clock menu more-arrow no yes

Filed under:

Processing the Numbers, Football Edition |
The Big Peach Bowl Preview

New, 24 comments

Previewing a slugfest in Atlanta that should be one of the better games this postseason

Spruce Derden-USA TODAY Sports

All statistics are courtesy of Football Outsiders, home of the F/+ Combined Ratings for college football.
The Fremeau Efficiency Index (FEI) was created by Brian Fremeau; check out his website BCFToys for other goodies.
The S&P+ rating was created by Bill Connelly; check out his college football analytics blog, Football Study Hall.
Hat tips to Addicted to Quack's kalon and FO's 7th Day Adventure column for the inspiration.

But first, let’s talk about a couple of things.

A Few New Stats

Well, not really new, but perhaps new to you! These have been around for most of the season, I just haven’t been using them in the previews. There are two stats pertaining to line play1, and five metrics relating to special teams play that have been added to the Glossary. I did a little writeup on them as well, which can be found behind the button.

1 | Which I would have been talking about if I’d realized they were available, which I didn’t until the Missouri game. Whoops.

First up are the Fremeau Special Teams Efficiency Components, which I’ve referenced in places over the season but never really introduced or defined. The following is taken from the Glossary:

Fremeau Special Teams Efficiency Components - The special teams component of F/+ is based on Brian Fremeau’s Special Teams Efficiency, which is made up of the following five components of special teams play (per FootballOutsiders):

  • FGEField Goal Efficiency, the scoring value per field goal attempt earned by the field goal unit as measured against national success rates.
  • PREPunt Return Efficiency, the scoring value per opponent punt earned by the receiving team as measured against national return rates.
  • KREKickoff Return Efficiency, the scoring value per opponent kickoff earned by the receiving team as measured against national return rates.
  • PEPunt Efficiency, the scoring value per punt earned by the opponent's receiving team as measured against national return rates.
  • KEKickoff Efficiency, the scoring value per kickoff earned by the opponent's receiving team as measured against national return rates.

Fairly straightforward — each punt, kickoff, and field goal generates some amount of scoring value, although I’ll admit I’m not 100% certain how Brian calculates that for punts and kickoffs at the moment. Once that’s taken per play and opponent-adjusted, you have these handy little efficiencies.

Next up is ASR or Adjusted Sack Rate, which is a measure of success in pass protection (when calculated for the offense) and a measure of success rushing the passer (when calculated for the defense). It’s simply the sack rateA adjusted for opponent strength and scaled to 100; the higher the rating the better. As noted, this is calculated both for the offense (OASR) and defense (DASR).

A | ( sacks / [sacks + passing attempts] )

Finally, we have ALY or Adjusted Line Yards, which attempts to separate out the portion of rushing success attributable to the line. The rushing yardage for every rushing attempt by a running backB is taken and weighted as follows:

B | WR runs are considered misdirection plays, and QB runs are typically scrambles.

  • Runs for a loss are weighted 120%.
  • Runs for 0-4 yards are unweighted.
  • Runs for 5-10 yards are weighted 50%.
  • Runs for 11 or more yards are not included.

This all makes sense, right? If a run’s stuffed behind the line, that usually means the line failed somewhere and allowed the defense through to make a play. Short runs are almost entirely due to the push of the line, so there’s no reason to remove credit for those yards. Once a running back gets into the second level, the line’s done its job and the rest is up to the back, so splitting credit seems appropriate. Runs in excess of 10 yards beyond the line of scrimmage are almost entirely produced by the running back, so it doesn’t make sense to credit the line at all. The best part about these weights is they were determined from analysis of play-by-play data, and you can read more about that here.

