ON YOUR MARK

Contents

• Chapter 1 Leaving it to Chance
• Chapter 2 Setup, Rules and Equipment
• Chapter 3 The Right Dog
• Chapter 4 Behavior
• Chapter 5 Reinforcement
• Chapter 6 Operant Conditioning
• Chapter 7 Exercises
• Chapter 8 Behavior Chains
• Chapter 9 Passing
• Chapter 10 Typical Practice
• Chapter 11 Preparing for Competition
• Chapter 12 Clinic Outline
• Chapter 13 Course Outline
• Chapter 14 Problems
• Chapter 15 North American Flyball Association
• Chapter 16 Getting Started
• Chapter 17 Putting on a Successful Tournament
• Equipment - Jumps, Mats, Boxes and Backstops
• Appendix
• Bibliography
• NAFA Rules and Policies

Chapter 1 - Leaving it to Chance

The sun was just coming up over the tall pines. The cool night air had left a cover of morning dew on the grass. This peaceful park setting was to be the site of an annual dog show. Handlers, in their best show attire, were busy grooming their dogs. Dignified judges, peering over their eye glasses, were busy inspecting the rings. The show committee had labored for months to ensure everything would meet the discerning eye of the Kennel Club representative. But, as observed by Ellen van Wageningen, of the Sun Times,

"This peaceful and dignified atmosphere surrounding the Grey-Bruce Kennel and Obedience Club's dog show and obedience trial was temporarily shattered when Ron English walked to the center ring Tuesday at noon. A knowing look sparkled in the eyes of the dogs lying in the shade waiting to compete. A bark of excitement escaped one dog, another let out a yelp. More barking. Ears perked up across the park. Soon organizers and participants in the Flyball competition were yelling at each other to communicate over the canine commotion. All eyes were on one thing - the Flyball course. Two teams lined up four dogs, a handler holding back each excited participant....the first dogs exploded from the arms of their handlers...by this time the excitement had infected more than the dogs, as spectators and handlers were yelling encouragement at the top of their lungs."

Flyball racing is a fast and exciting sport, not only for the spectators and handlers, but for the dogs as well. There are four dogs on a team with an alternate or two. As the teams line up the excitement builds. With each dog eager to compete, sometimes it's all the handler can do to hold their dogs. A whistle signals the start of the race. The first dog from each team is released. Racing side by side, each dog races down its lane, skillfully hurdling a set of four jumps set ten feet apart. At the far end of each lane is a Flyball box. Each dog must step on a Flyball box pedal releasing a mechanism flipping a tennis ball from the cup. Hopefully, the dogs will catch the ball (if not, the dog must retrieve the ball) and return over the four jumps with the ball. As each dog crosses the start/finish line another eager dog is released. The lead changes several times as dogs and handlers alike make mistakes. Sometimes dogs chasing loose balls will miss a jump, or get turned around all together and return over the wrong set of jumps. The first team to have all four dogs successfully complete a run wins the heat.

Flyball, an offshoot of scent hurdle racing, was introduced in California in the early 1970s by Herbert Wagner. Scent hurdle racing was already popular around much of the country. In scent hurdle racing the dogs must go over four jumps and pick up one of four articles - the article scented by the handler. Each article is numbered 1 through 4. The dogs wore racing jackets identified with the same number as their article. Herbert Wagner did a Flyball demonstration on the Tonight Show. Soon, dog training clubs around the US and Canada were putting on Flyball demonstrations at dog shows, nursing homes, county fairs and schools. Many of the clubs had scent hurdle teams, making the transition to Flyball easy. Furthermore, Flyball added another dimension to their programs. Dog training clubs, most notably Sportsmen's Dog Training Club of Detroit and McCann vs. McCann in the Toronto area, introduced Flyball to the Great Lakes area. Other dog training clubs in the area picked up on the sport and tournaments were soon organized. As more and more tournaments were held it became apparent that it would be necessary to agree on one set of rules. Thus, in November 1984, twelve teams from the Detroit/Toronto area founded the North American Flyball Association.

I don't remember where or when I first saw Flyball. I do, however, remember the first time Duffy, my Old English Sheepdog, went over all four jumps, triggered the box and returned over all four jumps with the tennis ball. She had only been introduced to Flyball a few weeks earlier and I was excited, as most people are when their dog puts it all together for the first time. Like most dogs involved with Flyball at our club, Duffy had some obedience training. She had earned an obedience title, Companion Dog Excellent (CDX), from the American Kennel Club. To train Duffy, I led her on lead back and forth over the four jumps, again and again for several weeks. Occasionally, quite by accident or with a little help, she would step on the pedal. Eventually, left to chance, even Duffy got the idea. That was the way Flyball was taught then, and still is in many training clubs - there were no books on Flyball, no videos, no training manuals. However, Duffy, like many other dogs, never became reliable. But then, an Old English Sheepdog and Flyball always pleased the crowd regardless of whether or not she was reliable.

Our second dog, Aubrey, a Golden Retriever, was trained and handled by my wife,Beckie. We started Aubrey much the same way. Before participating in Flyball our training club required dogs have at least an AKC title, Companion Dog (CD). Unlike Duffy, Aubrey had not been introduced to any of the open exercises, namely, jumping and retrieving on command. Needless to say, any Golden Retriever worth anything at all will retrieve. For Aubrey the training was quite stressful, particularly the way it was taught then and sadly is still being taught in many clubs by misinformed obedience instructors. None the less, Beckie proceeded to train Aubrey. Initially, Beckie kept Aubrey on lead. Again, as I had done with Duffy, Beckie ran Aubrey back and forth over the jump, and occasionally, quite by accident or with a little help, Aubrey, like the dogs before her, would step on the pedal. It wasn't long before Aubrey was too fast to keep on lead, so we set up gating. Everything went well as long as the gating was in place. As we began to remove the gating, Beckie would run alongside, leading Aubrey over the jumps off lead. I could not understand why any dog, let alone a happy, healthy, strong Golden Retriever, would choose to go around a jump rather than over a jump. Beckie seemed to be the only one benefiting from the daily workouts. Finally, again quite by accident, Aubrey got the idea of going over the jumps at one of our many nursing home visits. The floors had just been waxed. Aubrey, unsure of her footing, made every effort to stay on the mats. There was only one mat as we were doing the demonstration in a hall. The jumps just happened to be in her way. Fortunately, Aubrey's experience with the waxed floor carried over when we practiced at the club or put on other demonstrations. However, I am not endorsing this technique, nor any technique that would leave anything to chance. There are more effective techniques to shape desired behavior.

