Note: This article was originally published in 2015. Since then the race record was broken again, in 2021.

When it comes to fast races at Indianapolis, lately they’ve come in pairs. Case in point, 1990-1991, 2011-2012, and 2013-2014 (and perhaps even 2018-2019 and 2021-2022). Of those recent years, four of the fastest Indy 500s to date have the unique distinction of being held in consecutive pairs, and have curious similarities.

As of 2023, the race record for the Indianapolis 500 was set in 2021 by Hélio Castroneves (average speed 190.690 mph). Prior to that, the record was set in 2013 by Tony Kanaan (187.433 mph). While we are accustomed to the commonplace reporting of average speed, it should be noted that the race is actually scored by elapsed time. The average speed is merely calculated using the standard mathematical formula R=D÷T.

The race winner has completed the 500 miles under three hours officially on eleven occasions:

Year Winner Time Average Speed
Laps Under
2021 Hélio Castroneves 2:37:19.3846 190.690 2 18
2013 Tony Kanaan 2:40:03.4181 187.433 4 21
2014 Ryan Hunter-Reay 2:40:48.2305 186.563 5 21
1990 Arie Luyendyk 2:41:18.414 185.981 4 28
1991 Rick Mears 2:50:00.785 176.460 7 35
2019 Simon Pagenaud 2:50:39.2797 175.794 4 29
2022 Marcus Ericsson 2:51:00.6432 175.428 6 31
1986 Bobby Rahal 2:55:43.470 170.722 6 29
2011 Dan Wheldon 2:56:11.7267 170.265 7 40
2023 Josef Newgarden 2:58:21.9611 168.193 5 27
2012 Dario Franchitti 2:58:51.2532 167.734 8 39
2000 Juan Pablo Montoya 2:58:59.431 167.607 7 39
1989 Emerson Fittipaldi 2:59:01.040 167.581 7 43
2018 Will Power 2:59:42.6365 166.935 7 41

Unofficially, the race has been completed in under three hours one other time (but we will get to that later). In addition, a growing number of drivers (more than two dozen) have completed the full 500 miles in less than three hours in non-winning performances. Two race winners (Helio Castroneves in 2002 & and Alexander Rossi in 2016) missed breaking the three-hour mark by mere seconds. Rossi, in particular was the closest, finishing the race in 3 hours and 2.0872 seconds. Famously he coasted across the finish line out of fuel, which surely kept him from breaking the mark.


Several factors contribute to the race’s overall average speed, of course owing not little to the speed of the race cars on the track. That is why a contemporary Indy car 500-mile race (with lap speeds in the 220 mph range), will always beat out a stock car 500-miler (lap speeds in the high 190s mph and low 200 mph range). In the early/mid-1980s, NASCAR’s 500 mile records were faster than those in Indy cars. As Indy car speeds caught up and passed stock car speeds (and after restrictor plates were implemented to slow down the stock cars), Indy car superspeedway races were now superior in terms of closed-course competition speed records.

Racing speeds in Indy cars peaked in the mid/late-1990s, then dropped and basically plateaued in the 2000s due to all sorts of changes in the rules and the series (we won’t get into that). The constant annual uptick in speeds was now mostly a thing of the past, and can no longer be counted on to produce record speeds merely of its own merit. Perhaps the single factor that affects the average speed the most is the number of cautions and likewise the number of laps under caution. The correlation between caution laps and average speed is well reflected in the above chart.

Caution periods

As of the most recent race (2021), race lap speeds were in the 215-225 mph range. That translates roughly to 40.0-41.8 seconds per lap. Laps under caution were more in the 2 minute range (75 mph). For each lap run under yellow, it adds approximately 78-79 seconds to the elapsed time.

The current rules for cautions came into play starting with the 1979 race, and have been further amended in 1992, 2000, and a couple times in the 2010s. At the onset of caution, the cars must slow down and hold position. The field is “frozen”, and the pace car comes out to pick up the leader. The rest of the field bunches up, a method known commonly as the “pack-up” rule. The caution lap speed is about 75-80 mph.

Since cautions are attributable to an incident on the track (or debris), a race with an abundance of crashes (particularly crashes that involve wall contact) is going to be slower.