After the weighting process is complete, the runs are further adjusted for the strength of the opponent and for the game situation (namely, down and distance). This concept was originally generated for the NFL, and there they adjust for pre-snap formation (i.e., shotgun) as well. This is dependent on game charting data however, and we don’t have that for every snap of every game in FBS, so I don’t think that particular adjustment is included. Finally, in a similar fashion to ASR it’s scaled to an average of 100, and higher ratings are better. When calculated for the offensive line it’s presented as OALY, and when calculated for the defensive line (really, the defensive front seven) it’s presented as DALY.

A New Way To Determine The Edge

Every preview this season has featured a column called "The Edge" where I assess each team’s standing in a particular metric and determine which one has the upper hand. To do this, I eyeballed the ratings and ranks, made a determination based on some arbitrary criteria I decided on, and moved on. While this method ensures a unique assessment of the upcoming game, it’s not really justifiable from a statistical standpoint. That being said, the only methodology I’ve seen on SBNation that I felt was better is to classify it based solely on ranks, such that any differences in excess of 20 ranks resulted in the higher-ranked team having the edge, whereas anything within 20 ranks is considered a push. It’s better than my old method, but I still don’t really like it.

Instead, I’ve ginned up a statistically justifiable method based on standard scores. It produces more pushes than my previous method, and I think it’s better than what I was doing. Strangely enough2, most of the time the "20 ranks" method seems to line up with it nicely (and is a heck of a lot simpler to calculate), but now there’s some (published) justification for it. If you’re interested in the boring math stuff, it’s presented in all its nerdy glory behind the button.

2 | It’s almost like someone’s done this before.

Normality

The first step to calculating standard scores is to verify whatever set of data you’re working with is normally distributed. I knew most of the metrics I discuss are normally distributed, because they are all opponent-adjusted and that process is a form of normalization. Just to be thorough I went ahead and calculated normality plots for all 28 metrics I cover in these previews, which you can see in all of its screen-stretching excellence here. I provided the R2 value for each plot, and you’ll note most of them are quite high, which indicates linearity and in turn normality. The one exception is OASR, which is thrown way off by one team — New Mexico State, who gave up only 10 sacks against 452 passing attempts on the season. If you look at the rest of the plot it’s pretty clearly linear, so I’m not too worried about it.

Standard Scores

The standard score is a simple concept, and one we've discussed before. Put another way, the standard score describes, for a given set of data, how many standard deviations a particular data point is from the mean of that set of data. This works for normally distributed data, which is conveniently what we have here. Calculating the standard deviation and mean for each metric couldn’t be easier, so cranking out standard scores for this type of stuff is no big deal.

Standard Normal Distribution, Percentiles, and the Final Comparison

With standard scores in hand, you can leverage the standard normal distribution to calculate percentile ranks for each rating of every metric. The standard normal distribution is just a normal distribution with a mean value of 0 and a standard deviation of 1, and is probably something you’ve worked with before. It might have been a long, long time ago in high school, but you’ve definitely seen it. At any rate, those standard scores can be converted into percentiles, which are the percentage of data points which fall under the standard score on the standard normal distribution. Calculating the percentiles is an unnecessary step — you can certainly make a comparison with just the standard scores — but percentiles are something you’re also all familiar with from your standardized testing days, so it’s a bit more relatable. The only arbitrary part of this comparison is how many percentiles you deem to be significant. I’ve chosen 10 here, as I felt 5 was too conservative (and too much like what I was already doing) and 15 or more was way too liberal.

To summarize, after converting each rating into a percentile rank, differences in excess of 10 percentile ranks is deemed significant, and the higher-ranked team is given the edge. For comparisons within 10 percentile ranks, the difference is considered negligible and is deemed a push.

So, is this better?

Yes! The comparisons are driven statistically rather than arbitrarily, and given the nature of the data it’s a supportable method as well. It also passes the smell test, which is important with statistics. Previously I made interesting choices like giving the edge in S&P+ to the #1 team over the #3 team and the edge in F/+ to the #1 team over the #5 team, among many determinations that were laughable in hindsight. Finally, this method produces similar results to the "20 ranks" method employed by Bill C. (and others), who is maybe just a tad more experienced with all this than I am.

