Spikes injured?

[Deus Irae]

I'm not saying losing Spikes was a positive, I'm saying it has almost no impact statistically.
Obviously, if Belichick put him on the field, it was because he thought Spikes was the best option. And I'm pretty sure Belichick knows more about football than I do, so losing Spikes had to be a negative.

But that being said, I'm not looking at things in a vacuum. There's only 16 games in a football season, so 4 games is a good sample. In fact, you are the one who doesn't try to put things in perspective as you are looking only at the Packers game as proof that the suspension to Spikes was a setback to the defense. The fact is that the defense played their 3 best games WITHOUT Spikes. Yes, other factors need to be taken into consideration, but this is a fact, no way around it. Could Spikes have contributed positively in these 3 games ? Maybe...but based on the data from last season, he was not an impact player, so his loss had a minimal effect on the defense.

You're ignoring other factors when making your analysis, and Rob's trying not to do that. As for the defense playing its 3 best games without Spikes, that's simply a matter of opinion. I would argue that the 45-3 Jets game, with Spikes, was the defense's best game, because that game turned into quicksand for the Jets and it became party time for New England on both sides of the ball.

 

[Sicilian]

You're ignoring other factors when making your analysis, and Rob's trying not to do that. As for the defense playing its 3 best games without Spikes, that's simply a matter of opinion. I would argue that the 45-3 Jets game, with Spikes, was the defense's best game, because that game turned into quicksand for the Jets and it became party time for New England on both sides of the ball.

Not only did Spikes play in that game, but he had arguably the most important play on defense. I might be foggy on the details, but weren't the Jets driving for a tying TD (or at least a TD to bring them back within a score) when Spikes picked off Sanchez at the goal line? We then go down and score, and the game is over at that point. Huge turning point, of which he was the pivotal player.

 

[RDS11]

You're ignoring other factors when making your analysis, and Rob's trying not to do that. As for the defense playing its 3 best games without Spikes, that's simply a matter of opinion. I would argue that the 45-3 Jets game, with Spikes, was the defense's best game, because that game turned into quicksand for the Jets and it became party time for New England on both sides of the ball.

True, the Jets game could be included into their best game, but then it would be held against Rob's argument that Spikes was special against the run, as the Pats allowed 150 yards to the Jets, more than against the Packers.

The only fact Rob has been mentioning is the Packers game. Well, Spikes was starting against the Browns, and the Pats were run over for 230 yards. Colt McCoy was a rookie QB, with as much starting experience as Flynn. And they lost, big time. At least they won against the Packers, maybe without their starting QB, but that team had to be much better than the Browns were.

Again, that's why I was trying to avoid bringing an analysis based on 1 game. If you want to use only the Packers game as a statement on the utility of Spikes, than you have to look at the Browns game too.

As a statitician, I don't value intangibles, I value production. And Spikes wasn't a factor in 2010 : 1 INT, no forced fumble, no fumble recovery and no sack. That's too much zeros for a guy who was starting. I'm not a Guyton fan, but the 'other' guy at ILB had 3 sacks, 2 INTs, 1 forced fumble, 1 fumble recovery and 2 TDs.
Maybe it explain a bit why the defense had better statistics when Spikes was out.

 

[Deus Irae]

True, the Jets game could be included into their best game, but then it would be held against Rob's argument that Spikes was special against the run, as the Pats allowed 150 yards to the Jets, more than against the Packers.

The only fact Rob has been mentioning is the Packers game. Well, Spikes was starting against the Browns, and the Pats were run over for 230 yards. Colt McCoy was a rookie QB, with as much starting experience as Flynn. And they lost, big time. At least they won against the Packers, maybe without their starting QB, but that team had to be much better than the Browns were.

Again, that's why I was trying to avoid bringing an analysis based on 1 game. If you want to use only the Packers game as a statement on the utility of Spikes, than you have to look at the Browns game too.

As a statitician, I don't value intangibles, I value production. And Spikes wasn't a factor in 2010 : 1 INT, no forced fumble, no fumble recovery and no sack. That's too much zeros for a guy who was starting. I'm not a Guyton fan, but the 'other' guy at ILB had 3 sacks, 2 INTs, 1 forced fumble, 1 fumble recovery and 2 TDs.
Maybe it explain a bit why the defense had better statistics when Spikes was out.

Again, you keep ignoring obvious factors. As a statistician, you should know that doing so kills the accuracy/applicability of your data.