When cautions did not matter as much

The yellow caution light system first was installed at the Speedway in 1935. Through 1971, when there was an incident on the track, the yellow lights would come on and the starter would wave the yellow flag on the mainstretch. Drivers were instructed to slow down to a prescribed speed, and hold their positions. The pace car did not come out and pack up the field. As soon as conditions were deemed safe, the green light would come back on, and the racing would resume – sometimes unexpectedly and without warning. Track management wanted as many laps under green as possible. They wanted there to be as much “racing” as possible, and did not want drivers to lose any lead they built up every time a caution came out. Generally speaking, the caution would only come on for significant incidents (crashes and spins), and would only stay on for as long as it took to clean up the incident from the track. Stalled cars typically did not warrant a yellow, and instead drivers were instructed to park their disabled cars in the grass infield, and allow the race to continue without needing a yellow. The stalled cars would typically be abandoned, and remain parked in the grass until after the race was over. In some cases, if the car came to a rest near an opening in the infield wall/fence, it was usually pushed behind the wall/fence, but such was done on a case-by-case scenario. There were cautions for debris, but only long enough for a track worker to retrieve it. It was not unusual for a yellow light period to last less than one minute in duration for something as simple as picking up debris.

One will see that during, for instance, the decades of the 1950s and 1960s, the race record was broken quite regularly, despite the box scores listing numerous yellows. In addition, the race average speeds were not far below the typical race laps. Short yellows (and the relatively high speeds driven under yellow) played a role. By the late 1960s and early 1970s, the chief steward would announce a speed the drivers were supposed to maintain during yellow lights (for instance 110 mph under yellow versus 160 mph green race laps). The drivers adhered to the rules in principle, and did not pass other cars, but most observers began to note that many drivers were narrowing the gaps between cars, and slowing down only in the vicinity of the incident. As soon as cars were past the crash scene, some would speed up back close to race speed, then slow down again as they approached the scene again. This dance would repeat until the green ultimately came back out.

Since the yellows were short, and the cars did not slow down as much (relatively speaking) under yellow, cautions were not as much a factor when it came to average speed prior to 1972. This aspect came to an end in 1972 when USAC implemented the PACER Light System (a whole topic of its own for another day), and dropped the speed allowed under yellow.

Pit stops

Whenever a car comes into the pits and comes to a complete stop, the average speed is going to sink. Through the mid-1960s, the prevailing strategy was to make as few pit stops as possible. Large fuel tanks (and the use of gasoline) allowed cars to make it well beyond the halfway point before refueling, and thus some teams planned on only one pit stop. Over the years there were some cars that attempted and succeeded in completing the 500 miles without any pit stops, but the reduced pace they typically had to maintain precluded them from having a chance to challenge for the victory. There was no pack-up rule under yellow, and pit stops were generally considered undesirable, as they were costly in terms of track position. Pitting under yellow was not looked at as an advantage, and usually only occurred if the yellow light happened to come on when a stop was scheduled. Likewise, tires were only changed when needed.

After the 1964 race, when a minimum number of pit stops was made mandatory, crews still attempted to keep pit stops to a minimum. In the late 1960s and into the very early 1970s, some cars (even winning cars), changed no tires during the entire race. Changing no tires naturally reduced the time spent in the pits, and also allowed the tire manufacturers the ability to boast of their product’s durability and reliability. Unless a driver suffered an unfortunate puncture, spun out and flat-spotted the tires, or otherwise wore the tires out too early, the plan was usually to go without changing them.

When the bolt-on wings came onto the scene starting in 1972 speeds quickly climbed. The days of going the distance without changing tires were over. Fuel tank capacity was reduced from 75 gallons to 40 gallons in 1974, and naturally necessitated more stops for fuel. Tires had to be changed more often due to the increased speeds, however, through the 1970s and into the 1980s, tires (bias-plys) were still not changed every pit stop. It was not unusual for crews to inspect the tires before actually changing then, and at that, the right-side tires were the focus of the most attention. The left-front may be changed only once, or not at all during the race. All this contributed to a minor ‘slowing’ of the race due to more elapsed time in the pits.

In 1987, radial tires arrived at Indy, and strategies began to evolve with respect to tire wear and tire changing. Teams began changing all four tires almost every stop, so much so that in 1988, a fourth tire changer was permitted over the wall (previously only three men were allowed over the wall to change tires). Eventually as teams became considerably more proficient in conducting pit stops, changing four tires started taking less time than the refueling process. In the present time, tires are generally engineered to last about as long as a fuel stint.

Pit road speed limits

I have to credit Arie Luyendyk for pointing this one out. During an interview in the 2000s, he was asked if he thought his 1990 race record (185.981 mph) would ever broken. He theorized that one of the contributing factors keeping the races slower was the implementation of pit road speed limits. This was a considerable factor in the 1990s, but upon further investigation, it seems to have diminished in the 2010s.

