Final version of my "final 10 race performance" spreadsheet to rank the statistical results of champions and strong runs by non Champions.

[jaqua19]

Just wanted to share this with you guys!

I can't quite remember the exact formula I used. It included converting and calculating the points total over the final 10 races for each driver. I used the 2011-2013 points format.

The formula was something like (winsx2)+(top5s)+(top10s)/average finish. Then, that number was added to the total points scored, and divided by 10. It's basically a 'total points scored' statistic that weighs and considers top 10s, top 5s, wins, and average finish.

This is what the results were.

I'd love to hear some of your comments, and thoughts about these findings.

Both champions and non Champions are included in this list.

Sent from my SM-G975U using Tapatalk

 

[LewTheShoe]

This is the third thread to post this same mathematical adjustment in recent weeks, and I still don't understand it. I don't get the point of the adjustments... adding, multiplying, dividing in a way that renders the whole thing immaterial anyway. I say that because in every instance, the rankings are 100% explained by the actual points scored without any adjustments. The total impact of all the adjusting is... zero impact.

What is the insight that you are trying to reveal? Perhaps I'm too dense to see it. I notice the highest score of all is by a driver (Truex) who you described recently as a competent "system driver" whatever that means. It sounds like a pejorative term, as people use the phrase "system quarterback" to denigrate a football player who scores a lot of touchdowns and wins a lot of games and they don't have anything negative to use so they call him a "system quarterback."

I'm a data-loving guy, as you might know, but data isn't useful or interesting until it is converted into information.

 

[jaqua19]

This is the third thread to post this same mathematical adjustment in recent weeks, and I still don't understand it. I don't get the point of the adjustments... adding, multiplying, dividing in a way that renders the whole thing immaterial anyway. I say that because in every instance, the rankings are 100% explained by the actual points scored without any adjustments. The total impact of all the adjusting is... zero impact.

What is the insight that you are trying to reveal? Perhaps I'm too dense to see it. I notice the highest score of all is by a driver (Truex) who you described recently as a competent "system driver" whatever that means. It sounds like a pejorative term, as people use the phrase "system quarterback" to denigrate a football player who scores a lot of touchdowns and wins a lot of games and they don't have anything negative to use so they call him a "system quarterback."

I'm a data-loving guy, as you might know, but data isn't useful or interesting until it is converted into information.

I don't think I posted about this since April. Don't understand the point of your second paragraph, other than to dismiss what I've done here. That has nothing to do with any of this.

This isn't a real statistic. It's not advanced analytics. It was a fun project.


Now, obviously the total points scored over the final 10 races paint the whole picture.

So there is no discoverable insight. All I am doing is using the "box score stats", doubling the weight of winning, and adding it to the total points scored. Then dividing that total by 10, as there were 10 races.

So to answer your question about information..I calculated the total amount of points scored for each driver on that list over the last 10 races using the same points format. it's basically a converted points per race that weighs in the box score stats.

This list is ranking the best statistical performances, in order, of the playoff/chase era over the final 10 races. It does so by creating a converted points per race average that considers the total number of wins, top 5s and top 10s.


It's not serious. It's something I made for fun so I can rank the top chase performances of all time statistically.

If you want to call it redundant, go for it, it isn't to me and I got what I was hoping to find out of it.

Sent from my SM-G975U using Tapatalk

 

[jaqua19]

This is the third thread to post this same mathematical adjustment in recent weeks, and I still don't understand it. I don't get the point of the adjustments... adding, multiplying, dividing in a way that renders the whole thing immaterial anyway. I say that because in every instance, the rankings are 100% explained by the actual points scored without any adjustments. The total impact of all the adjusting is... zero impact.

What is the insight that you are trying to reveal? Perhaps I'm too dense to see it. I notice the highest score of all is by a driver (Truex) who you described recently as a competent "system driver" whatever that means. It sounds like a pejorative term, as people use the phrase "system quarterback" to denigrate a football player who scores a lot of touchdowns and wins a lot of games and they don't have anything negative to use so they call him a "system quarterback."

I'm a data-loving guy, as you might know, but data isn't useful or interesting until it is converted into information.

And if I can explain it even more simply, after really thinking about my thought process and curiosity, basically what I did is create a points format based off of the 2011 scoring system that scores an additional point for a top 10, another additional point for a top 5, and 2 additional points for a win. The drivers are scored based on finishing positions, and are given additional bonus points based on that.

I know that points are dependent on wins, t5s and t10s, obviously. But because wins, top 5s and top 10s are historically recorded for drivers, I wanted to find a way to factor them into a measurable performance of a driver over the final 10 races BEYOND their role in total points scored. Hence the apparent indirect creation of a points system that is more strongly influenced by posting wins, top 5s, and top 10s.

That number you see is a points per race average, that in my opinion, ranks the best Chase performances ever. A points per race average, enhanced by win total, top 5 total and top 10 total.

Hope this makes sense to you.

Sent from my SM-G975U using Tapatalk

 

[Team Penske]

And if I can explain it even more simply, after really thinking about my thought process and curiosity, basically what I did is create a points format based off of the 2011 scoring system that scores an additional point for a top 10, another additional point for a top 5, and 2 additional points for a win. The drivers are scored based on finishing positions, and are given additional bonus points based on that.

I know that points are dependent on wins, t5s and t10s, obviously. But because wins, top 5s and top 10s are historically recorded for drivers, I wanted to find a way to factor them into a measurable performance of a driver over the final 10 races BEYOND their role in total points scored. Hence the apparent indirect creation of a points system that is more strongly influenced by posting wins, top 5s, and top 10s.

That number you see is a points per race average, that in my opinion, ranks the best Chase performances ever. A points per race average, enhanced by win total, top 5 total and top 10 total.

Hope this makes sense to you.

Sent from my SM-G975U using Tapatalk

What gets me is a 3rd place is counted as a top 5 BUT also as a top ten.
Win should be 1st
2-5 should be top 5
6-10 should be top 10

Nothing in another series should be lumped together.
You don't hear about any driver other than KDB where they
count every win regardless of the series.

 

[pjmolo]

What gets me is a 3rd place is counted as a top 5 BUT also as a top ten.
Win should be 1st
2-5 should be top 5
6-10 should be top 10

To the best of my knowledge, top 5's and top 10's have always been tracked that way.

Last year's final standings from espn.com:

https://www.espn.com/racing/standings/_/year/2020

 

[jaqua19]

To the best of my knowledge, top 5's and top 10's have always been tracked that way.

Last year's final standings from espn.com:

https://www.espn.com/racing/standings/_/year/2020

Yep, it's always been like that. Top 10 is only a top 10 but a top 5 is a top 5 and a top 10.

And course a win is a win, top 5 and top 10. Which is how I factored it in here. A win basically grants the driver an additional 4 points per race, a top 5 grants an additional 2 points per race, and a top 10 grants an additional point per race.


Sent from my SM-G975U using Tapatalk