NUMBERS - Lets make predictions on questionable stats
In my last post, we did a little maths to sort out good teams from bad teams in week 2. Since we only had two data points against mostly questionable competition, any sort of analysis to this point is purely for entertainment. Week 3 hasn't changed much, so... let's get to entertainin'!
Here's week 3's Points Per Play numbers. In case you forgot, here's how I get them:
Points Per Play Ratio = Offensive Points Per Play / Defensive Points Per Play
where:
Offensive Points Per Play = Total Points Scored / Total Plays
and
Defensive Points per Play Allowed = Total Points Allowed / Total Defensive Plays
Oregon played Portland State, scores a metric fuck ton of points and is thus ranked number 1. Same for Alabama. How 'bout them Stanford Cardinal? Is that Toby Gerhart guy a distant memory now or what?
In the comments from the last post, Year2 mentioned using Yards Per Point as another way to measure team efficiency. Here's that calculation:
Offensive Yards Per Point = Total Offensive Yards / Total Points Scored (smaller number is better)
Defensive Yards Per Point = Total Defensive Yards Allowed / Total Points Allowed (bigger number is better)
Teams are ranked by their "Gap"
Gap = Offensive Yards Per Point - Defensive Yards Per Point
This assumes you have an efficient offense if you travel less yards to score points. If you put up a bunch of yards and score a bunch of points, you'll have lower Yards Per Point. Your team is beating defenses, getting big plays, scoring in the red zone, etc. If your defense holds offenses to less yards and less points, they're good at getting off of the field, getting turnovers, etc. Theoretically, a "bend but don't break" defense gets heavily rewarded as they'd allow a lot of yards, but no points. Subtract Off. Yards Per Point by Def. Yards Per Point and you get the "Gap" between your offense and defense (or your offense and oppositions offense if you want to think about it that way). Here's what the Top 25 looks like for week 3:
Hey there Florida, I see you're in the Top 10. Your Offensive Yards Per Point don't look that bad, so maybe things might turn out OK. If that doesn't work, we can go get ice cream sundaes after the game.
Now, lets make some stupid predictions!
Tonight! Miami (-3.5) vs. Pittsburgh
Miami: 1.887 PPP, 3.6 Gap
Pittsburgh: 1.509 PPP, 4.9 Gap
The numbers are split, so I'll go with Miami as they dominated their weaker competition and played respectably in a loss to tougher competition.
Oregon State vs. Boise State (-18.5)
Oregon State: 1.362 PPP, 5.4 Gap
Boise State: 1.813, PPP, 0.2 Gap
The Gap number suggests the Beavers should control the clock and keep Boise State's offense off the field, but the PPP numbers suggest Boise will score with whatever time they have. I'll go with Boise State seeing as they can score at will against Pac 10 teams.
LSU (-10) vs. WVU
LSU: 2.612 PPP, 13.4 Gap
WVU: 1.555 PPP, 4.6 Gap
Domination for the Tigers. Ellis Hugh wins.
Alabama (-7) vs. Arkansas
Alabama: 6.978 PPP, 27.9 Gap
Arkansas: 3.002 PPP, 8.9 Gap
Yeah... uh, this is when I stop believing the validity of these numbers based on the few games both teams have played. The Gap delta is ridiculous and would predict an Oregon-New Mexico blowout, but if I have to stick with them, I'll say the Tide will win big.
If any of these predictions pan out, I hope to become internet famous. Maybe I can get a TV deal out of it.
Now, one last bit. In the comments from the last post, nuftw requested a calculation that could measure a teams "fail" potential.
"Total Fail" = ( Total Offensive Yards + Punt Yards ) / ( Punts + Turnovers + Turnovers on Downs )
"Offensive Fail" = Total Offensive Yards / ( Punts + Turnovers + Turnovers on Downs )
The idea is "yards per fail." How far can your team go before it fucks up. Looking at the good teams is kinda boring, so with that in mind, I'm pleased to announce the Fail 25 Rankings for On-Field Ineptitude: Week 3.
I feel like I should give something out to New Mexico, but can't think of anything that degrades, yet inspires (as any fail award should). Here's the best I could do:
Lobos, may Leonardo Dicaprio's confident stride and uplifting smile inspire you to achieve great fail this season. (But I really hope to make this a weekly thing, so if you could fail less than other teams every now and then to spice things up, that would be great.)
Also, HAHAHAHAHA VANDY BEAT OLE MISS!
FanPosts are user-submitted, and thus NOT representative of EDSBS editorial or any of our opinions unless posted by us ourselves. Please refrain from posting blatant spam or self-promotion, because this makes us hate you. Thanks!
5 comments
|
0 recs |
Do you like this story?
Comments
I was told there would be no math
That said, Bill Snyder is impressed with your fail potential ranking system, and would like to offer you the position of Schedule Consultant (before a SEC school snags you, of course.)
Ellis Hugh, Elle Sue, Dave, whatever…
by vincentnakamura on Sep 24, 2010 12:42 PM EDT up reply actions
SO MUCH DATAZ
Basically, this means that for New Mexico to get a td, they need to get the ball at the opposing 20 yd line. That sounds about right.
I’m also even more worried about the CMU game this week. Playing the #5 team in the country is never easy, even at home.
Reality has a little-known Northwestern bias
NU prez knows how to get PUMPED
I only have a sample size of one, but I’m pretty sure once your fail potential drops below 22, your head coach starts physically abusing his assistants. Lets keep a close eye on this one…
by vincentnakamura on Sep 24, 2010 12:47 PM EDT up reply actions

by 



















