Defining NASCAR's Lead Changes
By Sean Wrona
These trends are connected. A large number of cautions makes it harder for cars to fall a lap down and provides more opportunities for more cars to attempt to win races via strategy alone, thereby artificially inflating the number of lead changes when teams use alternative strategies. 2010 and 2011 were the first two full seasons after the wave-around rule was introduced so perhaps it shouldn’t be surprising that NASCAR saw a glut of lead changes in those seasons despite a reduction of both cautions and lead-lap finishers.
NASCAR’s basic statistics ignore that there are, in effect, four different kinds of lead changes that can occur in a race:
Using these four definitions, I reviewed every race of the 2012 NASCAR Sprint Cup Series season and classified each lead as a natural, strategic, resumptive, or incidental lead. The initial laps of the race are always considered to be natural unless the pole-sitter does not start on the front row, in which case I would count the lead change on lap 1 as incidental. If a driver who initially led the race strategically or incidentally later resumed the lead after an exchange of pit stops, I continued to count the future laps led as strategic or incidental. Out of 681 total lead changes in 2012, 265 were natural, 344 were strategic, 52 were resumptive, and 20 were incidental. Although all but two races had 10 or more official lead changes in 2012, only six had 10 or more natural lead changes (and 11 of them had three or fewer). Excluding the three plate races (where this is commonplace) these races were the Coca-Cola 600, the fall Richmond race, and the fall Martinsville race. The last two are very interesting given that most people think short track races don’t have many lead changes. This may help explain why these tracks are fan favorites while intermediate superspeedways are frequently not.
There is much more of interest in these data than simply determining the true competitiveness of races. Many fans frequently argue that certain drivers are too reliant on their crew, their crew chief’s strategy, or pure dumb luck. The main point of classifying lead changes into four categories is to provide a metric for determining whether the driver or the crew played a bigger role in the team’s success as defined by leading races.
Although Jimmie Johnson led a series-high 26 races and 1,745 laps in 2012, Denny Hamlin led the most races (17) naturally, two more than Kyle Busch and Johnson, who led only 15 races in this manner. Astoundingly, Keselowski led only seven races naturally, but I would hardly use that as an indictment of him as a driver. Keselowski is a more strategic-minded driver than most of his peers and in an era when there are generally more strategic than natural lead changes, he and Paul Wolfe have correctly deduced that strategy is what wins races and titles, while Busch’s desire to lead every lap leads to more inconsistent results in that regard, even if it’s more fun to watch.
By contrast, Johnson led 21 races strategically, the most on the circuit, with Matt Kenseth and Keselowski leading 18 and Jeff Gordon and Hamlin leading 17. Busch and Tony Stewart were noteworthy in being two of the few drivers to lead more races naturally than strategically (15 to 13 in Busch’s case, and 11 to three in Stewart’s case), perhaps indicating that their crew and/or team may be letting them down. Clint Bowyer, Johnson, Kenseth, and Stewart led the way with two races led incidentally.
This analysis can be made even more nuanced than simply comparing races led, however. I invented the rather basic average percent led statistic on my website to measure dominance equalized over all races. A driver would have an average percent led of exactly 50 percent for leading half of any race regardless of length. The pure laps led statistic is too biased towards drivers who dominate races with greater numbers of laps such as short track races; for instance, Mark Martin actually led more laps than Gordon in 1998, but Gordon’s average percent led of 21.39 percent was considerably higher than Martin’s 15.71 percent, which serves as a better indication of how that season actually went. Applying this idea to the current framework, I can calculate average percent led for each driver for each type of lead change. Drivers with a higher natural percent led are more often outracing the competition while teams whose drivers have a higher strategic average percent led are more often outthinking. Using this metric, Busch was the most dominant natural leader, even though Johnson was more dominant overall.
In the table below, which lists each 2012 leader’s average percent led broken down by category, I have included resumptive percent led as part of natural percent led since the driver who resumed the lead after the end of a pit cycle presumably would have led the whole time discounting the pit stops.
Percentage of Laps Led, Specified by Lead Change Types