Notre Dame fans need treatment for hypertension. The last three games have been decided in the waning moments, with the the last two needing gifts from the opposing team to secure a victory.

Against Michigan State it was quarterback Kirk Cousins overthrowing a wide open Larry Caper in the back of the end zone followed by a poor pass on the next play that was intercepted by safety Kyle McCarthy.

This week it was a head-scratching timeout from Purdue head coach Danny Hope that allowed quarterback Jimmy Clausen to avoid spiking the ball and run an extra play for the deciding score of the game.

To be certain, it was a defining moment for the Irish signal caller.

On the road in West Lafayette and without many key personnel, the Irish offense played well enough to win. The Irish defense was also improved, holding Purdue running back Ralph Bolden (second in rushing yards per game entering Saturday’s contest) in check for the duration of the evening.

Notre Dame dominated the time of possession (8:50 advantage), first downs (23 to 16) and plays (79 to 64), but questionable coaching strategy (more on this later) and poor fourth quarter defense nearly lost the game as the Irish surrendered a ten point lead before notching the winning score with 0:25 remaining.

Offense

It wasn’t like the first three games, but on the road, without running back Armando Allen and wide receiver Michael Floyd, and with a half-injured Clausen, the Irish offense was still relatively effective and efficient.

The second quarter was particularly dominant as head coach Charlie Weis used the previously ineffective Wildcat formation and a physical running game en route to 14 unanswered points.

Notre Dame gained 383 yards at a modest 4.8 yards per play. Weis used balanced play calling (43 runs and 36 passes) that resulted in nearly equally balanced production in both yards (167 yards rushing to 216 passing) and first downs (ten first downs rushing, 11 through the air).

The efficiency was also on display as the Irish recorded only eight negative plays, scored on four of five trips inside the red zone, and converted 40 percent of third downs despite needing more than five yards 80 percent of the time.

About the only negatives were allowing four sacks and a reliance on the big play. Six big gains accounted for 120 yards (31.3 percent of the total offense) without which the Irish only averaged 3.6 yards per play, the lowest total of the season and about the average for last year.

Given how banged up the offense was, this is entirely understandable.

Rushing

The Irish running game was productive, even though Purdue knew it was coming.

Notre Dame gained 167 yards on 43 attempts for a 3.9 yard per carry average, numbers skewed due to the four sacks. Without sacks the average rushing attempt was good for 4.8 yards per carry.

Additionally, the Irish recorded four big runs for 75 yards, the most for both categories in any game this season.

Running back Robert Hughes led the way with 68 yards on 15 carries (4.5 yards per attempt) and a touchdown, reminding many of the promise he showed as a freshman. Wide receiver Golden Tate also chipped in 57 yards on only nine carries (6.1 yards per carry) and another touchdown.

Passing

The Irish had their worst passing performance of the young season, but it was largely expected given the circumstances surrounding Clausen’s injury and the absence of Floyd.

Clausen and backup Dayne Crist combined to complete 55.6 percent of 36 passing attempts for season lows in yardage (216), yards per attempt (six), yards per completion (10.8), big gains (two), and big play yardage (45). The yards per attempt and completion dip to five and 9.5 respectively without the two explosive plays.

But the passing game wasn’t about production, it was about closing.

Playing through noticeable pain and with limited mobility, Clausen delivered when it mattered most. It wasn’t his most efficient or productive outing (57.7 completion percentage, 171 yards, one touchdown, one interception), but it certainly was gutsy.

The Irish offensive line also had their worst pass protection outing of the year, allowing four sacks on only 36 passing attempts (one sack per nine attempts). However, Clausen’s injury certainly played a part in those rather dubious results.

Defense

It’s no coincidence that arguably the best defensive performance of season came when the Irish played well on first down and against the run.

Sporting improved—albeit not great—tackling and a more conservative game plan from co-defensive coordinator Jon Tenuta, Purdue averaged only 4.8 yards on first down. Of Purdue’s 26 first down plays, 13 resulted in two or fewer yards.

Despite this more conservative approach, big plays were problematic and prevented what could have been a fairly flawless outing.

Purdue gained 363 yards on 64 plays for 5.7 yards per snap. However, eight plays went for 199 yards and accounted for almost 55 percent of the total offense. Without these plays the Irish defense held Purdue to under three yards per play.

Many of these gains were on third down as the Irish allowed a 46.7 percent third down efficiency despite keeping Purdue in third and five-plus yards on 86.7 percent of their attempts.

