Thursday, February 02, 2012

Racing - Avoiding Crashes

Someone on BikeForums posted a great question the other day - how do the rest of us avoid crashes?

I never thought much about what I do to avoid crashes. I tend to focus more on skills of the rider, less on the environment. This means focusing on things like practicing touching wheels, bumping, cornering properly, braking, and, while on group rides, riding with strangers whose skill levels vary greatly.

One generalization I'll make is that the racers that avoid crits tend to be less adept at pack handling skills. Yes, they're very strong, usually much stronger than an average crit racer. But strength doesn't make a skilled bike racer, skills make a skilled bike racer. Avoiding time to practice this skill means you can't develop it. This is why a high level non-cyclist athlete (like a runner) would have to start as an entry level racer. It's not fitness in question, it's skill.

Criteriums offer the chance to practice the same corner again and again. I can watch different racers take the same turn, and it'll become obvious pretty quickly who's cornering well and who isn't. I'll avoid the ones that make me uncomfortable and follow the ones who demonstrate, if inadvertently, their skill at riding.

At any rate the BF member posted some of the following things he does to avoid crashes. It's a good list of standard things to do, i.e. the foundation for a set of guidelines.

1) I avoid being in the middle of the pack, instead opting to be on the edges. This gives me more of an 'out' if something happens. Sometimes this leaves you in the wind a little more (or a lot in a cross wind).

2) I try to stay to the front of the pack if possible. Not always easy with the way the pack moves.

3) If I can't be near the front, I stay near the rear. Staying in the rear takes a lot more effort, but lets me see things happen well before I usually get to them.

4) Stay to the inside of as many turns as possible. When someone falls in a turn, they slide to the outside. Staying inside means less likelihood of someone taking you out (but increases the accordion effect if you're not near the front).
Of interest to me was point 3, staying near the rear. I spent a bit of time working on a yet-unpublished clip of the 2011 Tour of Somerville Cat 2 race, and there's one crash where I was far enough behind where I could safely make a sweeping move to get around the crash. One rider slows me up a bit (he was just getting around the crash himself) but you can clearly see that I pass something like 1/3 of the field in my move.

Once the original post person ("original poster" or OP) put up his question, someone replied with a good but non-specific response.
...experience helps, a lot. Just like car driving experience... you develop a sixth sense that something is not right, and take evasive action before the actual wreck even gets started.
That, of course, got me thinking again. "Sixth sense" can be just a well-trained response. For example, blind people have claimed to be able to feel walls from a couple feet away. Their skin tingles and they turn away. Under normal situations the blind people so affected seem to be able to back up this claim. However, when the folks running the experiment put ear protection on the same blind people (ear muffs), the now-deaf people slammed into the walls.

What happened?

The blind people heard minor differences in the echo of sound of whatever they and others were doing (footsteps, breathing, etc) and instinctively turned away from the wall. The skin tingling was a conditioned response to this stimuli (or rather to this different stimuli).

What was really happening is that the blind people heard different levels of echoes and they developed responses based on this.

Likewise a sixth sense for avoiding crashes is more about being able to read the riders around you, body language, faces, pedaling style, group riding style/skills, etc.

Other than a mechanical (unclipped sprinting out of a turn) and having a guy swerve across my front wheel intentionally, as far as crashing in a race, I haven't crashed since the early 90s. I've gone off course, slowed, stopped, etc, but not crashed.

(And to be totally honest, I've unclipped probably 10 times in that time period without crashing so that one crash was kind of my fault).

I think there are two things that make the nebulous stuff, the sixth sense stuff - reading others' body language and technical practice.

Reading others typically means making a quick judgment on whether I'll trust someone or not. I'll use prior experience in addition ("profiling"), so maybe a rider is smooth but I know that they make abrupt path changes, therefore I'll adjust for that. I've sat 1-2-3" off of a non-racer's wheel, someone I don't know, on a shop group ride, simply because everything about their riding announced to me that it'd be safe.

There are racers who've been in pelotons with me for literally 20 or more years - I trust them implicitly, understanding that even if they make an error they're equipped with the skills to help make it through the incident.

I'll also not-trust a rider and back off another rider for any number of reasons. I avoid squirrely riders, even working hard to get ahead of them. Ironically I usually think that seeing someone follow someone else too closely is a red flag - if I see someone too close and they don't look super savvy, I avoid the cone area behind them where the mayhem will occur when they crash.

"Super savvy" - ever seen a master at something? Musician, martial arts, car driver, bike rider? They have a quiet, steady confidence, very fluent, automatically do things that others have to concentrate on doing. Those are the riders you want to follow.

