It started out as a seemingly easy task, take all the NASCAR stats I could find and compile them into my own database. This way I could run queries of things I found interesting and just get information that normally I’d have to rely on other sites, such as Racing-Reference.info. Make an Access database and enter in data, that’s easy, right? WRONG!
What I learned quickly, and this isn’t getting into the obnoxiousness of Microsoft Access features, is no matter your source, odds are your information is wrong. And I’m not talking about Racing Reference or even DriverAverages.com (another fantastic, but lesser known site). I’m talking about information direct from NASCAR itself.
NASCAR historian and statistician Greg Fielden painstakingly compiled race recaps and results going from NASCAR inception in 1949 until 1989 in a four volume book set entitled Forty Years of Stock Car Racing. These books, along with its lone supplemental book for years 1990-1993, are a fantastic source of NASCAR history and stats. I found out recently he even has a similar book about the Convertible Series that was running during the early days, but do not count as NASCAR Cup Series statistics. Fielden (along with Peter Golenbock) also helped put together the NASCAR Encyclopedia, billed as the “complete record of America’s most popular sport.
Why do I gush about Fielden? Well, besides the fact he’s been able to do what I’ve always dreamed of doing in NASCAR (he also worked for CBS and TNN as their statistician for their broadcasts), it is because he and NASCAR are the only sources of information for any site that houses NASCAR stats. For example, Racing-Reference is built around Fielden’s work, whereas Driver Averages is built upon NASCAR’s stats.
The question then becomes, why does this matter? It matters because sometimes data from each source doesn’t matches the other. You’d think that if two people were keeping track of the same thing, you’d have identical results? Well, in this case, you do not. Fielden even devotes four pages in NASCAR Encyclopedia explaining why the data doesn’t match.
When it comes to the early days of NASCAR, not every track kept accurate accounts of the races, be it how the cars started or even how many laps many of the drivers complete. It makes sense when you think back to how things might have been in the late 40s and 50s when tracks were Mom and Pop operations. There was no digital scoring, everything was done by hand and stopwatch. Even re-watching races from the late 1980s I’m still amazed there were not more scoring issues even then. Most of the time it was simply the top ten finishers that were published in newspapers, with no other documentation.
If you haven’t guessed, I’m a Fielden supporter in that he reviewed everything he could to come up with the most accurate set of data there could be. That even means making common sense judgement calls (more on this later) rather than just looking at it as black and white numbers. This might be hard to imagine (sarcasm), but NASCAR has always been wishy washy on enforcing things, which brings us to the first and most glaring of issues with NASCAR official statistics (and that common sense piece).
Simple question, how many wins does Bobby Allison have in his NASCAR Cup Series career? You ask NASCAR fans this, most will tell you 84. That’s what the official record books say, that’s how he’s billed in anything NASCAR related. Well the reality of the situation is he has 85 career wins, but NASCAR has omitted one for reasons that aren’t very clear.
On August 6, 1971 something unique was done that day, the larger Cup cars raced against smaller Grand American cars. The Cup cars included makes/models such as Plymouth, Ford, Mercury, and Chevrolet. The Grand American cars included Mustang, Camaro, and Firebird entries. That day Allison drove a Mustang to a three second win over Richard Petty, in a Plymouth. Because he raced a Mustang, NASCAR considered the win in the Grand American Series rather than the Cup Series. But here’s the problem, the start, the top-5, the top-10, the laps led and the laps complete ALL count in his official NASCAR statistics, just not the win. Going a step further, as Fielden points out, the victory COUNTS for Mustang in NASCAR’s list of car makes/models with victories in the Cup Series. Where’s the logic in any of this? Either nothing counts, or it all counts, there can’t be an in-between.
According to the book, there was a meeting between historians and NASCAR in 2002 to try and hash out some issues from the past. It appears some have been updated, whereas others (mainly Allison’s win total) have not been, nor seem they ever will. For example, in the official listings for the final 1956 points, for many years there was the 39th place finishing Tom Harbison. Problem is there never was anyone named Tom Habison who raced in a NASCAR event that year or any other year. Now the record simply shows “unknown” as the 39th finishing driver.
While creating my database I’ve seen a number of odd situations where I don’t know where to turn to get a straight answer from. For example, in the official NASCAR database for race 24 of the 1975 season, Frank Warren is credited for driving Dean Dalton’s #7 car to a 12th place finish. If you pull open Fielden’s book, he has Dean Dalton as getting credit for the finish. As expected, Racing Reference sides with Fielden and Driver Averages sides with NASCAR. If you look for a third opinion, you get that of UltimateRacingHistory.com, who has on their site for that race a note that states: “Dean Dalton qualified car #7 and may have started the race.” So how do you decide this? Honestly, there’s no good answer because either way you choose, you then don’t have a cross reference anymore.
