The Most (Un)Reliable Motorcycle Brands: Part II

Jul 19, 2023

A couple of weeks ago, we published a ranking of the most reliable motorcycle brands based on Riders Share's internal data. We measured tows per rental days excluding tows arising from flat tires and motorcycle accidents.

The #1 reason for a tow was the battery, regardless of brand. Our data suggests Royal Enfield could make a serious improvement in its brand reputation and reliability simply by purchasing better batteries or wiring for the batteries; it is unclear which of the two is the cause of the problems.

The brands at the top were a surprise (BMW and Indian). These brands attract a more experienced riding demographic - and let's face it, you probably won't survive 40 years of riding if you go over 120 miles per hour while riding drunk. As denoted in the article, we quickly realized risky riders need more tows, even when they don't crash the motorcycles.

So which brands are more "tolerant" of risky riders? Which brands can take some clutch abuse?

When looking at the rankings for low risk riders only, the results were pretty much the same. But when we redid them only including risky riders (inexperienced, young, bad driving records, etc.), the rankings changed drastically:

KAWASAKI 0.27%

DUCATI 0.32%

HONDA 0.43%

BMW 0.62%

YAMAHA 0.63%

Others 0.70%

SUZUKI 0.75%

TRIUMPH 0.95%

INDIAN 0.99%

HARLEY-DAVIDSON 1.07%

CAN-AM 1.46%

ROYAL ENFIELD 2.00%

We had to take out KTM because their results became statistically insignificant after reducing the population to only the risky segment. Our definition of risk is proprietary and based on neural networks that include other variables than the ones mentioned above.

Royal Enfield continued to dominate the bottom position, but there were several position changes at the bottom. Indian and Harley don't look so good anymore, perhaps because they attract some of the oldest and more experienced riders. The Japanese start to live up to their reputation when you factor in abuse. Ducati's tow rate is surprisingly similar for both risky and non-risky riders, suggesting perhaps it is not that unreliable after all.

Why don't you guys control for mileage or age of the bike?

These comparisons are not apples-to-apples if we compare an older Ducati vs. a new one, right? Well, our data suggests that is wrong. We couldn't find statistically significant differences in tow rates among bikes unless their mileage was over 80,000 and their model year older than 2005...and even then, very few bikes met those 2 conditions, so the significance is weak. In other words, it didn't appear mileage or year mattered that much. Bear in mind, Riders Share only allows motorcycles model years 2000 or newer, so we don't have data for bikes built before the 2000 model year. Starting on 2023, we are only allowing motorcycles 2005 or newer.

What about trip duration & miles per trip?

We looked at the tow rates based on mileage and the results were the same as looking at them on a per rental days basis, but we had missing or incomplete mileage data on 20% of trips so we went with days instead.

However, after thinking hard and long about it, we found a flaw in our data: if someone goes on a 10 day trip and they require a tow on the first day of the trip, usually we refund the 9 days but our system still registers it as a 10 day trip. This is a significant bias because longer trips usually can only have a max of one tow, but still keep the number of rental days. So longer trips are showing lower tow rates, but maybe not for the right reason. Plus, most trips are 1 day only anyways.

So if we remove this bias from the data, and remove risky riders, we get:

BMW 0.15%

INDIAN 0.17%

HONDA 0.38%

DUCATI 0.38%

HARLEY-DAVIDSON 0.39%

Others 0.41%

YAMAHA 0.52%

KAWASAKI 0.66%

CAN-AM 0.69%

SUZUKI 0.82%

TRIUMPH 0.87%

KTM 1.65%

ROYAL ENFIELD 5.69%

In short, Ducati does a lot better than in the regular rankings. but most other brands rank roughly the same.

What about maintenance?

Obviously, proper maintenance is the most important factor in predicting the reliability of a used motorcycle. At Riders Share, we require owners report the date of their last service when they list the bike, as well as the condition of the tires. However, this is insufficient to determine if the bike has had a long history of proper maintenance, and sometimes renters go on really long trips that put on a lot of miles on the bike, making those reports obsolete.

When you buy a used bike, you also don't really know the maintenance history of the motorcycle. You can get a Carfax report, but these are typically incomplete for motorcycles. A very large percentage of riders perform their own oil changes and there is no way for services like Carfax to pick that up.

So the data above is indicative too of what kind of maintenance the typical BMW or Indian owner gives their bike. They attract an older demographic, perhaps more responsible and wealthier, so they can afford to service the bike in timely fashion and not cut corners.

How is data collected?

We received emails complaining that our data wasn't trustworthy, so we are going to expand upon how we collect it.

Users order tows on Riders Share's website when they are stranded and have the option to select what type of assistance they need (ex. battery jump, flat tire, tow, etc.). Similarly, users file insurance claims on the website, so we have the ability to quickly remove tows caused by accidents from the dataset.

Additionally, we pause listings with mechanical issues until we receive proof these issues have been resolved, so it is rare that the same bike suffers a mechanical issue twice. This wasn't always the case, which is why our dataset starts in January 2021.

Following a suggestion from Dana Hooshmand, here are the confidence intervals. We can conclude that most brands are the same and only BMW/Indian are likely outperforming the average, but that's without controlling for the type of rider on the bike.

Only Royal Enfield is worse than average within a 90% confidence interval, although several brands come close:

Conclusion: This Rankings Are Best For Evaluating Used Bikes

Other than Royal Enfield, it appears the rankings are very sensitive to who rides the bike, and who maintains the bike. It is difficult to establish a clear difference in reliability by brand due to those two factors. As long as you do the regular maintenance, take care of your battery and don't abuse the clutch, your new bike is likely to never need a tow due to mechanical issues unless you have really, really bad luck.

Of course, if you are buying used, this is where the data above can come in useful. Buying a used BMW or a used Indian at a dealership, where you know nothing about the prior owner, minimizes the chances that the prior owner didn't perform maintenance and reduces the chances of a mechanical breakdown. Personally, my 3 year old, well maintained Kawasaki Vulcan started missing shifts, and it has under 10,000 miles, so I'll be looking at an Indian or BMW to replace it. I only bought it because there was no Indian dealer near me at the time, and I have rented Indians on Riders Share before and love them.

For motorcycle OEMs, there are a few useful tidbits. First and foremost, they need to focus on their battery and electronics, as they seem to be the main cause of tows for riders. Service departments need to clean them or suggest replacement of the battery if needed. Second, they can improve their brand perception simply by incentivizing maintenance of their bikes, and educating customers beyond providing the service manual (many people don't read them because they are too long, so cheat sheets would be nice).

Maybe a central database with the maintenance records of all bikes, like a credit score for bikes, would help riders make better buying decisions, while incentivizing proper maintenance? Perhaps that's a good startup idea?

Thank you for reading,

Guillermo Cornejo

CEO | Riders Share