Cardinal Numbers: Takeaways From the USC Game

For those of us obsessed with the collection, analysis, interpretation and presentation of data, Terry Johnson sobers us up in time for TCU without raining on the Cardinal's pigskin parade following last Saturday's epic upset win over 41-point favorite USC. Read on for a deeper appreciation of the impact of turnovers and the need to light a sustainable fire under the struggling Cardinal offense.

Stanford's stunning 24-23 win over #1-ranked USC provides a wealth of story lines.... Stanford pulls off one of the biggest upsets in the history of college football.... Tavita Pritchard leads Stanford to victory in his first college start.... Mark Bradford dedicates the game to the memory of his father and catches the game-winning TD pass.... Jim Harbaugh proves he can walk the walk to back up his talk.... Compelling stories, all of them.

Those stories, and others like them, are the main topics of discussion this week, as they should be. Those will continue to be the stories for which this game is remembered. Given the magnitude of this win, any statistical analysis is secondary. Many years from now, people still will be talking about this game, and they won't care what the statistics were.

However, for those who are interested in understanding exactly how Stanford was able to pull off such an enormous upset, there is one obvious, compelling story line that emerges from the statistics: turnovers.

The number that mattered in this game was five – the number of turnovers by USC. Stanford committed just one turnover, giving Stanford a turnover margin of +4. From a statistical standpoint, that was pretty much the ball game. A team with a turnover margin of +4 just about always wins.

Aside from turnovers, Stanford's statistics against USC were unimpressive. In fact, some of Stanford's basic offensive and defensive statistics against USC bore more than a passing resemblance to Stanford's statistics in the previous week's game against Arizona State – a game that Stanford lost 41-3:

Key Stanford Statistics
Against Arizona St. & USC
 vs. ASUvs. USC
Yards gained235235
Yards allowed440459
First downs1316
First downs allowed1819

Stanford's statistical performances against Arizona State and USC were similar in some basic ways. Yet the results, obviously, were dramatically different. That's what five turnovers forced by an aggressive, play-making defense can do.

By forcing turnovers, Stanford was able to compensate for its difficulty in stopping the USC offense. For example, in the second half, USC punted just once, on its first possession of the half. USC's other seven second-half drives resulted in two touchdowns and five turnovers. USC wasn't able to score enough points to put the game out of reach because Stanford kept taking the ball away. That allowed Stanford to stay close.

The turnovers also helped Stanford compensate for its continuing difficulties on offense. Stanford's offense struggled to gain any traction against USC. But turnovers gave the offense a helping hand. Stanford put seven points on the board on one "pick-six" interception. Later, Stanford set up its winning touchdown drive by returning another interception into USC territory.

Analysis of Stanford's turnover margin over the years shows that the team that wins the turnover battle usually wins the game. I've done a game-by-game analysis of Stanford's turnover margin in the 142 games played by Stanford from the beginning of 1995 to the present, including the current season to date. In those games, the team that won the turnover battle had a record 92-27, which is a remarkable winning percentage of .773. (The turnovers were even in the other 23 games.) The analysis shows that the winning percentage increases as the size of the positive turnover margin grows:

Record of Team That
Won the Turnover Battle

(Stanford Games 1995-2007)
+4 or better12-01.000

As you can see, when a team has had a turnover margin of +4 or better, that team has won every time. Stanford's turnover margin of +4 against USC therefore was a very strong indication of a likely Stanford victory. Even before Bo McNally's interception on the last play of the game, Stanford's +3 turnover margin had put Stanford in an area in which there has been a very comfortable winning percentage (.864). In fact, a team with a positive turnover margin of any size has had a high winning percentage.

The relationship between winning the turnover battle and winning the game was strongest under Willingham, and not as strong under Teevens and Harris. That seems to have been the case because Stanford's teams under Willingham were almost always competitive, so that an edge in turnovers often was enough to make the difference in a game's outcome, one way or the other. On the other hand, under Teevens and Harris, Stanford had too many uncompetitive games, too many blow-outs. Too many times, the game wasn't close enough for turnovers to matter. Too often, winning the turnover battle just wasn't enough to overcome the team's shortcomings and put the game in the win column. Turnovers were necessary just to make those teams competitive. And when those teams lost the turnover battle, they hardly ever won the game.

