Friday, July 26, 2013

Brickyard 400 Preview Video and Discussion of New Gen-6 Car




Brickyard 400 Favorites
Jimmie Johnson (led 73 laps and finished 2nd)
Tony Stewart (finished 4th)
Jeff Gordon (led 11 laps and finished 7th)

Threats for First Brickyard Win
Kyle Busch (finished 10th)
Brad Keselowski (led 5 laps and finished 21st)
Greg Biffle (finished 24th)


Thursday, July 25, 2013

Is another track record in store for this weekend's Brickyard 400?

The new Gen-6 car is fast. But how fast? 2013 is the first season with the new car, and we keep hearing how speeds have picked up, breaking track records several times this season.

Today we investigate how much faster this new car is, considering if it does better on certain tracks than others. We'll use that data to predict what the pole-winning speed could be at this weekend's Brickyard 400 in Indianapolis.

The chart below shows the increase in speeds so far in 2013 compared to the same races last year:


The huge jump at Kansas is due to a repaved track surface. This gain would have happened whether or not we had a new car.  Excluding Kansas, here is what we see:


The average jump in pole speed at these tracks is 1.7mph, or 1.1% above the speeds from last year. Twelve tracks saw faster speeds, while three tracks saw drops. The minimum positive gain is 1.0 mph. 

The tracks with reduced pole speeds were Sonoma, Dover, and Michigan.  These tracks are all so different (road course, intermediate, and superspeedway). There is no trend here. The tracks with the biggest gains were Fort Worth, Bristol, Charlotte, and Loudon.  Again, these are very different tracks. The data suggests that improvement in track speed is happening because of the car, not because of any particular type of track.

Let's keep going with our analysis. Below are the improvements in pole speeds from 2011 to 2012, both of these years using the previous Gen-5 car.


Again, the biggest gains are due to track repavings.  Removing the outliers (above 3 mph), here is what remains:


This chart shows that twelve tracks saw a pole speed increase, while almost the same number, eleven, saw a decrease.  The net average change was 0.1 mph, or practically zero. This indicates that the new car is responsible for a meaningful difference in speed.  A normal year would show almost no change in speed, but the new car caused an increase of 1.7 mph. We know this trend can't continue forever.  20 years of 1.7 mph increases would be a 34 mph jump. NASCAR would never allow the cars to get that fast. (Or would they?)

As we have seen above, consider that the biggest effect to speed increases have come from track repavings.  On average, a fresh track surface can increase speeds by 10-15 mph. This factor is ten times bigger than the effect of a new car.

Turning to this weekend's Brickyard 400, what can we predict about the pole speed?

Here is the full history of pole speeds at this track:


Remember that every year there are minor changes in rules, equipment, and tires, along with gradual wear and tear in track conditions. Notice the rapid growth in the first ten years (13.9 mph gain) before dropping 4.5 mph in 4 years, and then settling in the 182/183 range.  

The pole speeds in the last two years were:
182.99 mph in 2011
182.76 mph in 2012

That's a drop of 0.23 mph.

By considering the tracks that had similar drops in 2012 and looking at how those tracks have fared in 2013, we can expect an improvement in speed this year between 1.07 and 2.16 mph.

That would allow us to predict a pole speed this weekend in the range of 183.84 to 184.92 mph.

The track record is above 186 mph (set in 2004), so we should *NOT* expect to see a track record this weekend.

Let's see how accurate this prediction turns out to be. If I am wrong, you get your money back.


Friday, July 12, 2013

Video Previewing New Hampshire and Discussing Latest Article



(Updated post-race with results)

New Hampshire Favorites
Clint Bowyer (finished 13th)
Denny Hamlin (finished 21st)
Tony Stewart (led 84 laps but ran out of gas late to finish 26th)

Potential First-Time New Hampshire Winners
Carl Edwards (finished 8th)
Matt Kenseth (led 33 laps and finished 9th)
Dale Earnhardt, Jr. (finished 14th)



Thursday, July 11, 2013

The Difficulties of Passing at New Hampshire: Why Crew Chiefs Matter More

This weekend, NASCAR visits Loudon for the first of two races there. New Hampshire's track is 1 mile in length, and features very flat banking.

Unlike what we saw at Daytona last week, where there were many passes and lead changes, races at New Hampshire behave more consistently. It's much more difficult to get around other cars.

This observation leads to our key takeaway: passing will be at a premium this weekend. It will be important for drivers to start up front, make their moves early, and stay up front for the duration of the race.

Unlike other tracks, at Loudon drivers can't wait until the end of the race to shake things up. The image below shows the average number of spots gained by drivers in the last 10% of races at New Hampshire (2005-2012). Notice that for most drivers, the average gain is 1 spot or less. It's very rare for a driver to move more than 2 spots on average late in a race at New Hampshire. This means drivers need to get themselves into position early. Don't bet on a big surge late in the race.


Next up, consider the difference between average running position and average finish at New Hampshire races. Look how closely connected these variables are. A driver’s position in the early and middle part of the race is typically a good predictor of the finish. Because of this, don't expect to drivers to hang back early in the race, waiting to make a move at the end. It will be too late for that. The race is short (barely over 300 miles) and the flat banking makes passing difficult, so drivers need to get up front as early as possible.