Through 1988, the pit road was concrete, and by that time, was becoming quite rough. The cracks and bumps kept the speeds down, although there was no dedicated speed limit in the pit lane, even during cautions. Prior to the 1989 race, the pit lane was paved over in asphalt, making a much smoother and safer ingress and egress. At the same time though, drivers were starting to go much faster through the pits. The 1989, 1990, and 1991 races all saw the cars (especially the leaders) flying down pit road, particularly during green flag pit stops.

In November 1990, at the NASCAR Winston Cup season finale at Atlanta, a fatal crash occurred in the pit area. Ricky Rudd locked up his brakes, and spun into the car of Bill Elliott. Elliott’s car was being serviced, and tire changer Mike Ritch was struck and fatally injured. The series reacted by implementing a bevy of pit rules in 1991, designed to make pit stops safer for everyone involved. Numerous complex rules were tried out and many were scrapped, but one rule that firmly stuck was the implementation of a strict speed limit on pit road, enforced at all times. USAC and CART followed suit in 1992, imposing similar speed limits. CART had a near-miss incident at Long Beach in 1991 when Michael Andretti and Emerson Fittipaldi touched wheels exiting the pits. Andretti’s car became airborne and flipped halfway over, then came down almost on top of Fittipaldi’s car, landing in another team’s pit.

For the 1992 Indianapolis 500, USAC imposed a 100 mph speed limit on pit road during cautions. Penalties for speeding infractions were in the form of warnings or fines, but starting in 1993 became stiffer, drawing a dreaded stop-and-go penalty. Also starting in 1992, USAC started closing the pits at the onset of a caution. Cars were required to bunch up behind the pace car before they were allowed to pit. Usually it took one lap to adequately pack-up the field, then the pits would open. For 1993, the pit road speed limit was enforced at all times, not just under caution.

As years went by, the pit road speed limit at Indianapolis was gradually reduced. From 1992 to 1998 it was 100 mph. In 1999-2001, it was reduced to 80 mph. Since 2002, it has been 60 mph. The elapsed time for the in-lap, the stop, and the out-lap was now longer than it used to be. The agonizingly slow 60 mph trek down Indy’s long pit lane added as many as 10-20 seconds per pit stop. Multiply that by six pit stops, and that could be up to two whole minutes added to the length of the race.

Using the 1990 race and 2015 race as a gauge, however, the pit road speed limits seems to have nay affect anymore on race average speed. A pit stop in 1990 changing four tires and adding fuel (40 gallons) took about 18 seconds. In the 2010s, pit crews have improved their speed and efficiency substantially, resulting in much faster stops. Fuel tank capacities have been reduced to 22 gallons, and a vented fueling hose has replaced the need for a dedicated vent hose man. The refueling process naturally takes less time, and pit stops are closer to the 7-8 second range. The time lost slowing down for the pit road speed limit has almost been made up by the stop itself taking less time. As of 2015, on average a pit stop sequence (timed from exiting off of turn 4, pitting, and re-entering the track in turn 2), takes less than 5 seconds longer than it did without speed limits in 1990.

The 1972 record

Back in the early days, the race record for 500 miles was broken nearly every year. This was mostly owing to the advancements made in speed and equipment each year. It was not unusual to shave a minute, or as many as ten or fifteen minutes, off of the record time from the previous year. This trend continued well into the early 1970s, by which time lap speeds escalated rapidly. But by the 1980s, the rising trend in race average speed had stalled somewhat.

Mark Donohue set a race record of 162.962 mph in 1972. The race was slowed by only five yellows for 14 laps. At the time, there was no “pack-up” rule during cautions. Cars were instructed to slow down to a prescribed speed and hold their positions. As it happened, the PACER light system was in use from 1972-1978 to (theoretically) maintain the spacing of the cars around the course. Caution periods were generally not an official statistic back then, but it was recorded that the yellow light was on for 20 minutes and 36 seconds. In fact, two of the five yellows were less than one minute in duration. Donohue’s 1972 record stood for twelve years, despite the rising speeds. How did Donohue’s record stand so long? The answer lies mostly in three factors. Rain, crashes, and the “pack-up” rule.

Rain shortened the race in 1973 (332.5 miles), 1975 (435 miles), and 1976 (255 miles). A race record average speed would not be recognized in a race that did not go the full distance, although none of those three races came close to challenging the 1972 record. A bevy of crashes and cautions plagued the 1979, 1980, and 1981 races, preventing any of those from breaking the record. In addition, the 1979 race ushered in the “pack-up” rule during cautions. At the onset of a caution period, the pace car would enter the track, pick up the leader, and the rest of the field would “pack-up” behind the pace car. This practice was fairly standard at other races, but was eschewed by Speedway management up to that point. The new procedure naturally lengthened caution periods, contributing to a longer elapsed time and thus, slower overall average speed.