That’s great and all, but can we get on with the preview already?

Right! This time we’re looking at the Chik-fil-A Peach Bowl, which will be contested by the SEC’s Ole Miss Rebels and Big 12 co-champion TCU Horned Frogs. The game kicks off Wednesday, December 31st at 11:30 CST / 12:30 EST in the Georgia Dome, and will be televised by ESPN. These are two teams that were knocking on the playoff door most of the season, with only tough luck (TCU) and a devastating injury (Ole Miss) preventing them from getting there. Some of the more talented3 members of the RBR staff did a great roundtable on this game for your reading pleasure. It’s good stuff — check it out if you haven’t already.

3 | That’s pretty much anybody else on the staff. I win at spreadsheets though!

Before we get to The Goods, let’s take a look at how these two teams got here.

The Resume — Ole Miss

As a reminder, the schedule tables do not include games against FCS teams, primarily because the advanced metrics are not calculated for those teams. In the event of a schedule that does NOT include FCS schools, the lowest-rated FBS opponent by F/+ will be omitted from the table so everyone’s on a level playing field. I should have done this for the playoff preview article from a couple weeks ago, but sometimes I’m not very bright — I’ll be making it up to the Buckeye fans later on this week.

#9 — OLE MISS REBELS
Team F/+ S&P+ FEI OF+ DF/+
OLE MISS 29.3% (4) 248.2 (7) 0.242 (7) 9.2% (21) 18.8% (3)
ALABAMA 38.4% (1) 270.7 (1) 0.315 (2) 19.8% (3) 19.7% (2)
MISSISSIPPI STATE 27.8% (6) 250.7 (5) 0.218 (10) 12.2% (14) 14.8% (9)
ALABAMA POLY 23.5% (12) 234.2 (16) 0.214 (11) 18.0% (5) 6.1% (38)
LSU 19.4% (18) 234.0 (17) 0.135 (27) 3.0% (48) 14.0% (11)
ARKANSAS 18.6% (20) 230.9 (19) 0.171 (17) 8.9% (23) 10.3% (19)
BOISE STATE 15.9% (25) 235.6 (13) 0.107 (36) 7.3% (31) 9.1% (26)
MEMPHIS 9.3% (40) 213.0 (38) 0.085 (41) -2.2% (71) 5.6% (40)
ORANGE TEAM 7.5% (43) 212.5 (39) 0.084 (42) -2.0% (69) 7.6% (31)
TEXAS A&M 3.8% (53) 209.2 (49) 0.027 (52) 7.5% (30) -5.4% (90)
UL-LAFAYETTE -9.8% (86) 200.3 (63) -0.141 (100) 0.2% (57) -10.8% (111)
VANDERBILT -18.4% (111) 175.9 (111) -0.190 (121) -13.5% (120) -4.7% (82)
AVERAGE 12.4% 224.3 0.093 5.4% 6.0%

(Bold) numbers indicate national ranking.

Observations

  • Average F/+ Opponent: West Virginia (F/+ #34)
  • Average S&P+ Opponent: Georgia Tech (S&P+ #23)
  • Average FEI Opponent: Florida (FEI #39)
  • Average Offense: Texas Tech (OF/+ #41)
  • Average Defense: Temple (DF/+ #39)
  • Best Win: Alabama (F/+ #1)
  • Wins against F/+ Top-25: 3 (#1 Alabama, #6 Mississippi State, #25 Boise State)

That’s, uh, quite a schedule. Thanks to being a member of the well-regarded SEC West, Ole Miss faced six teams among the Top-25 of the F/+ rankings and nine among the Top-50, which may be the best mark in the country4. Their average opponent in S&P+ was equivalent to Georgia Tech, who was Top-25 in that metric. They also logged arguably the best win of any team in the country by tipping the Crimson Tide back in October. That rough schedule took its toll, as a physical road loss at LSU started a three-game skid that included a horrific, season-altering injury to leading receiver Laquon Treadwell just before he crossed the goal line against API, preventing the winning points and producing a second loss. A shell-shocked Rebels squad coughed up six turnovers two weeks later at an improved Arkansas, completing the fall from likely playoff participant to also-ran. A win the following week against rival Mississippi State prevented a drop out of the New Year’s Six, and here we are. Make no mistake though — Treadwell is done for the year, but one benefit of the bowl season layoff is the opportunity to heal up all those nagging injuries. When healthy, this defense is arguably the country’s finest, featuring elite talent and veteran playmakers all over the field.