Nevertheless, after nearly a year of running Aubrey back and forth over the jumps, over and over again, we decided Aubrey was ready to compete on the team. Although she was not particularly fast, she was reliable, or so we thought. In her first tournament, much to our surprise, she seemed to have forgotten everything she had learned. Later we would attribute this apparent lapse of memory to a phenomenon that Karen Pryor, a behaviorist and dolphin trainer, refers to as the "new tank syndrome". It seems that regardless of how well dolphins learn a given behavior, the first time they are placed in a new tank they seem to forget all previously learned behavior. Unfortunately, we didn't recognize it as such at the time, and we went back to running Aubrey back and forth over the jumps, again and again.

After months of training we finally thought Aubrey could compete again. We entered several tournaments the following year and Aubrey had indeed become quite reliable. Soon after, we noted Aubrey was faster as well. At the time we credited the improvement to experience. To some degree we were right, but not entirely. The improvement was three fold - experience, breaking down the exercise and obedience training.

By now we had started to break the Flyball exercise down, primarily at the advice of Jane Justice. Jane had a Rottweiler in Flyball. Rosco had learned Flyball in a very short time, but, unlike others dogs that had learned Flyball in a short time, Rosco was reliable. Jane had broken down the exercise, working the jumps and the box separately. This seemed to be a very successful technique, and to my knowledge had never been done before in Flyball.

Furthermore, Beckie, discouraged with obedience training, began taking lessons from Ted Aranda, a top dog trainer. Ted had done extensive work in operant conditioning and motivation, and had great success applying those principles to obedience training. Ted also encouraged handlers to break down the obedience exercises in small increments, teaching each exercise independently and then building the behavior in blocks or chains. Ted was working with Beckie and Aubrey to reduce the level of stress Aubrey had displayed with various obedience exercises. What Aubrey was learning in Ted's obedience class had carried over to Flyball. Soon, we began to apply those same principles to Flyball. We found, while the exercises were different, the principles of shaping behavior, backward chaining, positive reinforcement and reducing stress could be applied to Flyball, just as Ted had applied them to obedience. Indeed, many of the principles had been used for years to train tracking and other working dogs. Furthermore, Ted introduced me to some work by Karen Pryor, a dolphin trainer, behaviorist and author of "Don't Shoot the Dog", "Lads before the Wind", and "Teach Your Dog Frisbee" . She had been using the techniques of operant conditioning, shaping behavior, backward chaining, and positive reinforcement for years to train dolphins as well as dogs, although she admits that her experience with dogs is limited. By now we had our third dog, Wyn, a Border Collie puppy, and I was not going to leave anything to chance.

Dog Training is an active ongoing process that should not be left to chance. There are numerous techniques of documented success that trainers can apply. Techniques based on behavior modification studies and operant conditioning can be used to train top Flyball dogs. A systematic approach to shape your dog's behavior takes more time to see results than simply running your dog back and forth, again and again. However, we have found the dogs are more reliable, highly motivated, and self reliant. When we purchased, Wyn, our first Border Collie puppy, several trainers also purchased Border Collies, convinced as I was that Border Collies were naturally suited for Flyball. I took a systematic approach to training Wyn. Of course, I wasn't the first, nor the only one, to use these techniques in Flyball. Wyn and those trained this way turned out to be excellent Flyball dogs. Of the others, whose training was left to chance - many also turned out to be excellent Flyball dogs. Many more did not. Even today, many of those that happened by chance to be excellent Flyball dogs still have occasional problems--problems that are not easily corrected.

While the techniques pointed out in this manual have been successful for me, it is not to say it is the only way to train a dog. Even left to chance many dogs do very well. Before we explore the techniques to shape our dogs behavior, it is necessary to look a little closer at the sport of Flyball, the set-up, some rules, and the equipment we will need for training. In later chapters, we will explore behavior and the principles of shaping behavior. In the end your dog will be happy, self-reliant, highly motivated, in control, intense and reliable. Of course, you could leave it to chance.

Chapter 2 - Set up, Rules and Equipment

As mentioned in the first chapter, Flyball racing is a fast and exciting sport. There are four dogs on a team with an alternate or two. Racing side by side, each dog races down their lane. At the far end of each lane is a Flyball box. Each dog must step on the Flyball box pedal, releasing a mechanism that flips a tennis ball from the cup. The dogs catch the ball (if not, the dog must retrieve the ball) and return over the four jumps with the ball. As each dog crosses the start/finish line another dog is released. The first team to have all four dogs successfully complete a run wins the heat.

Figure 2-1

Each lane is from 10 to 20 feet apart. The start/finish line is 6 feet from the first jump and each jump is 10 feet apart. The Flyball box is set 15 feet from the last jump. At the far end, placed behind the Flyball boxes, is a backstop. The backstops prevent loose balls from rolling into the spectators, under chairs and into the opposing team's lane. Figure 2-1 shows a typical ring setup.

Flyball jumps are solid, white, and are to have an inside width of 24 inches (between the uprights), with the height varying from 8 inches to 16 inches (minimum and maximum heights). The jumps are set four inches lower than the shoulder height of the smallest dog on the team. The jump height shall be adjustable in one inch increments. The uprights are to be no more than 36 inches high, nor less than 24 inches high.

The Flyball box can be any design as long as it performs the desired function; namely, the dog depresses a pedal on the front of the box that releases a tennis ball for the dog to catch (or retrieve should the dog fail to catch the ball). Near the back of this book is a list of several commercially available Flyball boxes, as well as a few designs, if you happen to be handy with tools.

Since this book was written, NAFA has added some restrictions to boxes used in tournaments. The maximum width, height, and depth is 24", 18", and 30" respectively. Furthermore, the ball must be thrown such that it lands on the ground 2 feet in front of the release point.

Other than grass, the ideal running surface is soft, durable sponge vinyl ring matting. Rubber matting, in addition to being heavy, does not give sure footing, nor does it cushion the shock of landing. Carpet alone can cause severe rug burns after only a few runs; however, carpet with vinyl matting is an excellent surface for Flyball.

The North American Flyball Association is the ruling body for NAFA-sanctioned Flyball competition. A complete set of NAFA Rules and Policies can be purchased from NAFA. Before you compete in or host a NAFA sanctioned Flyball tournament you should have a thorough understanding of NAFA Rules and Policies. Without going into great detail, the following rules generally apply:

• The handler must stay behind the start/finish line.

• Start and Passing: The first dog cannot reach the start/finish line before the judge signals the start of the heat. The second dog released cannot reach the start/finish line before the first dog reaches the start/finish line on the return run - the same being the case for the third and fourth dogs and any dogs that must run again.

• The heats will be restarted for a false start (first time only), or if the box malfunctions (again first time only).

