We know that with each turnover a team’s odds of winning is reduced. Turnovers not only stop your team from scoring in the short term, they also reduce the number of plays and number of yards a team gains, two key ingredients to winning.

The model I use to create the probabilities below uses 2,678 past games to give probabilities for future games based on estimated metrics.

The four scenarios below show an increase of 1 turnover each and a corresponding decrease in the other metrics. You can see how turnovers affect the win probability on the far right.

In those 2,678 games no team with the metrics in scenario 1 has lost and only 2.2% of those in scenario 2 lost.

The long and short of it is that if the Tigers don’t turn the ball over they’re going to win. Turn it over once and the odds are still highly in their favor.

Things get interesting around the 3 turnover mark, but the model I use shows Tiger fans can expect 1.5 turnovers on Saturday.

Obviously, you can’t have 1.5 turnovers, so I would expect 1 or 2 turnovers for Clemson on Saturday and that means the probability of a win is high in either case, but 15% lower if the Tigers turn the ball over twice.

In case you’re wondering, the tipping point is 5 turnovers – that’s where the probability dips below 50%.

**Legend:**

H/A: Home/Away

eYards: Estimated Total Yards Gained

eYPP: Estimated Yards Per Play

eTYP: Estimated Total Yard Differential from opponent

eYPPD: Estimated Yard Per Play Differential from opponent

eTO: Estimated number of turnovers

ProbW: Probability of winning based on estimated statistics

You can also download current and previous episodes and subscribe to the Podcast via iTunes by clicking here.

Thanks to Brandon for participating and Clemson graduate Adam Eargle for the podcast artwork and all SUR graphics. Need a graphic artist? Check out some of Adam’s work here.

]]>What’s not identical is the average defense faced. Stoudt has faced tougher defenses on average by a long shot due to Watson’s injuries.

Note: The information in the category of “Plays” is taken directly from play by play data and drive summaries and may not equal “snaps”. Also, drives where both individuals quarterbacked are excluded.

]]>The Tigers streak of not losing as a favorite ended on Saturday. From November of 24, 2012 until Saturday 111 teams not named Clemson lost 297 times as a favorite.

Here’s the list and number of losses as a favorite in that time frame.

]]>They eyeball test looked worse than that. I suspect that’s because the model I use accounts for turnovers, but doesn’t “know” two of those were pick sixes.

The turnovers reduced the odds of a Clemson win to 4.7%. The fact that two were returned for touchdowns (unknown to the model) moved that number significantly closer to 0.

]]>

Here are his numbers for those two games combined.

Perhaps the most exciting number on the page is the percentage of explosive plays (16+ yards) that those 32 passes generated.

]]>