4 | Alabama Poly is right there with them, and probably some particularly unlucky ACC team.

Similarity — TCU

  • Offense — Rushing: API (RUSH OS&P+ #12)
  • Offense — Passing: Arkansas (PASS OS&P+ #28) (Really.)
  • Defense — Rushing: Arkansas (RUSH DS&P+ #7)
  • Defense — Passing: Boise State (PASS DS&P+ #40)

In terms of Rushing OS&P+ TCU is basically API, with only 0.2 points separating the two. The Tigers produced a balanced attack in that fateful game against the Rebels, generating 248 rushing yards (against 254 passing) at 5.4 yards a pop. The Study Hall from that game indicates API’s rushing efforts were slightly above the national average, which for Ole Miss is not a great game. As far as passing is concerned, I hesitate to take too much from the Arkansas game due to the nature of Arkansas’ offense, which included all of 16 passing attempts on the day. The next closest comp was Alabama, and let’s just say Blake Sims did not have one of his better performances that day. I’d expect a healthy Rebel secondary and pass rush to give Trevone Boykin some fits, but we’ll get into that later.

Ole Miss isn’t exactly known as a running powerhouse under Hugh Freeze, and the fact TCU’s rush defense is among the country’s best is not going to help matters. The Arkansas game also comes with the intangibles caveat I alluded to earlier, but the Rebels managed all of 63 yards at 1.9 yards per carry. I don’t need advanced stats to tell you that ain’t going to cut it. The passing comp is interesting as Boise State got to play with Schizo Bo, who chucked 3 picks alongside 387 yards at 10.8 per attempt. That was the first game of the season at a neutral site5, so the next closest comp might provide a better picture. That would be Alabama again, and Perfect Bo showed up for that one. It usually depends on which Bo Wallace takes the field, but in reality injuries will paint the picture here — we’ll also get to that in a bit.

5 | Rust, jitters, etc. Who thought after WVU that Blake Sims was the answer?

The Resume — TCU

#6 — TCU HORNED FROGS
Team F/+ S&P+ FEI OF+ DF/+
TCU 28.7% (5) 239.5 (10) 0.229 (8) 9.6% (20) 15.9% (6)
BAYLOR 27.0% (9) 243.9 (8) 0.218 (9) 13.7% (11) 11.7% (16)
OKLAHOMA 22.3% (13) 237.3 (12) 0.155 (22) 13.8% (10) 6.3% (36)
KANSAS STATE 17.5% (21) 220.3 (25) 0.146 (24) 10.5% (16) 2.1% (51)
WEST VIRGINIA 12.6% (34) 224.5 (22) 0.098 (38) 2.8% (49) 9.4% (25)
MINNESOTA 12.1% (35) 212.0 (40) 0.115 (34) 3.8% (44) 5.0% (42)
TEXAS 3.0% (57) 217.7 (32) -0.010 (64) -5.7% (94) 11.9% (14)
OKLAHOMA STATE -2.8% (68) 195.5 (73) -0.058 (80) -3.4% (77) -2.8% (71)
TEXAS TECH -5.4% (76) 193.6 (77) -0.077 (85) 5.2% (41) -9.9% (109)
IOWA STATE -11.8% (91) 175.9 (112) -0.084 (87) -0.8% (64) -11.3% (113)
KANSAS -17.4% (107) 180.3 (102) -0.134 (98) -12.0% (117) 0.2% (62)
SMU -32.0% (127) 153.3 (127) -0.256 (125) -19.2% (128) -10.9% (112)
AVERAGE 2.3% 204.9 0.010 0.8% 1.1%

(Bold) numbers indicate national ranking.