• A dog must run again for the second false start, an early pass, missing a jump, dropping the ball before reaching the start/finish line, or if the dog is assisted by the handler, a spectator, or the box loader.

• A team will forfeit the heat for interference, if the dog fouls in the ring, if the box loader calls a malfunction and the judge finds the box to be working, or if the box malfunctions a second time during any remaining heats within the race.

You should carry a water jug with you, plenty of treats and plastic bags to pick up where your dog leaves off. You will need a training collar and a plain leather or web collar - one that does not tighten around the dog's neck when it is being restrained. As previously mentioned, sometimes it is all a handler can do to hold their dog. Some handlers prefer Flyball collars - web collars with an extra strap or handle sewn on the outside. One individual even used a tracking harness to restrain her dog. A tracking harness is non-restrictive when the dog runs, and yet the dog can still breath while being restrained. Most handlers have learned to restrain their dogs by holding them around the shoulders or better yet, they train their dog not to pull on the lead.

I prefer a three foot leather lead. Although, the more common six foot leather or a cloth web lead is fine. I seldom use a lead when training. A lead may not be necessary and can even hinder the training - its just one more thing to deal with. However, a lead is required when there are strange dogs around or when taking your dog to and from the building or where there is any traffic. The degree of control you have on your dog and the degree of control others have on their dog will dictate whether or not to use a lead.

You will need a long line for retrieving, teaching the command come and quick turns. In the early stages of teaching the retrieve some dogs will not want to bring the ball back. By having the dog on a long line you will guide the dog back. A 40 foot web cloth lead is preferred for corrections.

Finally, you may want a retractable Flex-lead. However, Flex-leads have limited applications. A Flex-lead will retract and allow you to work with the dog without getting the lead tangled up with the jumps or around the dog's legs. With some dogs it is virtually impossible to run with them and a long line will just get tangled up. Flex-leads can thus be used in warm-ups and practice to correct the dog for a slow turn, for going after loose balls or chasing other dogs. Flex-leads are also great for taking the dog for a walk.

Chapter 3 - The Right Dog

---

For an up to date analysis of all breeds that have competed in NAFA sanctioned tournaments, visit the NAFA Breed Analysis page.

---

Flyball competition is generally divided into divisions based on a team's best time. Thus, divisions allow for a variety of breeds. Teams with one or two Newfoundlands are not likely to compete against teams with four Border Collies.

Furthermore, jumps are set four inches below the height of the smallest dog on the team. Many teams have a smaller dog - allowing them to lower the jumps for the other dogs. While this can be an advantage, it is not to say that four large dogs are necessarily at a disadvantage. Indeed, many NAFA records have been set by four dogs jumping the full sixteen (16) inches.

From time to time, competition is offered for teams with multibreeds only - all four dogs racing must be of a different breed. Multi-breed only competition, divisions, and the jump height make Flyball fun for a variety of breeds. As of the fall of 1990, 712 North American dogs had earned points toward a Flyball title - 210 had earned the title Flyball Dog (FD), 156 had earned the title Flyball Dog Excellent (FDX) and 166 had earned the title Flyball Dog Champion (FDCH). Table 3-1, list the many different breeds (and number) registered with NAFA as of the fall of 1990.

---

Table 3-1

Breed	Number
Airedale Terrier	3
American Eskimo	3
American Pit Bull	2
American Staffordshire Terrier	2
Australian Cattle Dog	9
Australian Kelpie	1
Australian Shepherd	15
Beagle	1
Bearded Collie	6
Belgian Malinois	1
Belgian Sheepdog	7
Belgian Tervuren	24
Bernese Mountain Dog	1
Bloodhound	1
Border Collie	148
Border Terrier	4
Borzoi	1
Portuguese Water Dog	1
Boston Terrier	1
Boxer	4
Rottweiler	7
Breed Unknown	38
Brittany Spaniel	2
Cairn Terrier	1
Chesapeake Bay Retriever	1
Schipperke	1
Scottish Terrier	1
Chocolate Labrador Retriever	2
CKC Spaniel	1
Cockapoo	1
Smooth Coat Collie	2
Cocker Spaniel	28
Corgi	11
Dachshund	1
Dalmatian	4
Doberman Pinscher	43
English Springer Spaniel	14
Flat-Coated Retriever	10
German Shepherd Dog	21
German Shorthair Pointer	13
Giant Schnauzer	3
Golden Retriever	59
Great Dane	1
Heeler	1
Irish Setter	2
Jack Russell Terrier	3
Keeshound	4
Kerry Blue Terrier	1
Labrador Retriever	58
McNab	1
McTavish	1
Miniature Poodle	5
Miniature Schnauzer	7
Mix Breed	47
Newfoundland	3
Nova Scotia Duck Tolling Dog	4
Old English Sheepdog	1
Queens Heeler	2
Rough Coat Collie	1
Saluki	1
Samoyed	2
Shetland Sheepdog	47
Siberian Husky	2
Silky	1
Soft Coat Weaton Terrier	1
Staffordshire Bull Terrier	4
Standard Poodle	6
Vizsla	1
Weaton Terrier	1
Weimaraner	2
Welsh Terrier	1
West Highland White Terrier	1
Whippet	1

---

Neither females nor males have a particular advantage. I prefer training females. Males, on the other hand, are generally stronger and faster, but females are lighter and quicker off the box. With some breeds the natural stride of the large males, compared to the female, can be a problem - for example, the jumps only being 10 feet apart can be a problem for the larger male Doberman Pinscher.

I am not a strong supporter of puppy testing. While puppy testing might indicate extremes, they will not indicate whether or not a dog is going to be a good Flyball dog. None of my dogs showed interest in retrieving at 6 or 7 weeks. To see them now you might have a hard time believing they were several months old before they would bring me a tennis ball.

If you are buying a puppy buy one from working parents. These puppies will generally have all the instincts and intelligence not found in a majority of the conformation dogs. There's not much you can tell about a dog's natural instincts when the dog spends the majority of its time in the kennel or being paraded around a show ring. The instincts that make the dog a good working dog will also make it a good Flyball dog.

If there is a standard for a Flyball dog, the dog should be friendly and outgoing, fast and agile, one who loves a tennis ball and has a dedicated handler and trainer. Dogs that are not outgoing and friendly can make excellent Flyball dogs, provided they are not aggressive when excited, while playing or otherwise. During competition or practice the dogs can become excited. In playful excitement, dogs have been known to bite other dogs and even handlers. To my knowledge none has ever bitten a judge or spectator. Nonetheless, aggressive behavior whether out of playfulness, because of the dog's natural instincts or otherwise is not allowed in Flyball. Aggressive dogs will be excused from competition and reported to the NAFA. Depending on the nature of the aggression or in the case of a second offense, the dog will be barred from competing in an NAFA-sanctioned tournament. Reinstatement will be difficult.