We’ve already talked about how important the starting position is for maintaining a lead and finishing strong. Notice in the chart below how closely these two stats are related: as with the previous plot, there’s virtually a linear relationship. How can a driver maximize starting position? The answer of course is qualifying: expect that to play a big role in this weekend’s race.


Building on our common theme: we asserted that passing cars on the track is very difficult at New Hampshire; the data illustrates this claim very nicely. Below are the net green flag passes per race by driver. Notice that it's rare for somebody to gain double-digit positions on the track. Generally speaking, drivers will only pass a net total of 10 or fewer cars during the race.


With everything above discussing the importance of starting up front and staying up front, remember that there are still some big variables that can result in surprising winners:

  • 2008: Kurt Busch stayed out on the track (instead of pitting with the leaders) during a late caution, before the race ended early due to rain.
  • 2009: Joey Logano collected his first career Cup win after earning a free pass from being a lap down, lucking out with the right amount of fuel because of a rain-shortened race.
  • 2010: Tony Stewart ran out of gas on the last lap, giving the win to Clint Bowyer.

Statistically, if you want to see an exciting race at New Hampshire, it will come not from passing but from unpredictable weather, pit strategies, and fuel gauges. And so the human element becomes important for mitigating these factors: look to the crew chiefs to have a big impact.


Sunday, July 7, 2013

Reviewing the Daytona Picks

On Thursday, I made my picks for yesterday's 400-miler at Daytona. How well did I do?

Daytona Favorites
Tony Stewart (finished 2nd)
Dale Earnhardt, Jr. (finished 8th)
Jeff Gordon (late crash: finished 34th)

Threats for First Daytona Win
Clint Bowyer (finished 4th)
Kurt Busch (finished 6th)
Carl Edwards (crash on last lap: finished 29th)

Overall I picked 4 drivers in the top 8, plus two who crashed out. Not bad.

Thursday, July 4, 2013

Danica, Denny, Dale Jr, and Daytona: A Mid-year Update




(This is a cross-post with BSports StatsInsights)

With Daytona coming up this weekend, we are now starting to visit tracks for the second time this season. This is a natural place to review some mid-year stats, and preview what we might see between now and September, when the Chase officially starts.

First, let's review the drivers who have shown the greatest improvement from average start to average finish this season:

2013 Drivers Who Improve Most from Start to Finish
6.2       Danica Patrick
6.2       J.J. Yeley
4.4       Dave Blaney
3.9       David Reutimann
2.9       David Stremme

Notice that Danica Patrick leads this list. She has done a good job this year of holding her own during the races. Her next goal should be to improve her qualifying efforts, as that will pay big dividends for the rest of this year and later in her career.

One thing to notice is that this list generally featuresdrivers who don't start or finish particularly well. These are generally back-of-the-pack teams who are able to move up positions during races by avoiding crashes, and letting the attrition inherent to NASCAR benefit them.

Now to the flip side: let’s look at all the drivers who lose the most spots in-race. These are the drivers with the biggest drop from average start to average finish:

2013 Drivers Who Decline Most from Start to Finish
-10.0       Denny Hamlin
  -9.3      Kyle Busch
  -7.3      Kasey Kahne
  -5.8      Mark Martin
  -5.7      Matt Kenseth

Some of these drivers on this list aren’t surprising, giventheir volatile style of racing. We also see that the entire Joe Gibbs Racing team is on here (Hamlin, Busch, Kenseth). Even though Denny Hamlin has missed a few races, he may end up missing the Chase, an even bigger prize, because he just drops too many spots during a race. Many of those performances are due to factors beyond his control, but it all adds up to hurt his overall performance.

In terms of Hamlin's chances of getting into the Chase: based on the current projections, it looks tough for him to make the cut. He'llneed to crack the top 20 in points, and have 2 wins. Right now, the main challenge will be getting into the top 20 ranking. A 7.5 average finish is championship-level performance, and so far Hamlin has not been able to demonstrate that level of consistency or quality this season - but he may be saving the best for last; we’ll just have to wait and see.

Denny Hamlin's Chase Chances are looking slim:
26% chance of winning 2 races between now and Chase
Needs an average finish of 7.5 to be ranked in top 20.

In terms of who will make the Chase, there are three names outside the top 10 that I think will make a strong run, and three drivers currently in the top 10 that will struggle to hold their positions. These selections are all based on my multi-race projection system comparing this year's race-by-race performance to previous years.

Drivers Who May Struggle to Make Chase
Dale Earnhardt, Jr.
Greg Biffle
Joey Logano

Drivers Projected to Make the Chase
Kasey Kahne
Jeff Gordon
Brad Keselowski

Finally, with Daytona coming up this Saturday night, let’s try to make some predictions on who could win. As always, these names are derived by my single-race projection system to find drivers who could win the race.

Daytona Favorites
Tony Stewart
Jeff Gordon
Dale Earnhardt, Jr.
  
Threats for First Daytona Win
Clint Bowyer
Carl Edwards
Kurt Busch