Observations

  • Average F/+ Opponent: East Carolina (F/+ #58)
  • Average S&P+ Opponent: South Carolina (S&P+ #57)
  • Average FEI Opponent: Utah State (FEI #61)
  • Average Offense: Iowa (OF/+ #56)
  • Average Defense: Colorado State (DF/+ #57)
  • Best Win: Oklahoma (F/+ #13)
  • Wins against F/+ Top-25: 2 (#13 Oklahoma, #21 Kansas State)

Well, there’s no argument to be made for TCU having a stronger schedule than Ole Miss, or even a remotely comparable one. Not to poke at a sore spot, but they don’t have a comparable schedule to Ohio State either, and that’s why they are here instead of the Sugar Bowl6. The fact they are as highly regarded as they are in spite of that schedule is useful information, and also not very surprising. TCU completely obliterated most of their schedule, including one of the most eye-popping scores I’ve ever seen in an 82-27 vaporization of Texas Tech7. In other words, they beat the crap teams the way an elite team should and beat the good teams along the way, and that’s all the advanced stats ask of you.

6 | Yes their loss was a heck of a lot better than Ohio State’s, but the Buckeyes had a better season. It happens.

7 | Which may or may not have been achieved unethically.

As yall are probably aware, TCU’s only loss was on the road to Baylor in a 61-58 shootout won by a last-second field goal. What you may not know is TCU allowed Baylor to come back from a 21 point deficit with 11 minutes remaining, which is hard to believe the most Baylor thing ever. At any rate, this is not your classic Gary Patterson squad with the suffocating defense and mediocre offense. 2014 TCU has Air Raid disciples Sonny Cumbie and Doug Meacham co-directing the offense, and they’re more interested in point production than point prevention this time around8. Watching the chess match between them and the outstanding Ole Miss defense will be a total blast.

8 | Not that their defense isn’t suffocating — in fact, it’s damn good.

Similarity — Ole Miss

  • Offense — Rushing: Minnesota (RUSH OS&P+ #41)
  • Offense — Passing: Baylor (PASS OS&P+ #11)
  • Defense — Rushing: West Virginia (RUSH DS&P+ #17), but not really
  • Defense — Passing: Texas (PASS DS&P+ #4)

As noted in the previous section, Ole Miss isn’t a very effective rushing team, and while Minnesota loves to run they aren’t very good at it either. They didn’t break 100 yards on the Horned Frogs, managing just 2.5 yards a carry. The passing numbers are a little more sobering, as Baylor absolutely torched TCU by the end of that game. To be fair, Baylor does that to everyone, and a brief glance at the box score suggests the Horned Frogs were able to contain the Bears for a good portion of the game. However, when all was said and done they allowed Bryce Petty and company to log 510 yards through the air, including nine completions in excess of 20 yards. It’s hard to replace elite cornerbacks9, and the loss of Jason Verrett to the NFL has undoubtedly taken a toll on the TCU defense. That being said, the Rebel offense showing up in Atlanta is not the one that blitzed the SEC West this season — but we’ll get to that in a bit.

9 | Cut to the Tide fans nodding solemnly.

Simply put, TCU has not faced a defense remotely like the Rebels. Ole Miss is top-5 in all four defensive S&P+ components — there is no soft spot on this defense. West Virginia is sneaky-good against the run, but they are still a full 16 points short of Ole Miss in Rushing DS&P+. A better comp may be TCU itself, depending on how much stock you place in practicing against a great defense every day. The Study Hall from the Texas game is illuminating, as TCU performed well below the national average in the passing game against a strong Texas defense. The key stat from that game was turnover points margin, +29 in favor of TCU — that’s Insanely High. Texas coughed up six turnovers in that game, allowing TCU to coast to an easy win. Take those away and things might have gotten interesting in a hurry. In short, expect the Rebels to give Trevone Boykin, Josh Doctson and Kolby Listenbee all kinds of fits.