In summary, I prefer females to males, medium size dogs from 18 inches to 22 inches at the withers, friendly, outgoing, fast and agile, and tennis ball crazy. They need not be black and white, but black and white dogs go nice with our black and white uniforms. Border Collies and female Doberman Pinschers seem ideally suited for Flyball. However, most any breed if trained properly is capable of running under six seconds. There are few teams that could not use a dog that runs under six seconds. Some of the smaller breeds jumping 8 to 10 inches will not run under six seconds, but they do lower the jumps for the team. Again, there are few teams that wouldn't love a small dog on their team. The dog need not be a natural retriever, but it helps. Any dog, purebred or mixed, from the Humane Society or from a long line of champions, will make an excellent Flyball dog, provided the dog does not have any abnormal problems due to breeding, training or previous experiences.

Chapter 4 - Behavior

---

Table 4-1 Favorable Behavior	Table 4-2 Unfavorable Behavior
Focused and intense Quick start Driving over the jumps Trigger the box Catch the ball Reliable retrieve Returning over all four jumps Good pass Self-reliant	Dog fails to catch the ball Dog fails to retrieve the ball Dog hesitates at the box Slow or wide turn Slow outrun or return run Dog is easily distracted Dog cuts the jump Bad pass Dog drops the ball Dog knocks down a jump

---

Just as there is a distinction between favorable and unfavorable behavior, trainers need to make a distinction between unfavorable behavior, undesirable behavior and problem behavior. Unfavorable behavior is associated with training problems, such as cutting the jumps or dropping the ball. Unacceptable behavior might include not coming when called, and uncontrolled play. Problem behaviors include housebreaking, chewing, digging and aggressive behavior. The only problem behavior we will address is aggressive behavior.

To correct behavior we have several options as pointed out by behaviorist and trainer, Karen Pryor. The trainer could shoot the dog; correct the behavior with negative reinforcement (collar correction); put the behavior on cue; shape the absence of the behavior (reinforce any and all behavior that is not the unfavorable behavior); change the motivation; ignore the behavior (extinction); or focus on training an incompatible behavior.

Shooting the dog, putting the behavior on cue, shaping the absence of the behavior and changing the motivation are techniques generally reserved for serious behavior problems. It is not the intent of this manual to deal with serious behavior problems. Therefore, our discussions will be limited to using negative reinforcement, ignoring the behavior and training incompatible behavior; that is, training favorable behavior that is incompatible with unfavorable behavior.

Correcting behavior with negative reinforcement is effective in both unfavorable behavior and unacceptable behavior provided the trainer keeps in mind two concepts. First, corrections are never used to teach the dog a new behavior. Second, corrections vary in intensity on a scale of one to ten, with one being very low in intensity and ten being very high in intensity. If you are trying to shape behavior you correct unfavorable behavior by starting with an intensity of one. I seldom progress beyond a simple "eh-eh." On the other hand, to stop or correct unacceptable behavior, start with an intensity of ten.

The other technique to correct unfavorable behavior is to ignore the behavior. If you elect to ignore the behavior, always follow up by training the dog to a behavior incompatible with the unfavorable behavior. If, for example, you elect to ignore the dog when it cuts around a jump, and I often do, then always follow up by training an incompatible behavior; namely, going over the jump. To train an incompatible behavior, you may have to regress to a point where the dog can not fail - to a level where success is guaranteed.

Chapter 5 - Reinforcement

Reinforcement is not a process of bribes and rewards. A bribe is a means to solicit, to request or plea for a response that has not yet occurred. Rewards occur after the behavior. Since you cannot explain to the dog that the reward is for past behavior, the dog does not correlate the reward with the behavior. This can be said even of rewards for behavior that occur only moments or seconds before. Rather, dogs learn to associate rewards with ongoing behavior. The same can be said for punishment. Since you cannot explain to the dog that the punishment is for behavior that occurred a few moments before, then the dog cannot correlate the punishment with the behavior. Bribes, rewards and punishment will not shape a dog's behavior. Only reinforcement paired with the behavior can effectively shape behavior. The same is true of corrections. Only corrections paired with the behavior can effectively shape the behavior.

When training dogs, immediate feedback is essential. If reinforcement is not immediate, any ongoing behavior will be reinforced in lieu of the desired behavior. Furthermore, unless reinforcement occurs while the behavior is ongoing, it cannot be considered reinforcement, but rather a bribe, a reward or punishment. Although food and toys, even a game of chase, are often used to elicit, draw forth or bring out the initial response, they should never be used to reward a response, nor solicit, request or plea for a response.

Reinforcements derive their ability to shape behavior because they serve as motivaters and provide information or feedback to the dog. Both are essential to the learning process. Positive reinforcement increases the probability that the behavior will occur again. A correction is something the dog will work to avoid. Since reinforcement and corrections are feedback to the dog, it must be timely and predictable. Commands, signals or cues that are intended to get the desired response must predict positive reinforcement. It is not enough that the dog simply respond to the command, signal or cue. More important is that the correct response predicts positive reinforcement. Further, failure to respond must predict a correction. Consequently, the dog will learn to avoid behavior that predicts a correction and will look to repeat behavior that predicts positive reinforcement.

Primary reinforcements are those that reinforce behavior independent of an association with other reinforcers. They are often associated with satisfying some natural instinct or drive. Examples of primary reinforcement include food (hunger), a toy (need to play) and a game of chase (need to hunt). It is important to note that while a ball may satisfy one dog's need to play it may not satisfy another. Therefore, select an appropriate primary reinforcement. Primary reinforcement can also be negative such as a collar correction. All animals have natural instincts to withdraw or escape adverse stimuli, just as you or I would withdraw our hand from a hot stove burner.

A conditioned reinforcement's ability to reinforce behavior results from an association with a primary reinforcement. Conditioned reinforcements are as real and as potent as primary reinforcement. For example, most dogs learn to associate the owner picking up a set of keys with something pleasant. The keys become a real and potent conditioned reinforcement. Conditioned reinforcement are useful in dog training when it is impossible to immediately reinforce the behavior with a primary reinforcement. The conditioned reinforcement acts as a bridge. The trainer signals the dog that "Yes - you are right" with a conditioned reinforcement. Moments later, the primary reinforcement becomes a reward for a behavior that has already been reinforced with a conditioned reinforcement. Although the primary reinforcement has no effect on the behavior being shaped, it does help to establish a positive attitude toward training and reinforces the conditioned reinforcements ability to reinforce.