Rushing

Over 40 percent of Purdue’s plays were runs that accounted for only 20.4 percent of their total yardage.

The Irish front seven played their best game of the year holding running back Ralph Bolden to 75 yards on 17 carries. The 3.9 yards per attempt pales in comparison to the 6.8 yards per carry he averaged entering the weekend.

As a team Purdue only rushed for 74 yards at 2.8 yards per carry. Subtracting the two Irish sacks this average increases to 3.8 yards per attempt, but falls to 1.8 yards per rush when three big runs of 55 yards aren’t included.

Moreover, the Irish only allowed 2.5 yards per rush on first down, a dramatic improvement from their performance against Nevada, Michigan and Michigan State.

In other words, the run defense was stout against a good rushing team that averaged 210.7 yards per game and six yards per carry coming into the contest. More production like this is needed in the future.

Passing

While the run defense improved, the secondary play continues to be a work in progress. Soft coverage was still problematic for an Irish defensive backfield that allowed three scores through the air.

Quarterback Joey Elliott completed nearly 58 percent of his passes for 289 yards and three touchdowns at 7.6 yards per attempt and 13.1 yards per completion. The Irish also surrendered five big pass plays for 144 yards (28.8 yards per play). Without the big plays the numbers are far more respectable (145 yards, 4.4 yards per attempt, and 8.5 yards per completion).

Despite blitzing with less frequency, Notre Dame did manage to notch two sacks, with freshman Manti Teo getting the first of his young career.

Special Teams

Kickoff coverage continues to be a strength for the Irish as Notre Dame allowed only 86 yards on five returns (17.2 yards per return). Kicker Nick Tausch is also improving and averaged 67.4 yards per kick Saturday.

But the punting production has significantly deteriorated and the Irish covered punts poorly for the first time this season. Eric Maust averaged only 37.8 yards per boot, but netted only 25.5 yards per punt as Purdue gained 49 yards on two punt returns (24.5 yards per return).

The Irish must find an answer for the punting woes as they are surrendering valuable field position at every exchange.

Summary

It is difficult not to question Weis’ play-calling and coaching decisions in this game.

The Irish scored 14 unanswered points in the second quarter, largely due to the success of a rejuvenated Wildcat offense, creative use of Tate in the backfield, and inspired running by Hughes. Despite this success, these packages and plays were largely absent in the second half.

What happened to “if it ain’t broke, don’t fix it?” The Boilermaker defense didn’t adjust and or prove they could stop it, Weis simply stopped calling it. Why abandon something that is working in lieu of an offensive strategy that places the burden of execution on a backup or half-injured quarterback?

Additionally, Weis continues to suffer from two strategic coaching deficiencies.

First, he seems to view punting and field goal attempts as offensive failures. Second, he either believes the best chance to win is by taking uncalculated risks on offense or has a control fixation and needs to make the play call that determines the outcome of the game.

The former left valuable points on the field early in the game, while the latter conspired to put the Irish defense in a bad position in the fourth quarter.

After Purdue went the length of the field and scored on their opening drive, Clausen directed the Irish offense to the Boilermaker 22 yard line. With one yard to go on fourth down Weis elected to gamble rather than attempt a 39-yard field goal.

Irrespective of the play call (running from the shotgun on fourth and one?), it is decidedly unwise to leave points on the field early in a road contest with an offense missing several critical players.

The more egregious error, however, occurred at the end of the third quarter.

The Irish defense had performed well except for the opening drive of the game. Notre Dame was up two scores with less than 19 minutes to play. It was fourth and ten on the Purdue 34 yard line. And Weis decided to gamble again and go for it.

The success rate of converting a fourth and ten is minimal, and even smaller with a backup quarterback.

Backup quarterback Dayne Crist was sacked and the Boilermakers used the momentum—and good field position—to reel off a scoring drive that covered 59 yards in only 3:21.

At times gambling isn’t needed to win. At times it’s better to play the percentages. At times it’s better to punt, pin the opposition deep, force them to move the ball the length of the field, and maximize the chance of success for the defense.

One of those times is on the road, with a backup quarterback, facing fourth and ten, up two scores, with less than 19 minutes to play, and a defense that has held the opponent to one touchdown.

The Irish are 3-1 but the three victories have come against teams that have combined for only two wins (against Toledo and Montana State). With the exception of USC the schedule doesn’t figure to get much more difficult, but improvement will be needed to continue winning.

Luck, as much as skill, is responsible for the last two victories. And the ball won’t always bounce in Notre Dame’s favor.