If you don't care about how I calculate these categories or what the categories actually measure, you can skip down to the "How to Read the Chart" section.
Anyway, let's get into it.
The Math:
The math really isn't very complicated. In order to get all of these numbers to work on the same chart, I have to get them on the same scale. The way I get them on the same scale, is to rate how well the team is above or below average in each category. How I do that depends on whether or not the objective of that statistic, is for the team to acquire more or less of a certain number.
An example of when a team wants to get more than the average team in a certain category, would be a stat like offensive touchdowns. An offense would like to score more touchdowns than the average team. The simple way to find out how above or below average a team is in that category, is to divide the team's statistics in that category, by the conference average in the same category.
However, if a team is below average in a category, their number will come out negative. That won't work on a radar chart. So, in order to keep the numbers positive, I multiply the numbers by 100. That puts the ratings on a scale where 100 is average. Anything below 100 is below average. Anything above 100 is above average.
When the goal of a team is to have less than the average team in a certain statistic (like interceptions thrown), then the formula gets a little more complicated.
If I used the same formula from above, the teams with the best numbers would come out lower than 100. In order to flip that, and make those that are above average have numbers above 100, I have to do a little more math.
First, I take the team's statistic in a category divided by the Big 10 average, like I did before, but then I subtract it from 1. What that does, is gives me the percentage of how much above or below average a team was, alone by itself. From there, I multiply it by 100 to make the number bigger. Finally, I add 100, to get it on the scale I described previously, where 100 is average.
Equations:
- When the Goal is to Get More of a Statistic- 100*(Team Statistic/Big 10 Average in Statistic)
- When the Goal is to Get Less of a Statisitc- (100+(100*(1-(Team Statistic/Big 10 Avg. in Stat))))
This would be last year's version of Iowa's offense. I've used five categories to rate the offense and defense on. Below, I will break them down.
- Passing- I use quarterback rating. While it's not perfect (there are legitimate criticisms of it), it does try to take into account all the categories that are important: yards, touchdowns, interceptions, and the amount of attempts taken to put up those numbers.
- Rushing- For rushing, there wasn't one number that I felt would rate a team's running backs. So, what I did was find how much above or below average a team was in yards per carry and touchdowns per carry. Then, I took the average score of the two numbers. Hopefully, this captures how many yards a team gained on the ground and how many touchdowns they scored, all while taking into account the amount of rushes it took to rack up those numbers.
- Scoring- In order to see how dynamic a team's offense was, I used offensive points per play. I only take into account points scored by the offense (field goals do count). Special teams or defensive returns for touchdowns are removed from this statistic, in order to measure the actual scoring ability of a team's offense. I also do the same for a team's defense, so that they don't get punished for something like the special teams giving up a kick off return for a touchdown. Like the categories above, this rating also takes into account how many plays it took to put the points up on the board.
- Turnovers- Turnovers is pretty simple. I used turnovers per play. This should accurately reflect how good a team is at holding onto the ball or, in the defense's case, taking it away.
- Blocking- I wanted to measure more than the skill position players on a team. So, in order to attempt to measure how good an offensive line is, I used tackles for loss per play. Ideally, the better offensive lines will have a lower percentage of plays end in their running back being stuffed in the backfield or their quarterback being sacked.
We will use the 2010 defense as an example of how to read the chart. As always, the first thing to remember is 100 is average. Anything below 100 is below average performance, and anything above 100 is above average performance. What makes these ratings even more useful, is that how much the number is above or below 100, is also the actual percentage that team was above or below average in that category.
So, for the above chart, Iowa's defense was above average in every category last year, except for in getting off blocks (thinking about renaming that "Getting in the Backfield") or tackles for loss per play. In that category, a rating of 77, means that they were 23% below the Big 10 average. 100-77= 23. Pretty easy, and pretty handy. Looking at the other categories, Iowa was 5% above average in turnovers forced, 54% above average in scoring defense, 29% above average in rushing defense, and 13% above average in passing defense.
Conclusion:
If you read this post, I hope it was helpful. The last thing I want to do is write posts using numbers that nobody understands. Also, hopefully, this gives people insight into how I create these ratings, just in case you think I'm pulling these numbers out of thin air. I have actually put thought into these, and am open for suggestions as to how to make them better.
Finally, if you have any more questions, let me know, I'll do my best to answer them all. After all, I'm a Hawkeye fan like the rest of you. My hope, is that giving you a look behind the numbers adds to your football, and overall Iowa Hawkeye, experience. Go Hawks!
No comments:
Post a Comment