The schedule of reinforcement may be continuous or variable. With a continuous schedule of reinforcement the reinforcement is given for every appropriate behavior. Since learning is more rapid with a continuous schedule of reinforcement, and since we desire to establish a correlation between the reinforcement, the command and the behavior, follow a schedule of continuous reinforcement until the behavior is established. However, once the behavior has been established, constant reinforcement results in saturation. And, selective and variable reinforcement are essential to raise the criteria or maintain the behavior.

A variable schedule of reinforcement has been found to be effective in maintaining learned behavior. Furthermore, variable reinforcement produces more durable behavior. Consequently, all the things that can go wrong at a Flyball tournament are less likely to effect behavior established with continuous reinforcement, but maintained with variable reinforcement.

The bridge between continuous reinforcement and variable reinforcement is a schedule of selective reinforcement. You shape the desired behavior with selective reinforcement. Selective reinforcement is a process by which you reinforce only your dog's best efforts. Further, using selective reinforcement you can raise the criteria. The dog trying to overcome some obstacle will try several responses, ultimately hitting one that gets reinforced.

As your dog more consistently performs the desired behavior, your selective reinforcement begins to more closely resemble a schedule of continuous reinforcement - nearly every effort is reinforced. At this stage the trainer maintains the behavior with a schedule of variable reinforcement, then further shapes the behavior by again raising the criteria using selective reinforcement. By going from selective reinforcement to variable reinforcement, and then from variable reinforcement to selective reinforcement, and so on, you continually raise the dog's response to a higher level of performance.

Food is an effective motivation in training dogs because food is a primary reinforcement. Since food is a primary drive or motivater for dogs it has the ability to reinforce behavior independent of an association with other reinforcers. Food, unlike praise, generates a specific response; it motivates decision and action. Although praise is an excellent reinforcer, it does not in itself generate a specific response. Praise alone does not motivate decision and action. Food, like praise, can also be used to relieve stress.

Furthermore, food or other primary reinforcers when paired with a particular cue, establishes the cue as a conditioned reinforcement. The cue can be a command such as "Jump," "Come," "Sit," or "Stay," or the cue can be a phrase such as "On Your Mark" or "Lets go for a walk." The cue can also be an object such as keys, a tennis ball, a Flyball box or even a jump.

People that use food improperly make the mistake of assuming behavior shaping techniques do not work. Frequently, handlers continue to use food as a target or bribe beyond the initial stages of the learning phase. Or, food is given too late and the wrong behavior is reinforced. Furthermore, if food is not paired with a cue, or command, there is no correlation between the primary reinforcement and the cue. Sometimes, food is even given with a correction, giving the dog conflicting feedback. Also, if food is given for no reason or for a poor response to the command, you are giving your dog feedback that does not communicate to your dog what you expect.

Still others stay on a schedule of continuous reinforcement, never raising the criteria or never going to a schedule of variable reinforcement. If a constant schedule of reinforcement is maintained for any length of time, without raising the criteria, the behavior will fade. Eventually, it becomes nearly impossible to raise the criteria. If you try to raise the criteria after being on a schedule of continuous reinforcement for any long period of time, the dog will quit, discouraged that it is not being reinforced on the continuous schedule that it has come to expect. The dog may even perform the exercise several times, but try and raise the criteria and the dog is confused because it did not get the reinforcement and will quit trying. The food then becomes nothing more than a bribe that may work a few times at best. A variable schedule of reinforcement is essential to maintain the behavior or to raise the criteria.

Chapter 6 - Operant Conditioning

A technique used in operant conditioning is a term called shaping. Shaping is a process by which the trainer first determines how closely the dog is able to perform a desired behavior with minimal assistance. Once the initial level is determined the trainer uses food as a target to elicit a response from the dog. With each successive effort the trainer further shapes the behavior. The dog's behavior is further shaped each time by reinforcing successive approximations. Each successive approximation or response more closely resembles the desired behavior.

During the learning stage the dog is always set up to succeed as the behavior is shaped in small increments in the direction of the desired behavior. One school of thought is to set the dog up to fail - thus the trainer is given an opportunity to correct the dog. The dog learns very quickly what not to do, and is left to guess what the handler wants. Stress training is not effective. It's inhumane and cruel. While mild corrections are important in helping the dog during the learning stages and shaping the dog's behavior, corrections that stop the learning process should not be used to shape behavior.

Behavior shaping focuses on the dog's ongoing behavior. More important than the causes related to the past is what is happening at the moment. By observing the dog's behavior, we can begin to shape the desired behavior. If the behavior is not precisely what is desired, we shape the behavior starting at a point where we assume the behavior is yet unlearned. As you shape the dogs behavior remember:

• Choose an appropriate reinforcement.
• Use small reinforcements.
• The behavior controls the reinforcement.
• Corrections must be unavoidable, immediate and paired with the behavior.
• Reinforce avoidance responses, not escape responses.
• Follow escape responses with avoidance responses
• Be predictable.

Use small reinforcements. In shaping behavior, many small reinforcements work better than a few big ones. A dog will learn more quickly for ten small reinforcements, rather than one large reinforcement for an equal number of behaviors.

The behavior controls the reinforcement. When the reinforcement controls the behavior it no longer is reinforcement, but rather it is a bribe.

Corrections must be paired with the behavior--unavoidable, immediate and sufficient to interrupt the behavior. Corrections are feedback to the dog and lead to avoidance responses. Just as positive reinforcement makes behavior more likely to recur, corrections, which initially induce an escape response, also make certain behaviors more likely to occur, as the dog learns to avoid the behavior that predicts a correction. Corrections must be unavoidable and paired with the undesirable behavior; otherwise, the dog quickly learns to avoid the correction--not the behavior associated with the correction.

Furthermore, if the correction is not paired with the behavior the correction is merely punishment which the dog does not understand or associate with the undesired behavior. Therefore, the correction must be paired with a cue. Just as a child learns to associate the cue "HOT" with something very unpleasant, the dog should learn to associate the cue "NO" with something very unpleasant, namely, a collar correction. This is initially taught independent of any ongoing behavior for three reasons. First, because when the trainer needs to use the command it is highly unlikely that the dog will be able to learn to associate the correction with something unpleasant. The first time most handlers use the cue "NO" is when the dog is going through the garbage, chewing on a piece of furniture, or chasing a squirrel. Indeed, most dogs learn to associate the cue "NO" with something fun. What they cannot understand is that the owner is not equally as excited.

Second, when the dog is playing or being aggressive, corrections are rarely intense enough to effect the ongoing behavior. The dogs simply do not feel the correction. Yelling and corrections at that point, unless paired with something unpleasant independent of the behavior, only further excite the dog.

