Checking the Numbers: Bills-Patriots Game

by WolfpackSteelersFan on Sep 16, 2009 12:06 PM EDT Comment 9 comments

Hello, Patriots fans! I have just completed a new post analyzing the box score stats from the game monday night. This post gives an idea of how the defenses and offenses are performing compared to the 2008 season averages. Hope you enjoy!

On Monday night, the New England Patriots and Buffalo Bills matched up for a game expected to be an easy win for the Pats. Surprisingly, the Bills held an 11 point lead late in the game, but some Tom Brady heroics combined with an untimely fumble on a kick return did in the Bills for a 12th straight time against New England. There are plenty of recaps out there reviewing the game, but here I wanted to look at this game from my own analytical perspective. More...

In this post, I took the box score stats from the game and compared how each team did against their season averages from the 2008 season. If it were later in the year, I would compare them against each team’s average so far in the 2009 season. But, since the previous season usually serves as the baseline for predictions going into a new year, I think that analyzing against the previous season’s numbers is perfectly acceptable. Perhaps after week 6 or 7 (maybe 8), I will begin using current season averages as the basis for comparison.

So, anyway, on to the numbers! Below, I have four tables.

To read the rest, please click here.

The views expressed in these FanPosts are not necessarily those of the writers or SBNation.

0 recs | Comment 9 comments | Share

Story-email Email Printer Print

Comments

A lot to digest here

Terrific work, thanks for putting this together

Keep the faith!

by Marima on Sep 16, 2009 11:19 AM EDT reply actions 0 recs

Glad you like it!

If it is popular, I’ll try to do this every week.

For ideas on statistical analyses, email me at wolfpacksteelersfan@gmail.com.

by WolfpackSteelersFan on Sep 16, 2009 12:51 PM EDT up reply actions 0 recs

A fumble is because of the player

a strip is because of the defense

That play was a strip.

The more you know, the more you know that you don't know.

by SlotMachinePlayer on Sep 16, 2009 11:29 AM EDT reply actions 1 recs

True

But if the player has both arms wrapped tightly around the ball, it’s very difficult to strip. He should have just taken a knee in the endzone.

Still, good play on the defensive side to get the strip.

For ideas on statistical analyses, email me at wolfpacksteelersfan@gmail.com.

by WolfpackSteelersFan on Sep 16, 2009 12:50 PM EDT up reply actions 0 recs

Our O line needs

to do a better job creating running lanes. Maroney did a better job running in a forward direction, but had nowhere to run. Same for Taylor. Not sure what A Peterson does in Viking territory, run around the line when it clogs up? Maybe they could teach that to our guys. Don’t know. Maybe they create better running lanes with their line? Could use a few more screen passes of our own as we did in preseason to pretty good success.
That was the Bills after all, not the Steelers or Colts or one of the better teams. We can’t play that way against the Jets this Sunday, or OUCH!

'A warrior may change his metal, but not his
heart,' as the saying is upon Barsoom." Dejah Thoris

by ba-shiznit on Sep 16, 2009 3:16 PM EDT reply actions 0 recs

Nice analysis

You’ve put together a nice quantitative analysis to complement the abundance of qualitative stuff out there. At what point are you planning to incorporate 2009 numbers as the constant variable?

I like it. I hope to see this some more!

by Nikolai19 on Sep 16, 2009 6:32 PM EDT reply actions 0 recs

2009 Numbers

Thanks, appreciate it. I was thinking about 6 or 7 weeks to get a good number of games averaged together before using 2009 numbers. Four or fewer games seems too small of a sample size, and I’m not really sure that 5 or 6 is enough. It will be better to use same year numbers, though, since players change teams, retire, etc.

For ideas on statistical analyses, email me at wolfpacksteelersfan@gmail.com.

by WolfpackSteelersFan on Sep 16, 2009 7:06 PM EDT up reply actions 0 recs

A thought

I think you’re probably right. Something in the range of 6-7 weeks might be appropriate. While I was typing this, I had a quick thought regarding a way to incorporate the 2008 and 2009 numbers. Use the 2008 numbers like you’re doing now, and then use whatever meager 2009 sample you have, but then weight the analysis strongly in favor of the 2008 numbers when coming up with your percentages. Gradually decrease the weight until the 6th or 7th week and then use 2009 numbers exclusively. It’s probably a LOT of extra work, but I just had that quick thought and figured I’d mention it.

by Nikolai19 on Sep 16, 2009 10:28 PM EDT up reply actions 0 recs

Thanks!

That sounds like a good a idea. It may not be too much work, I’ll have to look into it. Depending on how much time I have, I may get it incorporated this coming week, but if not, it’s definitely something to work on down the road.

For ideas on statistical analyses, email me at wolfpacksteelersfan@gmail.com.

by WolfpackSteelersFan on Sep 17, 2009 10:25 AM EDT up reply actions 0 recs