 

[RDS11]

Not only did Spikes play in that game, but he had arguably the most important play on defense. I might be foggy on the details, but weren't the Jets driving for a tying TD (or at least a TD to bring them back within a score) when Spikes picked off Sanchez at the goal line? We then go down and score, and the game is over at that point. Huge turning point, of which he was the pivotal player.

He was also the guy in coverage in the playoff game allowing the long completition (30 yards to Keller) that lead to a 75 yards drive for a TD in the 3rd quarter, in 5 plays, killing most hope of a comeback.

 

[RDS11]

Again, you keep ignoring obvious factors. As a statistician, you should know that doing so kills the accuracy/applicability of your data.

Tell me what I'm ignoring, because I've provided enough data to prove that Spikes wasn't a factor in 2010, in more ways than one.

Fact :the defense was more efficient during Spikes absence.
Fact : Spikes backup (not the right term, as Guyton would play on specific downs) had much better production.

I agree that we do not have a 100 games sample to make judgment. True. But we can only look at the data we have, and that's 16 games. And based on that data, not on opinion or what if, the loss of Spikes was mostly irrelevant.

 

[Deus Irae]

Tell me what I'm ignoring, because I've provided enough data to prove that Spikes wasn't a factor in 2010, in more ways than one.

Fact :the defense was more efficient during Spikes absence.
Fact : Spikes backup (not the right term, as Guyton would play on specific downs) had much better production.

Both of your "facts" are non-facts being dressed up, or slanted takes being called facts. By ignoring external factors and realities, you destroy your argument. Just looking at the Spikes/Guyton comparison without noting the huge difference in snap totals (628 for Guyton, 349 for Spikes) makes your "fact" anything but a real truth.

I agree that we do not have a 100 games sample to make judgment. True. But we can only look at the data we have, and that's 16 games. And based on that data, not on opinion or what if, the loss of Spikes was mostly irrelevant.

You're attempting static analysis on a system where static analysis is pretty much useless.

 

[RDS11]

Both of your "facts" are non-facts being dressed up. By ignoring external factors and realities, you destroy your argument.

That's a very good non-reponse. You probably should have been a politician.

Please define ''External factors and realities''. Yes, there are such things in statistics when we value external factors, it's called multi-variate statistics, but here I'm guessing you are only toying with the what-ifs.
What if Spikes had missed the 1st 4 games instead of the last four.
What if the Pats would have let Spikes played on 3rd and long.

All good questions, for which there's no answer, just opinions.

Again, we cannot create data. We only have what is there. And nothing is showing that the loss of Spikes was a blow to that defense.
If so, than just prove me wrong with numbers to back it up.

 

[BradyFTW!]

That's a very good non-reponse. You probably should have been a politician.

Please define ''External factors and realities''. Yes, there are such things in statistics when we value external factors, it's called multi-variate statistics, but here I'm guessing you are only toying with the what-ifs.
What if Spikes had missed the 1st 4 games instead of the last four.
What if the Pats would have let Spikes played on 3rd and long.

All good questions, for which there's no answer, just opinions.

Again, we cannot create data. We only have what is there. And nothing is showing that the loss of Spikes was a blow to that defense.
If so, than just prove me wrong with numbers to back it up.

If you expect anyone to take you seriously when your "analysis" ignores the overall progress of the defense last year and the weakness of opponents over the 4 weeks that Spikes missed, then I dunno what to tell you. Analysis of that caliber is exactly why we had some idiots on this board lobbying to trade Brady a few years back. Just because you've used some numbers doesn't mean that you've made a valid argument.

Also, there are plenty of us here who have a more-than-passing familiarity with multivariate statistics. I'd be happy to take a look at whatever regression analysis you've come up with that's led to the conclusion that Spikes' presence is not beneficial to the defense. Of course, I'm fairly certain that you don't have any, and you're just throwing around what you hope are sufficiently big words to intimidate people from disagreeing with you.

 

[Rob0729]

I'm not saying losing Spikes was a positive, I'm saying it has almost no impact statistically.
Obviously, if Belichick put him on the field, it was because he thought Spikes was the best option. And I'm pretty sure Belichick knows more about football than I do, so losing Spikes had to be a negative.