Finally, in a complex behavior chain such as Flyball where one behavior immediately follows another, it becomes impossible to pair the correction with any one behavior. Within five to six seconds as many as 15 to 20 behaviors occur - some simultaneously. Pairing the correction with any one behavior cannot be done, unless the correction is paired with the behavior independent of the behavior chain.

Reinforce avoidance responses, not escape responses. If you correct the dog, the response will be an escape response. If you then reinforce that response with a piece of food you are giving the dog two pieces of conflicting information - one positive and one negative. Does the response predict positive reinforcement or negative reinforcement? Do not reinforce an escape response, unless the dog is stressed by the correction. If however, the dog's response to the cue is an avoidance response (namely, the dog avoids the behavior that predicts a correction), the response should be reinforced. All too often the dog will either respond to the cue, and get corrected anyway, or the dog will not respond, will get corrected and then be reinforced. The dog will eventually learn, but the learning process will be very stressful for the dog.

Follow escape responses with avoidance responses. Corrections should be feedback to the dog. However, the correction should not end there. Corrections should help the dog, tell the dog it is wrong, followed by giving the dog an opportunity to correct itself. Only in this way will corrections become valuable feedback to the dog, rather than a source of stress. The learning process is much improved when you are told, "No, you are wrong, try again .... Yes, you are right."

Be predictable. Being predictable is more than being consistent. Being predictable is the basis for all communication between you and your dog. Corrections must be predictable, so the dog learns to avoid the behavior that predicts a correction, not the correction itself. Reinforcement must also be predictable.

Additional laws for shaping behavior should include the following from Karen Pryor's book, "Don't Shoot the Dog":

• Raise the criteria in small increments.
• Shape one behavior at a time.
• Always put the current level of response on a variable schedule before raising the criteria.
• Plan your shaping program.
• If one shaping technique is not working, try another.
• If the behavior breaks down, repeat the shaping process.
• Avoid interrupting a training session.
• End each exercise on a positive note.
• When introducing new criteria, temporarily relax old criteria.

Shape one behavior at a time: Do not shape two behaviors simultaneously. It is not always easy to distinguish between two simultaneous behaviors and a single behavior. To avoid this pitfall, determine ahead of time what particular behavior it is that you want to shape. If it is quick turns, work on quick turns, not quick turns and catching the ball.

When shaping unlearned behavior the dog needs time to process the information. Working on two behaviors simultaneously is asking the dog to process too much information. Either one behavior will override the other or the dog will become confused and either quit or make mistakes trying desperately to please the trainer.

Always put the current level of response on a variable schedule before raising the criteria: Many trainers make the mistake of assuming one behavior has been learned when in reality the behavior has only been simulated. You can easily be led into believing the dog has learned a certain behavior because they have learned to simulate the behavior. Sooner or later behavior that is being simulated will break down. What appears to be a sudden lapse of memory is not uncommon in dog training. The dog, making steady progress, suddenly seems to have forgotten everything. Although different than problems associated with new environments (new tank syndrome), the results are the same. Fortunately, correcting the problem is equally as simple. Relax, reshape the desired behavior and in a few sessions your dog will be making steady progress once again. A good trainer will recognize the phenomenon, and rather than being discouraged and frustrated will be encouraged. For it is at this moment in the training that the dog is actually beginning for the first time to actually learn the behavior. Until now the dog has simply been simulating the behavior.

Only if the behavior is on a schedule of variable reinforcement can you be certain the behavior is indeed learned and under stimulus control.

Plan your shaping program: There is an old adage that goes "failing to plan is planning to fail." Planning is a dynamic process. The plan itself is only a snapshot of any process at any given time. A larger number of "what if" scenarios should be considered to handle each situation. Stay ahead of your dog. If your dog makes sudden progress be prepared recognize appropriate behavior as well as inappropriate behavior. Reinforce the appropriate behavior. Often dogs either by accident or even with some forethought put together the next progressive step. Its important in the learning process and to the ultimate stabilization of the desired response that the dog is reinforced upon selection of the desired response. Furthermore, unless prepared you might not know what and when to reinforce. Furthermore, you will not accidentally reinforce the wrong response. There is no cookbook when shaping behavior and building behavior chains. It requires close observation and analysis of ongoing behavior, and a mental picture of the desired behavior.

If one shaping technique is not working, try another: If the behavior is not being shaped in a reasonable length of time, review your shaping process and reinforcement to ensure you are not violating one or more of the essential laws. If not, review the path you have selected to shape the desired behavior. The shaping process can take many directions, with as many ways to shape a given behavior as there are trainers to train it.

Avoid interrupting a training session: It is important in the learning process that successive approximations are successive. Each time a training session is interrupted you lose momentum.

If the behavior breaks down, repeat the shaping process. There are any number of things that can cause learned behavior to break down. By simply progressing through each successive step used to shape a given behavior you can again establish previously learned behavior. If a problem develops in a tournament or even in practice, progressing through each successive step will identify the problem and the solution, often in a matter of minutes. In a tournament you might even correct a problem between heats.

End each exercise on a positive note: Ending on a positive note is twofold. First, if your dog becomes confused, stressed or even tired, back up to a point that your dog is confident and success is assured. Remember, as a trainer you do not have to win all the time. Although goals for each training session are important, meeting those goals might hinder training. Your most important goals should always be developing a positive attitude, respect and a bond. Second, by applying the principles of backward chaining, even with simple behaviors, we always end on a positive note. Chapter 9 will focus on the process of backward chaining and why it is a preferred learning process.

When introducing new criteria, temporarily relax old criteria: Until the dog gets used to the new criteria, learned behavior may appear to be forgotten. The behavior is not forgotten. The dog faced with the new criteria must focus on the new criteria.

This is perhaps most apparent with the jumping. Having learned to single step over the jumps set eight inches high and ten feet apart, the dog must learn to single step when the jumps are raised to ten inches. To make the transition easier we often move the jumps in a little closer the first few times that the jumps are raised. Gradually, we can move the jumps until they are again ten feet apart.

Raising the Criteria

A variable schedule of reinforcement has been found to be more effective in maintaining learned behavior than a continuous schedule of reinforcement. Variable (predictably unpredictable or intermittent) reinforcements produce more durable behavior.

