A reasonable title of this game would be Georgia Tech vs the UNC amateurs, UNC was still missing 12 players off of the Agent and Academic scandals that haunted the program last off-season. UNC struggled at the beginning of the year before their defense got all of their players back and became an above average defense. Luckily Georgia Tech played them early in the season and was able to take advantage with a 30-24 win. Inside we are going to look at the stats using Equivalent Points.

Continue reading this post »

I wanted to take a look at Georgia Tech's past season game by game to try get a better picture of how GT did than just saying our season was a disappointment. With the new excellent SBNation blog Football Study Hall football fans are able to learn more about their teams and the game of football than ever through the use of statistical analysis. As a Georgia Tech student this fits me perfectly. Football Study Hall introduced a concept called Equivalent Points. Equivalent Points tabulates all the results from each play for a college football season and assigns a point value for each yard line based on the average points scored when teams had the ball on that yard line. For a better explanation go here. I took the list of EqPoints and started charting Georgia Tech's game against Kansas using the play by play data. Here is what a sample drive looked like using the point values:

What this allows me to do it to take every play from a game and assign to it a point value for the amount of yards gained. However, Equivalent Points only takes into account the field position change. I say field position change instead of yardage change because Equivalent Points does take into account that a 10 yard gain from your opponent's 20 is worth more than a 10 yard gain from your own 20. This means Equivalent Points has no context, it does not matter what down it is or what distance it is or how much time is left on the clock, it is just the equivalent points expected from possessing the ball at that yard line, which is still very valuable information. With this information we can tabulate Georgia Tech's expected points for an entire game based on its change of field position. Here are the results:

While the total points is pretty close to the 25 we actually scored this number does not take into account turnovers and I am still unsure of how to use the data correctly. What I am more interested is the per play breakdown. Obviously our bread and butter is the run and it is reflected both in how many plays we ran, 51 runs to 16 passes, and the total points we accrued, 18.43 to 6.49. However, when looking at our per play numbers we actually earned more points per pass play than per run play. Alot of this can be credited to 2 passes, a 46 yard pass to Embry Peeples from our own 4 worth 1.268 points and a 40 yard touchdown pass on the same drive to Stephen Hill worth 4.343 points. With out those two passes our Equivalent Pass Points drops to about -.1 points. Again, this makes sense, we have a very high risk high reward passing game. While I very much enjoy this type of analysis and making these write-ups I do realize that there is very little context for these numbers. Equivalent Points does not take into account down, distance, or game clock. It is merely an average of the points gained from a team simply possessing the ball at that yard line. Luckily Football Study Hall has given us 2nd level Equivalent Points which take into account the down that the play is on as well as the yard line. Lets take a look at the same categories as before:

2nd Level Equivalent Points are much more volatile than 1st level, after looking at the data it appears that the biggest swings in points is from gaining a 1st down. This does make sense, converting for a new set of downs gives the offense 3 more chances to score before they have to decide if they want to go for it. Maintaining those chances is in essence the main goal of an offense, continuing to get chances to score. What can we learn from this data? Our predominately rushing offense seems to do a good job on first down to set us up for successful conversions on 2nd and 3rd down, as opposed to actually converting for a new set of downs on 1st down.  Other than that, I really do not know. Perhaps PJ is right by being so gutsy on 4th down? 2nd level points seem to reward him for converting a good percentage of these 4th downs. What do you guys think?  If you have any questions or want more information comment below and I will try and discuss things with you as best I can. I plan on doing this for every D1 game we played this year and if I have the time or energy I will do the defense as well. I encourage anyone who is interested on Equivalent Points to read up on them and start game charting on your own. It is a great way to pass the time until there is real football to watch.

Here are the breakdowns player by player:

As you can see Nesbitt is pretty important, it will be interesting to see the differences between Nesbitt and Washington.

This was a question a few of my friends and I recently discussed and I wanted to know what others thought. After the jump I'll talk about my answer.

Continue reading this post »

           I recently read the article from the Football Sabermetrics Fanshot about Expected Win Probability (EWP) developed for baseball being applied to football. I personally love in depth statistical analysis and looking back at the past using some new methods. For those of you who did not read the article, which I recommend you do, EWP is the square of the number of runs (points) scored divided by the sum of the square of runs scored and the square of runs allowed. 

      While I love stats I do not pretend to know the ins and outs of a stat but a thorough explanation is given HERE. Anyway this stat gives a baseline for how many wins a team is expected to earn during the season (EWP*# of games=Expected # of wins, for last season our EW was .65*14 = 9.1) any variance is attributed to “luck” or some skill that helps win close games (great kicker, clutch QB, CPJ winning game on 4th down, etc…). What I wanted to do with this stat was try and apply it to individual players. I wanted to determine exactly how many wins Nesbitt and Dwyer contributed this season. This is an impossible question to answer. I will explain why pretty much all of my assumptions are wrong as I work through the post but the number one problem with individual player football stats is that it is near impossible to completely separate one player’s contributions from the 10 other players on the field. BUT for this post lets assume we can.  The first question I had to answer is how do I separate the points that Nesbitt contributed from the rest of the team. What I did is calculate our EWP with Nesbitt’s points removed (from both the numerator and denominator) and then subtract that EWP without Nesbitt (.525) from our original EWP (.65) to get an EWP+(.125) and multiplied that number by 12(A historical comparison which I do later requires me to consider a 12 game season). That gives us Nesbitt’s Expected Wins Added (1.5). According to EWP Josh Nesbitt’s point contributions last season gave us 1.5 extra wins. I also calculated all of this data for each of Coach Paul Johnson’s QBs since 2004.