The Goods

Overall Quality
TCU OLE MISS THE EDGE
F/+ 28.7% (5) F/+ 29.3% (4) PUSH
FEI 0.229 (8) FEI 0.242 (7) PUSH
S&P+ 239.5 (10) S&P+ 248.2 (7) PUSH
Spread -3 Spread +3 TCU

When TCU Has The Ball
TCU OLE MISS THE EDGE
OF/+ 9.6% (20) DF/+ 18.8% (3) OLE MISS
OFEI 0.324 (23) DFEI -0.699 (3) OLE MISS
OS&P+ 116.3 (16) DS&P+ 128.5 (5) PUSH
Rush OS&P+ 127.7 (13) Rush DS&P+ 140.0 (4) PUSH
Pass OS&P+ 113.7 (36) Pass DS&P+ 155.2 (3) OLE MISS
SD OS&P+ 116.5 (25) SD DS&P+ 138.6 (3) OLE MISS
PD OS&P+ 123.6 (27) PD DS&P+ 169.3 (2) OLE MISS
OALY 110.2 (34) DALY 121.7 (10) OLE MISS
OASR 125.4 (35) DASR 120.0 (32) TCU

When Ole Miss Has The Ball
TCU OLE MISS THE EDGE
DF/+ 15.9% (6) OF/+ 9.2% (21) TCU
DFEI -0.607 (6) OFEI 0.251 (34) TCU
DS&P+ 123.2 (13) OS&P+ 119.7 (12) PUSH
Rush DS&P+ 134.5 (6) Rush OS&P+ 112.7 (37) TCU
Pass DS&P+ 112.5 (34) Pass OS&P+ 133.9 (9) OLE MISS
SD DS&P+ 119.1 (18) SD OS&P+ 123.8 (16) PUSH
PD DS&P+ 121.6 (26) PD OS&P+ 126.4 (23) PUSH
DALY 124.9 (6) OALY 104.8 (56) TCU
DASR 104.1 (61) OASR 92.8 (82) TCU

The Matchup on Special Teams
TCU OLE MISS THE EDGE
ST F/+ 3.3% (10) ST F/+ 1.3% (39) TCU
FPA 0.568 (4) FPA 0.534 (18) TCU
FGE 0.465 (10) FGE -0.12 (79) TCU
KE -0.226 (24) KRE -0.108 (43) TCU
PE -0.16 (38) PRE -0.282 (121) TCU
PRE 0.053 (19) PE -0.227 (22) PUSH
KRE -0.197 (90) KE -0.346 (4) OLE MISS

(Bold) numbers indicate national ranking.

Wondering what all these terms are?

