Some of you may recognize my name from my work on Chapel Bell Curve, the only (we hope) advanced stats podcast about UGA football. I’d like to ask at outset that you - and I can’t emphasize this enough - not let my (deserved) reputation as a (bad) podcaster color (ruin) what you think of me as an author. I can ruin each of my reputations individually, I-thank-you-very-much. Anyway, for those of you who have controlled their “exit this tab immediately” reflex, I’m very proud to introduce the first installment of What Advanced Stats Tell Us About UGA Football.
Going forward, I’m going to bring some of my thoughts on advanced stats to Dawg Sports each week. Working with the incredible team at #sbnanalytics, I’ll be able to bring you both advanced stats box scores for each game, and updated season statistical summaries for each week. My hope is that this can a) give access to good numbers for those of us who are already statistically inclined and b) provide an introduction to some of the statistical concepts we talk about to the uninitiated dawg.
This week, I’d like to focus our analysis on one of the most widely used advanced statistics: Success Rate. Success Rate’s ubiquity is well earned, as it provides us with a single number that allows for analysis of any given team both offensively and defensively.
Simply put, Success Rate (SR) is a raw measure of a given offense’s ability to gain the needed yards on a given down. If a play gains 50% of the needed yards to gain on first down, 70% on second down, or 100% on third and fourth down, it is counted as successful. If you divide the number of successful plays with the total number of plays run, you arrive at SR, and have a single number that helps you understand the efficacy of an offense. If you do the same calculation, but use the numbers from plays run against a defense, you end up with SR surrendered, which gives you similar information on the defensive side.
But how does this look in action? One of the most common criticism of advanced analytics of any kind is that they can’t replace the eye-test - an amorphous combination of intuition and anecdote that guides a lot of our everyday thinking. In spite of my love of statistics, I actually agree. No stat can replace human intuition, but that’s simply because they were never intended to to begin with. Rather, I view stats as narrative components of the incredibly complex story told in every football game. Sure, you can write a compelling story without - say - dialogue, or symbolism, but why would you want to? I use stats help confirm what I see, and to call BS on what my eyes may misapprehend. Let’s use the advanced stats for UGA’s latest tilt against Murray State as an example of this in action.
If you look at the top-line stats, you can see that the contrast in SR between these two teams is a stark one. The average SR for most CFB games is somewhere around 45%. UGA dominated the game by running successful plays about 68% of the time, all the while holding Murray to a comparatively paltry 30%. You at home, the stodgy “football-is-a-game-of-guts-and-hard-glares-at-respected-opponents” gen-xer, may be saying to yourself, “Well of course we dominated the game, you stupid millennial, anyone could have told you that. Why do we even need this?” Well, I would reply to you, my favorite straw man, it’s helpful because when we can confirm our read of a game, it helps us inform how we think of our team as a whole. So, for instance, if you were to analyze this game through a statistical lens, you wouldn’t be too worried about UGA’s anemic first quarter. Rather, the numbers would tell you that UGA’s total domination of not only SR, but SR broken down by down, quarter, and situation, is representative of an equally comprehensive domination on the field. It’s useful to look at the statistics from a game that is easy to analyze (like this one) because they can help us calibrate our thinking.
SR is also a great indicator of a team’s holistic quality when calculated over a season. UGA also measures up very well on these terms:
Even against an easy schedule, UGA’s season SR of 62% is superlative. Additionally, a Rush SR (Success Rate only on rush plays) of 65% tells us that the UGA run game is as formidable as we think it is. I said this on the Vanderbilt Review episode of CBC, but I think it’s important for us to remember that a win against Vanderbilt is still a win against an SEC opponent. Similarly, a 62% success rate over two games, regardless of the quality of the opponent, is a good sign for dawg fans everywhere. I believe in the axiom that great teams dismantle bad teams, and to this point, while we can’t confirm that 2019 UGA is a great team, they haven’t done anything statistically to disprove that notion, either. And two games in, that’s all you can ask for. (Also, check out the the SR number for UGA on 2nd down. Absolutely lethal)
Catch y’all in the Classic City next week, but until then, Go Dawgs!