wonderl33t
Club Member
Sacks Have Little to Nothing to Do with NFL Quarterback Play and Winning
- Thread starter Shouright
- Start date
Well the increase in number of pirates cannot be due to the increase in flesh eating zombies so it must be due to the increase in pirate surgeons performing breast enhancement surguries. Conversely, the increase in global warming can only be attributed to the amount of flying saucers flying into our atmosphere.
dlockz
Hall Of Famer
Getting your qb sacked a lot has nothing to do with winning and losing, what an amazing theory that reeks of not playing football lol.
I guess people with this theory didn't watch the Buffalo game.
I guess people with this theory didn't watch the Buffalo game.
J Tes
next generation
- Joined
- Aug 9, 2003
- Messages
- 2,362
- Reaction score
- 0
Post of the ****ing year. :lol2:
Nope. That goes to the Jeff Ireland like a boss thread
Shou, these are some interesting stats, and you've presented them in a unique way. Here are my observations:
1. I do not think the conclusion you've drawn ("sacks have no correlation to QB play and winning or losing") is worded as accurately as it should be. My reading of your analysis draws me to a differently worded conclusion altogether - something like "QB's with winning seasons vs. losing seasons seem to be sacked about the same percentage of the time - in fact QB's with winning seasons seem to get sacked a little more frequently". And "QB's with higher ratings vs. lower ratings seem to be sacked about the same percentage of the time - in fact, QB's with higher ratings seem to get sacked a little more frequently".
2. If you wanted to get a better picture of how sack % affects a QB's performance, you would look at one QB's games, and correlate his sack % in a game to his W/L outcome and his QB rating. My guess (although I'm not sure) is the number would be positive in both cases.
3. What would be very interesting to see would be a comparison of the various QB's on the above correlation that would show how strongly sack % is tied to QB performance for each QB. We could see once and for all how much RT is affected by sacks and how well he compares to his peers in that regard.
4. However, both 2. and 3. would need sufficient sample size (i.e. enough games played by an individual QB) to draw reasonable conclusions. I'm sure you're aware of the the statistical tests that can be run on the samples to determine their validity relative to the correlative tests performed on them.
That's my 2c - maybe I've completely misread your analysis, and if so, I apologize.
1. I do not think the conclusion you've drawn ("sacks have no correlation to QB play and winning or losing") is worded as accurately as it should be. My reading of your analysis draws me to a differently worded conclusion altogether - something like "QB's with winning seasons vs. losing seasons seem to be sacked about the same percentage of the time - in fact QB's with winning seasons seem to get sacked a little more frequently". And "QB's with higher ratings vs. lower ratings seem to be sacked about the same percentage of the time - in fact, QB's with higher ratings seem to get sacked a little more frequently".
2. If you wanted to get a better picture of how sack % affects a QB's performance, you would look at one QB's games, and correlate his sack % in a game to his W/L outcome and his QB rating. My guess (although I'm not sure) is the number would be positive in both cases.
3. What would be very interesting to see would be a comparison of the various QB's on the above correlation that would show how strongly sack % is tied to QB performance for each QB. We could see once and for all how much RT is affected by sacks and how well he compares to his peers in that regard.
4. However, both 2. and 3. would need sufficient sample size (i.e. enough games played by an individual QB) to draw reasonable conclusions. I'm sure you're aware of the the statistical tests that can be run on the samples to determine their validity relative to the correlative tests performed on them.
That's my 2c - maybe I've completely misread your analysis, and if so, I apologize.
Thanks for your thoughtful reply. There was little to no correlation game-to-game in 2013 between sacks and the stats of Ryan Tannehill's that were alluded to in the original post. His play as measured by those stats didn't vary as a function of sacks. Given the season data since 1994, I'd be surprised to find something different from that for other QBs.Shou, these are some interesting stats, and you've presented them in a unique way. Here are my observations:
1. I do not think the conclusion you've drawn ("sacks have no correlation to QB play and winning or losing") is worded as accurately as it should be. My reading of your analysis draws me to a differently worded conclusion altogether - something like "QB's with winning seasons vs. losing seasons seem to be sacked about the same percentage of the time - in fact QB's with winning seasons seem to get sacked a little more frequently". And "QB's with higher ratings vs. lower ratings seem to be sacked about the same percentage of the time - in fact, QB's with higher ratings seem to get sacked a little more frequently".
2. If you wanted to get a better picture of how sack % affects a QB's performance, you would look at one QB's games, and correlate his sack % in a game to his W/L outcome and his QB rating. My guess (although I'm not sure) is the number would be positive in both cases.
3. What would be very interesting to see would be a comparison of the various QB's on the above correlation that would show how strongly sack % is tied to QB performance for each QB. We could see once and for all how much RT is affected by sacks and how well he compares to his peers in that regard.
4. However, both 2. and 3. would need sufficient sample size (i.e. enough games played by an individual QB) to draw reasonable conclusions. I'm sure you're aware of the the statistical tests that can be run on the samples to determine their validity relative to the correlative tests performed on them.
That's my 2c - maybe I've completely misread your analysis, and if so, I apologize.
Oh, that's awkward...
OnDaChainWax
☠️ Banned ☠️
Nothing. Nothing at all. Not physically... Not mentally...
I wont even read this joke of a thread. Someone please mail me if i missed something beautifully sarcastic.
I wont even read this joke of a thread. Someone please mail me if i missed something beautifully sarcastic.
The team that wins the sack battle wins 70% of their games. When you are leading the league is sacks allowed, it is difficult to win the sack battle.
Sacks swing the balance of games by an average of 2 points. Point differential is the most important stat.
ANY/A differential is more highly correlated than YPA. Sacks lower a teams ANY/A differential.
Are you saying that the Dolphins somehow buck these league wide trends? If so, do you have proof?
You didnt use season data. You only used QBs who had at least 500 attempts in a season, which is a lot of attempts for any QB especially before all of the new rules were made protecting receivers and QBs. Most QBs who exhibit bad play would not have 500 attempts in a season. You selected a sample that gives you the outcome you wanted, which makes your analysis bogus.
He also arbitrarily (what does the salary cap have to do with this analysis?) used data from 1994 and forward. That also skews the data because it includes some QBs with only their years as veteran QBs.
It makes no sense why the salary cap year is the starting point for sampling.....selective sampling at that. Why not start with the rule changes that allowed for some of these pass happy offenses take form.
I'm convinced that he spends hours trying different scenarios until he finds the right combination to support his views. And we're suffering from confirmation bias.....