If you don't already frequent Football Study Hall, you should start. Bill Connelly is a statistician of the highest order and does some amazing work. He's got so many CFB-centered metrics on how stats affect games, it's just awesome.
My hope is do just look at some top level stuff and draw some conclusions. We're going to be looking at a handful of stats and how those played out throughout the season, specifically offensive and defensive total yards, pass yards, rush yards, points and turnovers. This will not include things like points on special teams (i.e. punt/kick returns) or momentum swings/big plays. This is just a 30,000 ft. look.
Let's do this.
|Total Yard Margin||-29||375||-65||312||245||355||136||-66||57||92||-217||-67||72|
|Pass Yard Margin||25||207||98||301||184||247||208||139||214||278||-15||120||278|
|Rush Yard Margin||-54||168||-164||11||61||108||-72||-205||-157||-186||-202||-187||-206|
Edit: I omitted the offensive and defensive stats to make the chart easier to read and less daunting. I also bolded any figure that was negative, to help add emphasis.
Ok, so mentally, draw a thick, red line between the West Virginia and Oklahoma games, because that's where things start to change.
And I'm looking at the Margin section, there are a LOT of negatives in those losses. I think the Margin section is where we're going to be spending a lot of our time as it highlights how well we did some things and how poorly we did some other things.
Tech had 5 games where they were outgained from a total offensive yards standpoint. Two of those games, Tech won. The only BIG negative margin here was Baylor. After looking at these numbers for a while, I've come to the conclusion that you can't really tell how the game went by just looking at the Total Off Yards. You just can't. You've got some games where Tech had a huge margin and blew the opponent out (SFA +375) and big margins and barely squeaked by (ISU +355). Now, that's offensively starting to point us in the right direction to determine what stats are most meaningful for a Tech victory.
Pretty similar with Total Yards. Tech only had one game where they were outgained in the passing game, and it was agaisnt Baylor (-15). You can glean some things from this, but really, about the only thing the data shows is that it's BAD if Tech is outgained through the air (although there's only one data point, so not really statistically significant).
You were probably expecting this, but this is where the stats start to impact the games. In the 8 victories, Tech was outgained on the ground 4 times, and had an overall rush margin/win at -18.5 yards. So even in the wins, Tech was averaging almost 20 fewer yards on the ground. Two outliers here would be TCU (-164) and ASU (-206!!!).
And then in the 5 losses, Tech was outgained in every single one of those, and it wasn't even close. In those losses, Tech averaged giving up 207 more yards on the ground than they gained on offense. I haven't been able to find a trend in rush yards margin and how that relates to wins/losses, meaning if Tech is outgained by X yards, Tech will lose. I do think I have found a trend in number of statistical categories and which team holds the more positive margins.
In the 5 statistical categories, Tech can have a positive margin in anywhere from 0 to all 5. Let's break down the season into categories in which Tech held the statistical advantage by number of categories (still with me?).
There was only one game where Tech held all 5 statistical advantages; Kansas. Tech was +245 in total yards, +184 in pass yards, +61 in rush yards, +38 in points and +3 in turnovers. And you can see, it was Tech's most complete game of the season (at least with our numbers here). Tech won 54-16 and once Tech started scoring points, it was over.
We're looking at Stephen F Austin, Texas St., Iowa St. and Arizona St. Tech won all of these games handily, with the exception of Iowa St. (which, I think, was because we lost big time in TO margin at -3 and gave up a special teams score). Only Arizona St. in this group outgained Tech on the ground. The other statistical shortcoming in those games was the turnover margin (I counted an even margin as a loss)
Tech only had one game where they held 3 statistical advantages, and it was West Virginia. Tech was outgained on the ground (-72) and was -1 in the turnover margin.
This is where it gets dicey. Up until now, Tech has won all the games listed above. The group of games in this 2 category were a mixed bag, but I would still think the losses in this group could be winnable games. Obviously, if Tech were to win these games, they would hold a statistical advantage in an additional category, but what I was meaning if the margins were closer to even, Tech could win. I think if Tech expects to win, they cannot be negative in rush yards margin AND turnover margin. You can't just win very many ball games losing those two categories.
The games where Tech held 2 statistical advantages were SMU (W 41-23), TCU (W 20-10), Oklahoma St. (L 34-52), Kansas St. (L 26-49), and Texas (L 16-41). In all of these games, Tech's offense struggled to move the ball consistently for long stretches in the game with maybe SMU being the lone exception. In all of these games, Tech was outgained on the ground and with the exception of Texas, was negative in the TO margin. This combination of rush yards and TO margin seems to be key. The only times Tech won while not holding those advantages were SMU (TO margin was 0) and TCU. I think we can see how if TCU was a little more consistent offensively, that game could have gone the other way (officiating notwithstanding).
Only one of these games and it was a little surprising as it was Tech's closest loss of the season, at Oklahoma (L 30-38). Tech only held the advantage on passing yards at +139 (which isn't all that big, considering OU held the rushing advantage at +205). Tech did everything they could to stay in this game with onside kicks and trick plays. I think that's where this game was made closer than the stats would suggest.
Only one of these games, too, and it was an ugly affair; Baylor (L 63-34). Obviously, Tech can't win any games holding 0 statistical advantages.
|Offense||Season Totals||Game Averages|
If you look at the Margins, rush yards and TO were negative on the season. I think if Tech is going to turn anything around this year, those numbers need to AT LEAST be closer to even (0), if not positive. If they continue to average in the negative, I think we will see another season like this one.
VTM statisticians, what do you think? What do you see in these numbers?