Rough pedaling style is a big sign. Not necessarily a particular cadence either, although more high cadence riders scare me than low cadence ones because it's harder to control a bucking bike at high cadence. Solo type riders seem to focus on cadence quite a bit, and they're also typically less comfortable in a field.

Poor form means the rider will be tired, i.e. they either normally have good form and are now exhausted (I fall in that category often) or they have poor form overall (some may claim that also applies to me).

Tired riders make mistakes, that's all there is to it.

I take my own experience into account. When I'm totally on the limit I get dizzy if I look around too quickly. I don't think it's just me - I just watched a Het Volk where a tired rider looks back in the final sprint, brushes someone, and falls over. Julian Dean once led out Hushovd for a Tour stage sprint, turned back, and fell over. It happens. So when I see a rider who looks like they're on the limit, ragged pedaling, eyes kind of glazed over, not looking around much, I figure they're at that dizzy stage.

Guys who turn in too early are scared of turns. That's a huge warning flag and it runs across all levels of the sport. To wit - watch any Tour mountain stage and you'll see some very strong racers who can't corner well. I totally avoid the early turners, or stay inside of them on the exit because they'll go wide. When things are moving right along in single or double file, cornering well usually means turning in later than earlier.

Those that turn in early by themselves, not following the field, they're especially dangerous. They're turning in early because they're scared - they don't trust the field to stay upright. But by turning in early they're setting themselves up to exit wide, right into the field, not the best way to make friends in the pack.

Of course, after they take out a bunch of the field, they'll probably pick themselves up and say, "Yeah, man, no one knows how to corner. The whole field ran into me."

It reminds me of the joke where an elderly wife calls her elderly husband.
"Honey, you should watch out. The radio is saying that there's a guy driving the wrong way on the Interstate."
"One guy? What do they mean one guy?! Everyone is driving the wrong way on the Interstate!"

Don't be that guy.


In SoCal one year, on some winding descent I'd never been on before (Lilac? near the Lawrence Welk place), I started it near the back of about a 20 or 30 rider group. I only stopped blasting by riders because I got to the lead rider of the group. Since I didn't know where we were going I just followed him. Later my friend overheard some comment like "That guy from Connecticut knows how to descend." No, I wanted to say, that's not right.

I know how to corner, therefore I can descend.

I've said this before and I'll say it again: I'm extremely risk averse.

That's why I do criteriums.

It seems illogical but it really isn't.

(Of course the fact that I can't climb or time trial helps in making this decision. If I could just drop everyone I'd do road races all the time.)

Based on some of that nebulous stuff, I feel much more comfortable in a criterium than a road race. I see more poor riding in road races than in crits, probably because a lot of guys who don't like crits don't like them because they don't handle their bikes well/properly in a group.

Road races also hit much higher speeds, regularly hitting 55-60-65 mph on descents in my personal experience. Crashing at that speed could be life changing; even a minor problem could be beyond a good bike handler's capabilities.

Therefore it makes sense that I stay with the slow crits, where we rarely exceed 40 mph and normally ride 25-35 mph on the straights. It's easier to read people when they take the same turn a few times; within a few laps I'm comfortable with who I'm comfortable with, and I already have flags on the riders I don't want to be near.

It's an interesting question, asked by someone who's consciously made moves to try and reduce risk. Ultimately the rider needs experience and knowledge to reduce the chances of crashing - experience as far as reading riders, groups, and making judgment calls, and knowledge in skills, tactics, and applying them both appropriately.


Mark said...

Nice article. One thing I notice, as I race both Master's crits and Cat-3's, the 3's don't corner nearly as well. They ride into the corner shallow, have to slow more than necessary, then jump out of it to regain lost speed because of the original approach angle (shallow). The master's approach the corners wider and can then maintain their speed much easier during and after the turn. As a big guy, I can only conduct so many accelerations. I prefer the Master's races. In the 3's, if I'm tail-gunning and have room (always looking to make sure!), as we approach the apex of the turn, I can carve my own turn (wide in-tight-wide out) and this helps. But only if I know I'm not going to cut somebody off.

Aki said...

That shallow turn-in is describing that fear-of-turning syndrome, the early apex. It's what gets more people into trouble than anything else. It's a pet peeve of mine, when I see pros turning really early, even when other guys around them are turning in properly. It means they're scared of the turn.

I mainly race 3s so I can't really speak about the Masters. I'm guessing that most of the Masters understand cornering lines within the context of a field. Sometimes an early turn in is good but in general, when in single file, it's not.

Having said that, forcing others to jump is a good tactic. So it all comes down to what you want to accomplish.

Anonymous said...