But that being said, I'm not looking at things in a vacuum. There's only 16 games in a football season, so 4 games is a good sample. In fact, you are the one who doesn't try to put things in perspective as you are looking only at the Packers game as proof that the suspension to Spikes was a setback to the defense. The fact is that the defense played their 3 best games WITHOUT Spikes. Yes, other factors need to be taken into consideration, but this is a fact, no way around it. Could Spikes have contributed positively in these 3 games ? Maybe...but based on the data from last season, he was not an impact player, so his loss had a minimal effect on the defense.

First, I would dispute that the Pats' defense played their three best games without Spikes. Their best defensive game was against the Jets in December when Spikes played. I thought their second best defensive game was vs. the Steelers although the Steelers got a lot of points and yards in garbage time to make the defense look like they had a far worse performance (they went into the fourth quarter 24-3 and the Steelers and Pats traded garbage time TDs back and forth) and Spikes played that game. The only game that I would say was in the top three performances that the Pats' defense played that was after Spikes was suspended was the Chicago game and they had Jay Cutler and the weather to help them (otherwise, I would have ranked them #2).

Second, comparing stats from 12 games vs. 4 games is not a statistically valid argument. The fact that the Pats were a young defense and even you admitted that it took time for the defense to gel as a unit, the games with Spikes in it are skewed because there are a series of games that the defense struggled through growing pains. By the end of the season, McCourty had emerged as a dominant CB and the rest of the defense improved.

Third, you are not using any analysis other than stats. You are looking at the box score to draw conclusions. That is looking at things in a vaccuum. The box score doesn't tell you that the Pats defense dominanted the Steelers' offense until the game was pretty much out of reach. The box score doesn't tell you that the same thing about the Bengals game. Stats in football are very misleading and should not be used exclusively for analysis. This isn't baseball.

Fourth, I never said that the Packers' game was proof that Spikes was missed. It is supporting evidence to my overall opinion, but it isn't the smoking gun to the argument.

 

[TheComeback]

The only analysis that matters is watching a guy play. And anyone who watches Spikes play knows he has talent, and knows helped this defense last year. That would be why he started so many games over Guyton.

You can analyze all the statistics you want, but they won't tell you the story of what's actually happening on the field. I believe a certain someone said that statistics are for losers, and that person is correct.

I suggest RDS stop digging through ESPN player profiles and start watching Spikes play. I mean, really watching him play, without making any presumptions.

 

[Rob0729]

True, the Jets game could be included into their best game, but then it would be held against Rob's argument that Spikes was special against the run, as the Pats allowed 150 yards to the Jets, more than against the Packers.

The only fact Rob has been mentioning is the Packers game. Well, Spikes was starting against the Browns, and the Pats were run over for 230 yards. Colt McCoy was a rookie QB, with as much starting experience as Flynn. And they lost, big time. At least they won against the Packers, maybe without their starting QB, but that team had to be much better than the Browns were.

Again, that's why I was trying to avoid bringing an analysis based on 1 game. If you want to use only the Packers game as a statement on the utility of Spikes, than you have to look at the Browns game too.

As a statitician, I don't value intangibles, I value production. And Spikes wasn't a factor in 2010 : 1 INT, no forced fumble, no fumble recovery and no sack. That's too much zeros for a guy who was starting. I'm not a Guyton fan, but the 'other' guy at ILB had 3 sacks, 2 INTs, 1 forced fumble, 1 fumble recovery and 2 TDs.
Maybe it explain a bit why the defense had better statistics when Spikes was out.

First, did you even watch the Jets game. The Pats were allowing the Jets to run the ball (they ran the ball 31 times) to run down the clock daring them to throw. Besides, Spikes had a key INT that stopped the only drive that the Jets really had a chance to score on other than the one they scored a field goal.

Second, you cannot analyze football unless you value the intangibles. Football isn't a stats driven sport. Bruschi never had incredible stats, but he was the heart and soul of the defense and invaluable. Wilfork is invaluable today, but he will never get close to a Haloti Ngata type of stats production. Especially in a two gap 3-4 defense like the Pats where they are taught to do their role and not focus on stats, it is tough to measure a player on stats.

Third, comparing Colt McCoy to Matt Flynn is laughable. Yes, the Pats defense played like crap in that game, but it is only one game in 12 while the Packers' game was one in four. Since you are a statician, you know that the Pats got eaten alive by a rookie QB only 8% of the time with Spikes and 25% of the time without him. Clearly statistical evidence that shows Spikes is far more valuable than you give him credit for.

Fourth, I never used the Packer game in a vacuum. I mentioned all four games why they should be discounted. I expanded my reasoning on one.