  • FEI: The Fremeau Efficiency Index, an overall team quality metric that is drive-based and opponent-adjusted. For a more detailed discussion of FEI, check out this section of the PTN Football Primer.
  • OFEI: The offensive component of FEI.
  • DFEI: The defensive component of FEI.
  • FPA: FEI Field Position Advantage, a measure of how much field position value a team earned against its opponents.
  • Fremeau Special Teams Efficiency Components - The special teams component of F/+ is based on Brian Fremeau’s Special Teams Efficiency, which is made up of the following five components of special teams play (per FootballOutsiders):
    FGEField Goal Efficiency, the scoring value per field goal attempt earned by the field goal unit as measured against national success rates.
    PREPunt Return Efficiency, the scoring value per opponent punt earned by the receiving team as measured against national return rates.
    KREKickoff Return Efficiency, the scoring value per opponent kickoff earned by the receiving team as measured against national return rates.
    PEPunt Efficiency, the scoring value per punt earned by the opponent's receiving team as measured against national return rates.
    KEKickoff Efficiency, the scoring value per kickoff earned by the opponent's receiving team as measured against national return rates.
  • ASRAdjusted Sack Rate, which is a version of sack rate (defined as sacks / [sacks + passing attempts] ) that has been opponent-adjusted. The metric is scaled based on an average rate of 100; the higher the rate the better. ASR is calculated for both the offense (OASR) and defense (DASR).
  • ALYAdjusted Line Yards, which is a measure of success in the running game specific to the line. This is accomplished by taking each carry by running backs only and weighting the yardage as follows:
    • Runs for a loss are weighted 120%.
    • Runs for 0-4 yards are unweighted.
    • Runs for 5-10 yards are weighted 50%.
    • Runs for 11 or more yards are not included.
    After the weighting process, the runs are further adjusted for game situation and opponent, and then averaged out per carry, resulting in adjusted line yards — a more detailed explanation of the entire process is available here. ALY is calculated for both the offensive line (OASR) and the defensive front seven (DASR).
  • S&P+: Another overall team quality metric, S&P+ is primarily play-based and consists of three components: Success Rate, Equivalent Net Points per Play, and a drive efficiency component. The "+" refers to opponent adjustments. For a more detailed discussion of S&P+, check out this sectin of the PTN Football Primer.
  • OS&P+: The offensive component of S&P+.
  • DS&P+: The defensive component of S&P+.
  • Rush OS&P+: S&P+ calculated on rushing plays for the offense — a good measure of a team's effectiveness at running the ball.
  • Rush DS&P+: S&P+ calculated on rushing plays for the defense — a good measure of a team's effectiveness at stopping the run.
  • Pass OS&P+: S&P+ calculated on passing plays for the offense — a good measure of a team's effectiveness at throwing the ball.
  • Pass DS&P+: S&P+ calculated on passing plays for the defense — a good measure of a team's effectiveness at defending the pass.
  • PD: Passing Downs, defined as later downs with medium yardage or more to go (3rd, 4th downs in excess of 5 yards to go), as well as 2nd down with more than 8 yards to go.
  • SD: Standard Downs, defined as all downs that are not Passing Downs.
  • SD OS&P+: S&P+ calculated on standard downs for the offense — a good measure of a team's offensive effectiveness on earlier downs and short yardage.
  • SD DS&P+: S&P+ calculated on standard downs for the defense — a good measure of a team's defensive effectiveness on earlier downs and short yardage.
  • PD OS&P+: S&P+ calculated on passing downs for the offense — a good measure of a team's offensive effectiveness on later downs and long yardage.
  • PD DS&P+: S&P+ calculated on passing downs for the defense — a good measure of a team's defensive effectiveness on later downs and long yardage.
  • F/+: The F/+ combined ratings combine FEI and S&P+ into one metric that serves as Football Outsiders' official rankings for college football. For a more detailed discussion of F/+, check out this section of the PTN Football Primer.
  • Off. F/+: The offensive component of F/+.
  • Def. F/+: The defensive component of F/+.
  • ST F/+: The special teams component of F/+.
  • Swanson Giddiness Index: Easily the most accurate predictor of success in college football, the Swanson Giddiness Index is a qualitative, completely unsupportable metric that is presented via the tone of that week's image/animated gif of Ron Swanson — beloved Parks and Recreation character and official spirit animal of Processing the Numbers.

Wondering what all of this means? Check out the PTN primer!

So, what do we know?

Well, that first table probably tells you all you need to know — these are two well-matched teams. Ole Miss is slightly better regarded in the overall metrics, but not to a significant degree. Vegas isn’t on board with any of them, as they’ve got TCU by a field goal in what will probably feel like a road game for the Horned Frogs.