 Record with
Record with
Willingham31-4.886 8-23.258
Teevens/Harris12-11.522 3-23.115

The statistical relationship between turnovers and winning is strongest on a game-by- game basis. However, there also has been some relationship between the turnover margin for a full season and Stanford's record for the season. Since 1995, Stanford teams that ended up with winning records have had an average turnover margin for the season of +6. On the other hand, Stanford teams that ended up with losing records during that time have had an average turnover margin of -1.4.

A positive turnover margin for the season makes a winning season more likely, but it has not been a guarantee of a winning record. Four Stanford teams over the last 12 seasons have had a positive turnover margin for the season, but ended up with a losing record. Three of those four teams were Teevens/Harris team, which just weren't competitive enough to convert a positive turnover margin into a winning record.

On the other side of the ledger, in the last 12 seasons, every Stanford team with a negative turnover margin for the season has ended up with a losing record. That's just one more reason to value a positive turnover margin.

Stanford's ability to force turnovers by USC was a highlight of the game, but it's not a sustainable strategy over the long run. Stanford isn't likely to force that many turnovers with any regularity. Stanford has had a turnover margin of +4 or better only eight times in the last 12 and a half seasons. We're not likely to see another game like the USC game anytime soon. Stanford will need to find other ways to win games.


Not to take anything away from the glory of this past Saturday's historic victory, but Stanford's offensive statistics against USC were not good, to say the least. As indicated above, Stanford's offense had an anemic 235 yards of offense for the second straight game. That's pathetic. Last year's all-time worst Stanford offense averaged 232 yards per game. For the last two weeks, obviously noting a number of spectacular plays down the stretch, the Cardinal offense has returned to that dismal level. Before the late-game heroics against the Trojans, Stanford had played nine consecutive quarters of football without generating an offensive touchdown We're back in the "Harris Zone" and that absolutely must change.

There were some extenuating circumstances against USC, because Stanford was forced to use its inexperienced back-up quarterback. A game, but very green Tavita Pritchard started slowly and obviously grew increasingly effective as the game progressed.

Stanford's first nine possessions ended with eight punts and an interception (not counting the one-play possession to run out the clock at the end of the first half). Stanford gained only 59 yards of total offense during those possessions, with only five first downs. Pritchard completed just 3-of-14 passes for 29 yards on those nine drives, and took four sacks. Stanford's rushing offense generated just 30 yards on 23 carries. For the first 42 minutes, Stanford's offense was dead in the water. That is a tribute to USC's outstanding defense, but it is also cause for concern.

When the Cardinal got the ball with about three minutes remaining in the third quarter, Stanford and Pritchard suddenly hit their stride. Stanford's next three possessions resulted in three good drives, ending in two Stanford touchdowns and a field goal. Stanford gained 176 yards of total offense and 11 first downs in barely more than a quarter. Pritchard completed 8-of-16 passes for 120 yards, and was not sacked. Stanford ran for 56 yards on 14 carries. Thus, the offense showed excellent progress as Pritchard grew more comfortable.

One obvious change in Stanford's offensive approach against USC was the inclusion in the game plan of some running plays designed for the quarterback. The coaching staff added some bootlegs, draws, roll-outs, and so forth to take advantage of Pritchard's mobility, and it certainly worked. Excluding sacks, Pritchard had 10 carries for 49 yards. Most of those rushing attempts came on designed plays, rather than scrambles. The quarterback's ability to run the ball added a new dimension to the offense, adding to an arsenal that clearly had been short on weaponry.