The bridge between continuous reinforcement and variable reinforcement is a schedule of selective reinforcement. You shape behavior with selective reinforcement. Selective reinforcement is a process by which you reinforce only your dog's best efforts. Furthermore, by using selective reinforcement you can raise the criteria. The dogs that learn if they don't first succeed in getting a reinforcement, they can try, try again. Therefore, without discouraging them, you can let them go without reinforcement for the wrong response and selectively reinforce just the responses you like. The dog trying to overcome some obstacle will try several responses until ultimately it finds one that gets it the reinforcement. For this reason it essential that once the behavior is established the dog is placed on a schedule of variable reinforcement. Otherwise the dog will become discouraged when it does not get the expected reinforcement. Raising the criteria with selective reinforcement will be next to impossible.

As your dog more consistently performs the desired behavior your selective reinforcement begins to more closely resemble a schedule of continuous reinforcement - every effort is worthy of reinforcement. At this stage the trainer can then maintain the behavior with a schedule of variable reinforcement or further shapes the behavior by again raising the criteria using a schedule of selective reinforcement.

To establish the final desired behavior use a loop process of selective reinforcement to raise the criteria in an incremental step and a schedule of variable reinforcement to maintain the behavior at the new level of performance. Then, again using selective reinforcement, raise the criteria in an incremental step to the next higher level of performance. Maintain the behavior using variable reinforcement. This process of raising the criteria, maintaining the response at the higher level, then repeating the process over and over until the final desired behavior is reached is a loop - more precisely a selective-variable reinforcement loop. Much like a computer program with a loop that starts at x = 1 and ends at x = 10, a selective-variable reinforcement loop starts where the dog simulates some behavior that is similar to or will lead to the final desired behavior. This is illustrated in Diagram 6-1.

It should be noted that getting used to new criteria will often interfere with previously learned behavior. As trainers we must relax the criteria for previous learned behavior when introducing new behaviors. For example, if the dog is comfortable with the jumps ten feet apart, when we raise the jumps for the first time it might be necessary to move the jumps in slightly. Once the dog is comfortable at the new height, we can gradually move the jumps until they are again ten feet apart.

Diagram 6-1 Operant Conditioning

Chapter 7 - Exercises

---

Table 7-1 Categories

Informal Exercises	Formal Exercises
Interaction Restrained Recall/Chase Play Running Hide and Seek Conditioned Reinforcement Target Training (Push) Catch Circle Right and Left Socializing Play Retrieve Follow Me (Puppies) Take Hold Give Grooming Name Attention Hand and Feet Games Crate Training Chase (Object) Comfort Zone Restrained Outrun	Jump Sit and Release Stay Command Stand and Release Retrieve (long line) Push/Catch Turn and Chase Down and Release Back Up and Call Your Dog
Foundation Exercises	Advanced Exercises
Long Line Recalls About Turn Stand for Exam Directed Jumping Trigger the box Retrieve (introduce a correction)	Quick Turns Out Run Passing Added Distractions Return Run Racing Side by Side Building the Behavior Chain

---

Unfortunately the exercises do not follow a simple straight-forward sequence. Rather, each exercise is related in a complex training process best illustrated in Diagram 7-1. However, to simplify the training process we can focus our attention to four (4) major events or behaviors indicated on Diagram 7-1 with a diamond - the retrieve, the outrun, the return run and triggering the box. All other exercises illustrated in Diagram 7-1 are merely shaping exercises in the training process or exercises designed to prepare the dog for actual competition.

That is not to say that these exercises are any less important. The reason we teach each exercise and work on each exercise regularly is to establish the behavior independent of other behaviors. In this manner problems either at a tournament or at practice can be corrected without setting up the jumps, backstops and boxes.

Furthermore, Diagram 7-1 illustrates a training process from beginning to end. In reality, the trainer will never finish the training. Always work on the foundation established during the training process. A typical practice that might routinely include a number of the exercises is illustrated in Table 7-1.

Also, Diagram 7-1 places the exercises in various categories. The student should realize that some dogs may progress very rapidly at one or more exercises and less rapidly at other exercises. Another dog may progress at an entirely different rate. You will find some dogs working on advance exercises, and at the same time, having difficulty with what may seem very simple and basic exercises to other dogs. Often the training process may be in a holding pattern, waiting until one or more exercises can be added to the process before proceeding on.

Diagram 7-1 offers an insight to the training process and will identify those behaviors that need to be established before you can expect to proceed to the next step in the training process. Also, if you are having difficulty with one or more of the exercises, Diagram 7-1 will help to identify where you have gone wrong. Perhaps you have taken part of the exercise for granted or even overlooked one or more steps.

Exercises typically taught in the beginning Flyball class are indicated with a single line box. Exercises taught in the advanced class are indicated with a double line box.

Informal Exercises

Socializing. Expose your dog to different surroundings. Visit nursing homes, hospitals, fun matches, fairs and shopping centers. Initially, do not put added stress on the dog by asking it to perform. Allow your dog to adapt to the new surroundings. When your dog is confident with the new surroundings or if you must introduce a dog to a new environment, refresh its training with easily reinforced behavior shaping exercises.

Regardless of how well a dog has learned a behavior, the first time it is placed in a new environment the dog will temporarily forget all previously learned behavior. This is another reason socializing your dog is so important. When in an unfamiliar or new environment reshape your dogs behavior using the progressive steps used to initially train your dog. When entering a tournament with new dogs, I prefer to introduce the dogs using a technique known as backward chaining. Backward chaining is discussed in great detail in Chapter 8.

Follow Me. This exercise is primarily for puppies from 8 to 12 weeks old. The exercise combines the "Come" with taking your puppy for a walk. For this exercise, take your puppy on a walk without a leash. The area must be free of distractions, preferably with no other people around. Most two and three month old puppies won't want to get too far away (wait until the dog is 6 or 7 months old and this won't necessarily be the case). Go for a walk together and when the puppy gets about twenty feet away, get the puppy's attention by calling its name and clapping your hands. When it looks at you, kneel down and call it to you. Give it lots of praise and continue your walk. Occasionally walk away from your puppy encouraging your puppy to follow you. This will establish yourself as a pack leader; you are teaching the puppy the command come and establishing a bond between you and your puppy. Equally important, you are establishing a comfort zone for your puppy. The puppy that is unsure at this age will feel uncomfortable twenty or twenty-five feet away; near you the dog will be comfortable and secure. As the puppy gets older the instinct to stay near you has been established.

Comfort Zone. I like to establish a comfort zone with a puppy. One way is to teach the puppy that when outside a certain distance from the handler, say four to five feet it can expect some discomfort. If the puppy does not immediately respond to the handlers command "Come" (when outside that comfort zone), the puppy will get a light but firm correction on the rear end. If the puppy is within the comfort zone it can expect nothing but fun and pleasant experiences. Although establishing a comfort zone for an older dog can be done, it is more difficult than with a puppy of 8 or 9 weeks. Puppies are unsure of themselves at this age and respond very well to establishing a comfort zone. Furthermore, once the puppy is 10 to 12 weeks old it might be difficult to reach down and give it a light but firm correction.