As a CAT 4, I think equipment (tire selection especially) plays into it. I always laugh when I see someone "giving it a go", but are on a training tire. I understand there are budgets, but Gatorskins on a corner is asking for trouble.

However - how do you stay out of crashes....I think a lot of it comes down to the fact that most of the stronger riders can stay in the top 15 or so, and guess what....that in itself self-selects the better bike handlers or those more confident.

And I disagree on the "inside corner" issue, especially if the guy to your outside likes to pinch off others on corners...but this leads me to this..

Communicate....want to ensure the turn is taken correctly...inside riders should communicate for others to stay wide, or not pinch. But this goes throughout the race. The more crits you race, the more you can kind of see potential trouble.

From my experience, most of the crashes I have seen are due to slide-outs, especially if there is a corner near the finish line.

Anonymous said...

In terms of the cornering...any theories as to why people take corners incorrectly? My gut-reaction is again, most people out there alone do not need to corner like that under speed, and therefore, they are unsure of the ability of a tire to hold its grip in a corner.

Now - this gets somewhat back to equipment selection, but I have to admit, I have been guilty myself of cornering early because it felt as if that move would ensure the least amount of potential slide out.

Aki said...

Anon, I appreciate your feedback. However I see a few errors in your thinking:

1. When I race on clinchers, I'm on training tires (I almost never own "racing clinchers" since I have at least one set of tubulars for racing). I've never had an issue with traction on clinchers except in the rain (and I'm a bit iffy in the rain regardless so that's my fault, not the tires').

2. Proper cornering involves everyone. The inside can work but too often riders come in too much (as you said). The inside line thing is a general "rule of thumb". I actually don't follow this rule in many situations, but it's my default mode when dealing with an unknown field. If it's necessary to communicate then someone is doing something wrong, typically the one diving into the inside (whether from the inside or the middle of the field).

3. Slide-outs are, coincidentally, related to turning in too early, caused by fear. In fact you say that directly, that you turn in early to avoid sliding out. What you should know is that, in general, turning in actually contributes to sliding out.

In car racing, you'll see a lot of skid marks exiting the corner, where guys who have turned in too early lose traction as they try and stay off the gravel/grass. Sliding out on a bike is extremely rare when taking the right line; it's much more common when turning in too early and then having to turn harder to clear the curb. A late turn in gives you a lot of room to maneuver, even if something weird happens.

There are two instances when people turn in early inadvertently, at least in most crits:
A. Scared. The rider doesn't trust himself enough to turn in later.
B. Rider is desperately trying to move up, having been caught out of position.

In A, there's nothing you can do at the moment. If you find yourself turning in early, buy and play Gran Turismo whatever version. It really punishes early turn ins and really rewards late apexes. In more real examples the Gerolsteiner rider that crashed at Milan San Remo (and grotesquely broke his leg) was turning in too early for many turns, virtually all of the ones on the DVD. I couldn't believe he made it that far before smashing into, in his case, a pole and a wall. Michael Rasmussen is a great example of an early turner, ditto Levi. Andy Schleck usually is (but at least he can follow a good cornerer). It's not a fitness/ability thing, it's about practice. Heck, I used to turn in early. I learned to turn in late partially so I could go a lot faster on descents. It's more fun that way :)

For B there's a bit more to the picture. There are many ways to (safely) discourage riders from moving up the inside when it's dangerous. The key is to either slow enough for the turn or get better position before getting so desperate. There were a spate of pros crashing in the last turn of crits, usually because the guy went in super hot, inside, and simply couldn't turn hard enough. It's like anti-lock brakes - your stopping power is determined, usually, by your tires, because if your tires had enough traction, you wouldn't need anti-lock brakes. Likewise, if you go into a max 40 mph turn at 45 mph, you can't blame your tires for failing, it's a matter of staying within the tire's parameters. I've drifted to the outside of the first turn (the sweeping left) at New Britain following a Cat 2 attack there at 42 mph. We couldn't get around that simple wide bend at that speed and we both had to first ease, then brake, so we wouldn't end up in the woods. We both looked at each other, laughed, and waited for the field.

There are tactical reasons to turn in early (watch my Rent race where I lead out SOC) but that's a different thing altogether.

I hope this helps and let me know if you have questions or comments.

Anonymous said...


If you have some video guidance on turning in early, late, correctly that be used as a guide, that would really helpful.

Aki said...

I thought about this, and no, I don't have a current clip or post in queue that covers cornering in detail. The closest is a yet-unpublished clip of the Palomar descent (but it's really long, not sure how to break it up). It's an example of how to do it, but as for actual instruction... I have to work on some stuff. There are a few posts from before:

Descending stuff.

Cornering while driving

Part of how I learned to corner

More cornering off the bike