The 1982 race and the 1983 race came close to the 1972 record. Both were longer by less than one minute. Translated to track position, Gordon Johncock (1982) and Tom Sneva (1983) would have been about one lap behind Mark Donohue.

The 1984 record

The inevitable finally occurred, and Mark Donohue’s 1972 race record fell in 1984. Rick Mears won the 1984 race at an average speed of 163.612 mph, about 43.88 seconds faster than Donohue. The 1984 race was slowed by only five yellows for 34 laps. The race went a notable 48 laps before the first caution, and most of the second half was fairly clean. Mears emerged two laps ahead of the field in the closing stages, and he cruised over the final 42 laps unchallenged.

Race laps by the leaders were in the high 190 mph / low 200 mph range, with Gordon Johncock (204.815 mph) setting the fastest lap of the day.

The 1986 record

The 1986 race was a record-breaking day practically from the drop of the green flag. Michael Andretti blasted from the outside of the front row to take the lead on the first lap. His lap of 202.940 mph was the first time the opening lap was completed at over 200 mph. The race saw only six yellows for 29 laps. Bobby Rahal’s average speed of 170.722 mph was a new record, and Rahal was the first driver ever to complete the Indianapolis 500 in under three hours.

The 1986 record is somewhat peculiar compared to other fast 500s, in that the longest green flag stint was only 45 laps. (laps 57-101). The six cautions were evenly spaced, but none of them were very long in duration. Though the overall race record was soon broken in 1990, Rahal’s halfway record (100 laps / 250 miles) of 176.251 mph remarkably stood until 2013.

The 1986 record can largely be attributed to a clean race, and a slight uptick in race lap speeds by the leaders. Only one car (Rich Volger) hit the outside wall, and it was not a heavy wall contact. One of the yellows was for debris, and two were for stalled cars. The other two yellows were for a spin by Johnny Parsons, who subsequently backed the car into the inside wall, and Arie Luyendyk’s infamous spin by the pit entrance on lap 194 which set up Rahal’s winning pass of Kevin Cogan. None of the incidents were serious, and all were cleaned up quickly. Lap speeds were in the high 190 mph range, flirting with the low 200s. Rahal’s second-to-last lap was run at 203 mph, and his final lap was run at 209.152 mph, the fastest single lap of the race.

The 1990 record

There is a reason the 1990 race was known as “The Fastest 500” for so many years. The 1990 race shattered the existing record (1986) by over 15 mph. Arie Luyendyk’s average speed of 185.981 mph set a mark that would take 23 years to break.

The race got off to a clean start, but was slowed in the first half by three yellows. Danny Sullivan crashed on lap 20, Tony Bettenhausen required a tow-in on lap 45, then on lap 63 the yellow came out again for two blown engines and a separate crash involving Pancho Carter. On lap 70, the green came back out, and the rest of the race was green, with exception of a brief, 5-lap caution when John Andretti bushed the wall in turn four and seconds later spun lazily in turn one on lap 141.

Bobby Rahal’s 1986 halfway record (100 laps) survived, but each and every 10-lap interval record from lap 110 on fell in 1990. Arie Luyendyk actually picked up the pace towards the very end of the race. The record stood until 2013.

Other notes

As promised above, I did want to mention one other unofficial stat. Officially the winner has completed the 500 miles in under three hours eleven times (see above). However, there was one other year that could have fallen under that category.

Without rehashing the controversy, it is worth mentioning that the 2002 race was completed (by at least two drivers) in less than three hours. Race winner Helio Castroneves was officially timed at 3:00:10.8714 (166.499 mph). He himself just missed out on completing the race under three hours. However, Paul Tracy and Felipe Giaffone (second and third place, respectively), were actually ahead of Castroneves on the track. Tracy was 19.4404 seconds ahead, and Giaffone was 18.2114 seconds ahead, amid the confusion. Had Tracy been declared the winner, his winning time would have been 2:59:51.431 and the average speed would have been 166.800 mph.


  • 1990 Indianapolis 500 Daily Trackside Report
  • 2012 Indianapolis 500 Daily Trackside Report
  • 2013 Indianapolis 500 Daily Trackside Report

EDIT: Updated article to reflect races from 2016-2019. (6/1/2019)

EDIT: Updated article; copyedits and other minor changes. (8/18/2020)

EDIT: Updated article; copyedits and other minor changes. (11/25/2021)

EDIT: Updated article to reflect 2022 race. (6/6/2022)