game that team has to play. Other stats like turnover differential, QB rating, TD-int ratio, yards per carry in the running game, etc. are all different from time of possession in that sense.