 

[Deus Irae]

That's a very good non-reponse. You probably should have been a politician.

Please define ''External factors and realities''. Yes, there are such things in statistics when we value external factors, it's called multi-variate statistics, but here I'm guessing you are only toying with the what-ifs.
What if Spikes had missed the 1st 4 games instead of the last four.
What if the Pats would have let Spikes played on 3rd and long.

All good questions, for which there's no answer, just opinions.

Again, we cannot create data. We only have what is there. And nothing is showing that the loss of Spikes was a blow to that defense.
If so, than just prove me wrong with numbers to back it up.

You ignored the vast disparity in snaps when you cited to 'production', for example. You ignored weather conditions, strength of opponents, single-game circumstances and other important factors.

You seem to be either incapable or, or unwilling to, understand the differences in situation, so there's really not much point in discussing this further, beyond saying that your analysis is essentially useless as any sort of attempt at 'objectivity' because of its shortcomings.

 

[RDS11]

If you expect anyone to take you seriously when your "analysis" ignores the overall progress of the defense last year and the weakness of opponents over the 4 weeks that Spikes missed, then I dunno what to tell you. Analysis of that caliber is exactly why we had some idiots on this board lobbying to trade Brady a few years back. Just because you've used some numbers doesn't mean that you've made a valid argument.

Also, there are plenty of us here who have a more-than-passing familiarity with multivariate statistics. I'd be happy to take a look at whatever regression analysis you've come up with that's led to the conclusion that Spikes' presence is not beneficial to the defense. Of course, I'm fairly certain that you don't have any, and you're just throwing around what you hope are sufficiently big words to intimidate people from disagreeing with you.

No I don't. But that's exactly what I said (when mentioning external factors), but then again these kind a statistics are not readily available on NFL.com and I don't have the time like folks at Coldhardfootballfacts to look at the correlation between Spikes play and the defense success.

I will repeat for the X-time, I wasn't saying Spikes doesn't have talent or anything else, the only thing I said is that his absence didn't have much impact on the Patriots D.

I just don't believe in the ''I saw him play, he's good'' when statistics don't back the statement. These eye tests can be tricky, especially coming from fans who are already bias for the player. And for now, the simple fact is that Spikes hasn't shown much in the production department. Will he become a Pro-Bowler ? Time will tell, most players have a breaktrought in their 2nd year.

Now if you want to talk about stats, take it any way you want, under any kind of scenarios or ''realities'', but Guyton is currently a more productive player than Spikes. And nobody needs multivariate stats to see this one, as both played the same position on the same team.

Yes, the last 4 games were not against high powered offense, but we shouldn't count them ? Spikes played against these teams too the first time around, and played also against the Browns, the Lions, the Vikings...not great offenses either.

While I agree that other factors should be accounted for when evaluating the defense, again, the loss of Spikes was statistically irrelevant. That's all I said.

The comparison to the trade Brady group is just ridiculous. First, we're talking about a LB who has played 1 year versus a 3 times SB winning quarterback. Putting words in mouth of others just doesn't make a valid point

 

[RDS11]

You ignored the vast disparity in snaps when you cited to 'production', for example. You ignored weather conditions, strength of opponents, single-game circumstances and other important factors.

You seem to be either incapable or, or unwilling to, understand the differences in situation, so there's really not much point in discussing this further, beyond saying that your analysis is essentially useless as any sort of attempt at 'objectivity' because of its shortcomings.

With this theory, no stat in NFL could be used for comparison.
For example, a sack on 3rd and long wouldn't mean the same thing as a sack on 1st and 10 ? A 18 yards draw on 3rd and 20 should be modified so that a 2 yards run on 3rd and 1 would be worth more ?


All true, and when enough games are played, with enough plays, one should assume that all thing will become equal. Football is not the best sports for this, as the number of plays over an entire season is not that much, but given the data that is at our disposal, we can assume that Spikes and Guyton data are comparable, at the very least.

 

[Rob0729]

I just don't believe in the ''I saw him play, he's good'' when statistics don't back the statement. These eye tests can be tricky, especially coming from fans who are already bias for the player. And for now, the simple fact is that Spikes hasn't shown much in the production department. Will he become a Pro-Bowler ? Time will tell, most players have a breaktrought in their 2nd year.

Based on this, you think Browns' NT Ahtyba Rubin is a better NT than Vince Wilfork. Both players had 2 sacks, but Rubin had 25 more tackles than Wilfork. In fact, Rubin had as many solo tackles than Wilfork had total tackles. My eyes tell me that Wilfork is a far superior player, but that stats says otherwise. Clearly I am just biased and Rubin is the superior player.