Continue reading this post »

           My last post dealt with how Georgia Tech's offense fared against the quality of the defense it was facing. In short we played a very tough schedule and fared very well against most teams, and the only times we underperformed as an offense our defense was not able to carry our team to victory. In this post I want to determine how our offense preformed compared to the rest of the ACC. Again I will use the yards per play stat, a simple and effective way to measure the success of an offense. I will also only use conference game statistics in this comparison just to try to weed out any variance. And all my stats are from cfbstats.com. A quick look at the overall stats shows that Georgia Tech's offense was 3rd overall with 6.1 yards/play, behind only FSU and surprisingly Virginia Tech. Although, looking further into the stats I would rank our offense 2nd in conference. Neither Georgia Tech nor FSU had the opportunity to pad their stats by facing their own bad defense (GTwill also noticed this), while Virginia Tech did not have to play arguably the ACC's best defense, their own. Beyond a general comparison I also wanted to look at each team and how other ACC team's offenses fared while facing this team, so I compiled the chart below.

           This chart is organized by the team we faced and the yards per play they allowed to other ACC teams. This shows a couple of interesting things. First off is that we only had one game where we did not perform better than the average ACC offense, Miami. That game was just as bad in the stats as it was to watch. Second is that our offense showed up to play both times against Clemson. We posted the 2nd and 3rd best performance against Clemson and in both games we needed every yard we gained. Finally I can not get over just how awful FSU's defense was, 4 teams gained more than 7 yards/play against them, which is considered to be an elite level of offense. Also their 2nd BEST defensive performance, 5.6 yards/play allowed against Maryland, was equal to Virginia Tech's WORST defensive performance, 5.6 yards/play allowed against Duke. One other thing I wanted to do was compare how well Georgia Tech's offense preformed against the SEC.

             Some quick observations; The 3 SEC defenses we faced would have ranked the 3rd, 5th, and 6th worst ACC defenses we faced, respectively. We also fared below average in only one game, and I think we all know which game that was. While we clearly were not playing the top defenses in the SEC we did very well against the teams we did play, more than a yard per play better than the average SEC team vs. Miss St. and more than a yard and a half better vs. Vanderbilt. I think this validates what all of us GT fans know, our offense works and it works at an elite level. The last thing that I take out of these charts is just how consistent our offense preformed. A simple scan of the charts puts Georgia Tech at or near the top of most every list. I can not wait to see how much more our offense improves in year three with Nesbitt at the helm.

              Thanks for all the comments in the last post, they were extremely helpful and insightful. Please continue to post any comments our questions. My next post will look at the explosiveness of our offense and whether or not this year was a "quick-strike" offense.

Now that football season is over it is time for some review of the great season Tech enjoyed this year. I am going to focus on the offense at first because anyone who watched a game this year saw how our defense preformed, we really don't need numbers to tell us. What I was interested in was comparing how well our offense preformed in comparison to the strength of the defense we faced. I measured the success of our offense by the success we had on each play. Besides scoring the most points possible an offense's main goal is to maximize it's plays. The best stat to measure this is simply yards-per-play. I used ESPN's game logs to take out kneel downs and when our starters weren't in. I also wanted to compare how our offense fared against the strength of the defense we faced. This year Georgia Tech faced 5 defenses ranked in the top 25 according to the S&P+ defensive metric. An explanation of this metric can be found here. Basically S&P+ determines the success of each and every play in a game. This provides a great way to determine a team's success on a play-by-play basis to eliminate factors such as number of possessions and bad starting field position. Once I had all the data I plotted them on this chart.

This graph simply shows what a lot of fans already knew. Our 3 worst offensive games were all games we lost, which is no surprise. We took care of business when facing truly awful defenses (Vandy, FSU, and Duke). To give you some perspective 5.5 yards a play is a good offense while 7 yards a play is a truly elite offense. In 4 games this year Tech's offense was elite. In only 2 games this year was our offense bad, of course in the Orange Bowl it was awful and I think this graph shows just how great that Iowa defense was. Our average yards-per-play was 6.22, a very good number. That ties us for 17th nationally and 2nd in the conference, only behind FSU. (stats from cfbstats.com) This graph also shows that our offense does its job nearly every time and it struggles only when playing an elite level defense. This shows just how much our offense carried this team. Whenever our offense struggled our defense was not able to pick up the slack. Hopefully Al Groh will be able to fix this and provide a safety net for our offense.

Please post any thoughts or comments. My next post will be how our offense preformed in comparison to the rest of the ACC.