This is again where I side with Fielden over NASCAR because I’ve seen other examples of things that don’t make sense in the official record, but are different with Fielden’s work and make sense. Neil Castles in the 34th race of 1967, according to NASCAR, completed 311 of 200 laps. Fielden puts him at 190 laps completed. NASCAR has Jim Bown leading five laps during the 1983 Budweiser 400 at Riverside, but Fielden has Joe Ruttman leading those laps. I side with Fielden in this case because his results have a box score showing lap leaders, whereas NASCAR is just the finishing results.
Even with all that going for him, Fielden’s credibility takes a big hit with NASCAR Encyclopedia as there are many errors. The most glaring is how about half the father/son combinations who share the same name (Sr/Jr) also share the same stats. For example, Duane Carter is credited with starts in 1950, then 35 years later he has races in 1985, 1986, 1990, 1992, 1994 and 1995. The problem is Duane Carter has a date of death of 3/7/1993, making two starts after he died. Reality is Duane Carter (Sr) ran one race in 1950, then his son Duane Carter Jr (aka Pancho) ran the rest. Another issue is, not an error but an oversight, with the date of deaths for some drivers. Despite the book being updated in 2002, Kenny Irwin (died 2000) does not have a date of death, whereas Dale Earnhardt (died 2001) does.
I’m not trying to nitpick the man’s work, but that makes you start to question how you can trust it 100%, which is what my main point is, you can’t. Even when NASCAR and Fielden agree, it might not be accurate anyways. Case in point is the magical Doug Cooper. For race 38 of the 1966 season, Cooper is credited in both sources as finishing not only 10th, but 25th as well. Thankful Racing Reference has a more accurate looking result with Bob Cooper 10th and Doug 25th. The caveat to this without any citation, how do we know that, while sounding correct, is correct? Good news is I know the person behind Racing Reference has done a great job researching things, based on our conversations, so I trust it. Sort of.
That back tracking leads me to my greatest headache of the past three years and something I cannot get anyone (NASCAR, websites, or an International Motor Sports Research Center) to respond back with a reasonable explanation of what I’m missing. There are three drivers and two years where if you add up the points attributed to them via race results, it doesn’t match their end of year totals in the official final driver standings. The affected are Ricky Rudd and Tim Richmond in 1982 and Darrell Waltrip in 1984.
Ricky Rudd’s official total via the NASCAR Media Guide (NMG), their NASCAR Historical Database (NHD) for media members, and Fielden have him at 3,537 points. NASCAR.com, the official site of NASCAR, has him at 3,467 points. I, using a calculator first and then Excel to ensure I wasn’t adding wrong, have his total as 3,542. Here’s what I found to be crux of the issue is the 22nd race at Darlington. In that race, he finished 31st and according to Fielden and NHD, he led one lap. The 31st place finish netted him 70 points, and by leading one lap, he should have received 5 bonus points. Going back to the totals above, the NMG, NHD, and Fielden omit his 5 bonus points and NASCAR.com omits the 70 points for the finish. The additional points do not change where Rudd finished overall, but annoying it doesn’t add up.
Tim Richmond has the same scenario but with the 8th race at Martinsville. For finishing 18th, he should have received 109 points, and according to Fielden and the NHD, he led 18 laps that day, thus 5 bonus points. NASCAR.com and I agree, based on that, he should have ended the year with 2,611 points. Instead NMG, NHD, and FIelden have him listed at 2,497 points, less the 114. Most notably, drivers can sometimes not be award points if their entry was late for that race. Richmond, at this point of the year, was running for J.D. Stacy, which was a full-time team and it seems odd they’d forget that entry. Nothing out there references a reason why he wouldn’t be awarded points, nor any mention of a potential penalty that year that could have voided those points. Unlike Rudd, had Richmond been credited with those points he would jump from 26th place in the final standings to 23rd.
Darrell Waltrip has the weirdest case of any of the drivers because no one can come to agreement on what is going on. In the 10th race of the year Waltrip won a controversial race at Nashville, earning him 175 for the win and 5 additional points for leading a lap (Neil Bonnett led the most). In the 30th (and final) race of the season at Riverside, Waltrip finished 34th, with 33 laps led (Bobby Allison led the most), so 61 points for the finish and 5 bonus points. I calculated him at 4,230 points in total, which Fielden agrees with, but NMG has him at 4,235 (additional 5 points at Nashville) and NASCAR.com has him at 4,169 (minus Riverside, but with the extra Nashville points, which matches the NHD that awards him extra points at Nashville and zero at Riverside). Again taking the zero points angle, Waltrip was driving for Junior Johnson, so forgetting an entry seems very bizarre. Anyway you go, Waltrip still finishes fifth in the final standings, but why do we have such discrepancies?
In the end I will be taking up FIelden’s mantra of trying to produce the best database with the available data, but with a twist of common sense. I’m surprised that taking a look back at the records has never been a priority for NASCAR, if not to correct obvious errors, but to ensure that the official statistics are actually 100% accurate.