 

[Deus Irae]

With this theory, no stat in NFL could be used for comparison.
For example, a sack on 3rd and long wouldn't mean the same thing as a sack on 1st and 10 ? A 18 yards draw on 3rd and 20 should be modified so that a 2 yards run on 3rd and 1 would be worth more ?


All true, and when enough games are played, with enough plays, one should assume that all thing will become equal. Football is not the best sports for this, as the number of plays over an entire season is not that much, but given the data that is at our disposal, we can assume that Spikes and Guyton data are comparable, at the very least.

Stats can be used for comparison, even in football, as long as you acknowledge that they aren't usually going to be the end all. That's not the issue here. The issues here, for me, are your claiming "fact" about things that weren't, your refusal to understand and/or acknowledge the non-stat things which have an obvious impact on the stats themselves, and your making an asinine argument about production without taking snap totals into account.

 

[RDS11]

Stats can be used for comparison, even in football, as long as you acknowledge that they aren't usually going to be the end all.

I agree, but some stats have an higher corrolation to victory than others. Running the football, for example, or stopping the run, do not have as hign an impact on victory as most would think.

The issues here, for me, are your claiming "fact" about things that weren't, your refusal to understand and/or acknowledge the non-stat things which have an obvious impact on the stats themselves, and your making an asinine argument about production without taking snap totals into account.

Explain to me how I used fact that weren't.

I acknowledged that the impact of the defense getting better cannot be attributed on Spikes being suspended. I agree to that, it's obvious. But how can you argue about Spikes lack of production ? No forced fumble, no fumble recovery, no sack. Guyton, playing on the same team on the same position posted better stats. I'm not trying to compare another DT to Wilfork like same other poster tried a few message prior. I'm looking at 2 guys, from the same team playing the same position.

Also, the number of snap totals should represent the importance of said player to that team. If the guy doesn't play, obviously he won't put great stats. But it also shows that the coaching staff believe that other players are better than said player. Spikes played 349 snaps according to Reiss. Guyton played 628, but in 4 more games.
Butler played 381 snaps, and he was benched for a long period of time following the 1st Jets game...to me, it's another indication that Spikes wasn't such a factor this year.

Snap totals and production are closely related. If you play, you are expected to produce. If you produce, you play. If you don't produce, you don't play.

 

[Deus Irae]

I agree, but some stats have an higher corrolation to victory than others. Running the football, for example, or stopping the run, do not have as hign an impact on victory as most would think.

Yeah, see, this is just not true. Well, more accurately, you're making assumptions "most would think", that are not really applicable to the discussion.

Explain to me how I used fact that weren't.

I acknowledged that the impact of the defense getting better cannot be attributed on Spikes being suspended. I agree to that, it's obvious. But how can you argue about Spikes lack of production ? No forced fumble, no fumble recovery, no sack. Guyton, playing on the same team on the same position posted better stats. I'm not trying to compare another DT to Wilfork like same other poster tried a few message prior. I'm looking at 2 guys, from the same team playing the same position.

Also, the number of snap totals should represent the importance of said player to that team. If the guy doesn't play, obviously he won't put great stats. But it also shows that the coaching staff believe that other players are better than said player. Spikes played 349 snaps according to Reiss. Guyton played 628, but in 4 more games.
Butler played 381 snaps, and he was benched for a long period of time following the 1st Jets game...to me, it's another indication that Spikes wasn't such a factor this year.

Snap totals and production are closely related. If you play, you are expected to produce. If you produce, you play. If you don't produce, you don't play.

I already explained it.

And your comment about snap totals and production just doesn't hold up when applied to a lot of instances, because it completely ignores the Patriots penchant for situational football.

SOSH has a thread breaking down the snaps on a week-by-week basis, based upon Reiss' numbers:

http://sonsofsamhorn.net/topic/60114-2010-patriots-defensive-snaps/page__st__100

You can see the numbers fluctuating wildly for Spikes and Guyton.

 

[RDS11]

I already explained it, and you went the smartass route with it.

And your comment about snap totals and production is just nonsense when applied to a lot of situations.

Obviously...NFL coaches are just dumbasses that don't put their best players on the field.

Belichick seems to think that playing Guyton more than Spikes gives him the best chance to succeed, but you disagree ?