There's no doubt that Stanford needs to do a much better job on offense. For the season, Stanford is now averaging 355 yards per game of offense, which is 9th in the Pac-10 and 83rd in the nation. The trend is in the wrong direction, with two straight games of 235 yards of offense. Stanford is scoring just 22.4 points per game, which is last in the conference and 92nd in the nation. Stanford needs to prove it can generate some offense and score some points, and needs to do it for a full game.


Since the first game of the season, this column has been tracking three metrics that should be characteristics of an "attack defense": tackles for loss, sacks, and turnovers created. As we've noted since the first game, Stanford is generating far more tackles for loss than had been the case in previous years. That continued against USC with six more tackles for loss. Stanford is now averaging 7.8 tackles for loss per game, which is third in the Pac-10 and 21st in the nation, compared to 3.8 tackles for loss per game last year.

Stanford didn't have many sacks in the first few games of the season. But the pass rush is now coming on strong, with 10 sacks in the last two games. Stanford is now averaging 3.2 sacks per game, which is 2nd in the Pac-10 and 15th in the nation. That compares to 1.2 sacks per game last season. With 16 sacks this year, Stanford already has exceeded its sack total from last year (14 sacks in 12 games). Very encouraging.

Prior to the USC game, Stanford wasn't creating many turnovers – only five turnovers in the first four games. Obviously, that changed in a big way as Stanford forced five USC turnovers. Stanford now has created 10 turnovers in five games, which ranks 5th in the Pac-10 and 50th in the nation. That's not great, but at least it's a marked improvement over what we saw in the first four games.

However, while Stanford is having some success with the big plays that you would hope to see in an "attack defense," Stanford's defense continues to get burned far too often. Against USC, the defense gave up 459 yards. This the fourth time in five games that the defense has given up 440 yards or more. Stanford is now giving up an unaccceptable 455 yards per game, which is last in the Pac-10 and 105th in the nation. Stanford is allowing 32.8 points per game, which is ninth in the Pac-10 and 98th in the nation. That's just not good, no matter how you look at it. Not trying to burst anyone's bubble, just to provide a sobering reality check.

Once again, as in previous games, Stanford's defense was victimized by big plays against USC. Stanford gave up touchdown passes of 63 yards and 56 yards, and a 47-yard pass that set up a field goal. Those three plays accounted for over a third of USC's yards (166 yards) and 17 of its 23 points. Giving up big plays may be the price of an attacking defense, but it's a pretty hefty price.

Random Numbers

USC had given up only three sacks in its first four games, but Stanford sacked the USC quarterback four times...

Stanford's conversions on 4th-&-20 and 4th-&-goal on its last drive were Stanford's only fourth-down conversions this season. Before that drive, Stanford was 0-for-5 on fourth down conversions this year...

Mark Bradford now has 139 career receptions. He needs only two more receptions to catch up with Glyn Milburn, Chris Walsh, and Ken Margerum, who are tied for ninth on Stanford's career reception list with 141 catches each...

Anthony Kimble's 752 all-purpose yards in the first five games already are more than Kimble's team-leading total of 734 all-purpose yards last season...

Chris Hobbs had his best game ever returning punts against USC, with two returns for 47 yards, including a career-long 29-yard return...

Stanford is second in the Pac-10 in "red zone" efficiency, scoring points on 92% of its red zone possessions. The problem, however, is that Stanford has been in the red zone less frequently than any other team in the conference...

Last year, Pannel Egboh and Tim Sims led the team in tackles for loss with 5.5 each for the season. This year, Clinton Snyder already has 6.0 tackles for loss, and Pat Maynor has 5.5 tackles for loss...

Stanford players have received four "Pac-10 Player of the Week" awards so far this season. That equals the total number of "Player of the Week" awards for Stanford players in the last three seasons combined...

Are you fully subscribed to The Bootleg? If not, then you are missing out on all the top Cardinal coverage we provide daily on our website, as well as our full-length feature articles in our glossy magazine. Sign up today for the biggest and best in Stanford sports coverage with (sign-up) and The Bootleg Magazine (sign-up)!

The Bootleg Top Stories