As we talked about the similarity section, TCU has not faced a defense remotely like the Rebels except in practice. Ole Miss does not have any overt weaknesses on that side of the ball, and as a result they match up favorably with the TCU offense. Boasting the stingiest scoring defense in the country, they have a significant edge according to FEI, and were just a smidgen away from having one in S&P as well. TCU will have to try to get their yards on the ground, because throwing on Ole Miss is not generally an advisable activity — Cody Prewitt and Senquez Golson are all-everything this year for a reason. S&P+ has the rush game as a push, but Ole Miss has a rather significant edge in ALY. That’s not terribly surprising as ALY does not consider quarterback runs, and almost a quarter of TCU’s rushing production came from Boykin. Aaron Green has a rather eye-popping yards per carry average, but again that was generated against a middling slate of rush defenses. TCU does have an edge in ASR, but it’s very slight, and I suspect it’s mostly because of the weirdness with the OASR distribution this season.

Ole Miss has the better defense but not as good of an offense, and that’s reflected by a lot of purple in the third chart. The Rebels will not run on TCU, not that they’ve really run on anybody this season. One area TCU appears a little softer in is the passing game, ranking 34th and 26th in passing and passing downs S&P+ respectively. This isn’t overly surprising as they rank a middling 61st in DASR, although that’s enough for an edge on Ole Miss in that department. If10 Good Bo shows up and can find ways to consistently get it to Evan Engram and Cody Core, Ole Miss will be able to move the ball on TCU. If he doesn’t…

10 | IF. BIG NEON LETTERS IF. MID-90s BLINKING TEXT IF.

Ole Miss has a solid special teams unit, but TCU has a significant edge on the Rebels. These defenses are both rather good on a drive basis, and as a result this game may come down to field goals. That’s a big win for the Horned Frogs, as they have a tremendous advantage in that aspect of the kicking game, boasting the #10 kicker in the country. Jaden Oberkrom hasn’t been asked to kick anything in the 50s (his long is 47 this year), and he’s missed a chip shot here and there, but 85% conversion rate is not terrible for a college kicker. Kickoffs may be interesting, as Ole Miss is very good at covering them (#4 overall) and TCU is very poor at returning them (#90 overall); the difference is much smaller on the other side. Punting also favors TCU, so barring turnover pandemonium I’d expect the Horned Frogs to win field position in this one.

This is clearly a defensive struggle on paper, and after looking over the numbers I’d say Bo Wallace is the key player in this game. This is Wallace’s final game as a Rebel, and if he comes in flinging picks everywhere Ole Miss has no chance of winning, period. As noted, he has to play well for Ole Miss to do anything on offense, and if he does the Rebels are certainly capable of keeping it close with TCU. Their defense is certainly up to the task, but if this comes down to special teams and field position, they’re in trouble.

Any intangibles to consider?

This is a neutral-site field in a dome, so location and weather are non-factors. I suspect you’ll see a few more Ole Miss fans at the stadium given the location, but it’s not going to be like playing at Vaught-Hemingway11. The big thing here is injuries, which have decimated Ole Miss in the latter half of the season. Treadwell is one of the country’s better receivers, and his absence is huge for the Rebels — if he were playing, I’d pick Ole Miss without a second thought. #2 wide receiver Vince Sanders is also out, so the pass game will be all on Engram and Core. In more good news, starting left guard Aaron Morris will also miss the game with a torn ACL; after shifting pieces around Ole Miss will start freshman Rod Taylor at right guard. As far as I know the defense will be at full strength.

11 | Not that Ole Miss is known for having a crazy home field a la LSU, but you get what I mean.

On the other side, TCU will still be without the services of B.J. Catalon, who suffered a concussion against West Virginia back on November 1st and hasn’t yet recovered. That’s when Green’s production stepped up, and he’s performed admirably as the lead back. The Horned Frogs will also be missing defensive end Mike Tuaua, who suffered a shoulder injury against Iowa State. Overall, TCU appears to be the healthier team going in.

THE PICK: TCU Horned Frogs — hate to pick against the SEC but I just don’t see it. I hope I’m wrong!

ROLL TIDE