As the puppy matures the comfort zone will automatically expand; however, that security of being near the handler (and the insecurity of being too far away) has been established and will be difficult to break. It is possible and even desirable to be able to expand that comfort zone naturally, by allowing the dog more freedom with age. A simple "No" can be used to establish a new boundary for that comfort zone, provided the command "No" predicts a negative reinforcement. Certainly, for Flyball the comfort zone should ultimately be expanded to 75 or 80 feet.

Restrained Recall/Chase. Have a friend restrain your dog. Give the cue, "Ready" or "On Your Mark," release the dog with the command "Come". The handler then turns and runs in the opposite direction, away from the dog. Later, we will use the restrained recall over one jump, then two, three and so on. The outrun is similar to a restrained recall, only with the outrun the dog is released by the handler to chase a ball. Later, we will use the restrained outrun over one jump, then two, three and so on.

Take Hold Give. Play retrieve is a great way to introduce your dog to the retrieve. However, a more formal command "Take, Hold and Give" should be introduced. The command "Take" must predict positive reinforcement. Initially, it should be paired with a treat. Next you should have your dog "Give" the article. Later, we can introduce the hold command, initially working on the command "Hold" for only for one or two seconds. Again, the command "Hold" and "Give" must also predict positive reinforcement.

Some dogs won't "Take" a tennis ball, others will not "Give" it back. Therefore, use a wood dowel, or some other article that the dog will be willing to take, hold and give with out a fight.

Play Running. Play run with your dog with and without jumps.

Grooming. Grooming is not only important for the dog's health and well-being, but important in establishing a bond as well.

Hide and Seek. Have a family member restrain your dog while you go into another room. Call your dog and praise it when it finds you. Again, this establishes a bond and a comfort zone.

Name Attention. Have your dog on a short leash or standing next to you. Each time it is distracted and looks away, call its name. The dog will look at you. When it does, reinforce its response with a piece of food.

Conditioned Reinforcement. The word "Ready" should signal to your dog reinforcement. "Ready" should predict reinforcement. As with the name attention exercise above, have your dog on a short leash or standing next to you. Each time it is distracted and looks away, call out "Ready." The dog will look at you. When it does, reinforce its response with a piece of food. The word "Ready" will be paired with a treat until the word "Ready" becomes a conditioned reinforcer.

"On Your Mark" should signal to your dog that it is going to be released to chase or retrieve an article. Initially, restrain your dog, give the command "On Your Mark...Get Set" and release your dog on the command "Go" to chase a ball, another member of the family or perhaps to retrieve a water dummy. I often will restrain my dog physically or on a sit stay, toss a water dummy off shore, give the signal "On Your Mark ...Get Set" and release the dog on the command "Go." Just as the dog reaches the dummy I signal the dog "Ready" to strengthen the ability of the signal "Ready" to reinforce the dog. A Whistle or the sound of a box being loaded can be substituted for the signal go, thus establishing each as a conditioned reinforcement.

The command "Leave It" and "No" must predict negative reinforcement before using the command in a training situation. Initially, with no distractions and your dog sitting at your side, pair the command "No" and later, "Leave It" with a collar correction. After one or two sessions, the commands "Leave It" and "No" will predict negative reinforcement and can then be used in training situations to give the dog a correction, one that the dog cannot avoid. The dog will likely never need another collar correction, other than on occasion to remind the dog that "No" and "Leave It" do indeed predict negative reinforcement.

Circle Right. Using the signal "Ready...Ready ...Ready," circle right. This is the beginning of establishing a quick turn.

Hand and Feet Games. Hand and feet games are important to establish that they are not something to be afraid of. Playfully pushing your dog away and then enticing it to come back with a piece of food will teach the dog the comfort zone is next to the handler. The same can be done with your feet, playfully pushing the dog away and enticing it to return with a piece of food.

Target Training. Similar to the "hand shake" encourage your dog to push your hand and reinforce the behavior with a piece of food. Later, we will transfer the target (our hand) to the Flyball box pedal.

Catch. Either toss the ball or hand the dog the ball pairing the command "Catch" with the exercise.

Crate Training. It must be stressed that a dog crate is not intended for long-hour durations for the convenience of an absent owner. If correctly used, a crate can have many advantages for both you and your dog. With a crate, among many other benefits, you can enjoy complete peace of mind when leaving your dog alone at a tournament to enjoy the privacy and security of a den. You can effectively confine your dog when it is over excited or bothered by too much confusion, knowing that your dog will be comfortable and safe. You can travel with your dog without the risk of the driver being dangerously distracted or the dog getting loose and hopelessly lost. Finally, the dog can easily adapt to the strange surroundings of a motel or a tournament site as long as the dog can retreat to a familiar secure den.

The crate must be large enough to permit the dog to stretch out. The most practical dog crate is the wire mesh type, available in a wide variety of sizes. Light-weight and easily handled, it allows total ventilation and permits the dog to look around or the crate can be easily covered for privacy.

A young puppy should have no problem accepting a crate as its own. Whether for a puppy or an older dog, initially place the crate in an area where there are people. For bedding, use an old towel or a piece of blanket that can be washed, and a freshly worn unlaundered article of your clothing such as a tee shirt, an old shirt or sweater (this will help in the bonding and make the dog feel more at home).

Establish a crate routine, closing the puppy in at regular 1 to 2 hour intervals during the day (normal nap times are a good opportunity to establish a routine). Be sure to remove any collars and tags which could get caught in an opening and strangle the dog or puppy. A dog may be crated all night, but, it is unlikely that a puppy will make it through the night without relieving in the crate. Until housebroken, puppies should be taken out during the night at least once.

If you are crating an adult dog for the first time, the crate must be introduced gradually, with every possible effort made to be sure that the dog's first association with it is pleasant. Place the crate in a location where the dog will feel part of the family, encourage the dog to investigate the crate, luring it in with treats - praising it enthusiastically. When it begins to enter the crate confidently, place its bedding and something of yours or a towel you have slept with inside and coax your dog to lie down and relax. Continue this pattern for several days - shutting the door only briefly while you sit beside the crate. As soon as you feel confident that the dog will remain quietly, you may safely leave it alone for extended periods of time - perhaps 1/2 to 1 hour.

Play Retrieve. One of the first steps is to test and develop the young dog's instincts to chase and retrieve. Find a long hallway (with all the doors closed), an ideal spot to work on retrieving since there are no distractions and nowhere to go. The object need not be a ball. Crumpled paper, a work glove, a squeaky toy or an old sock may be easier for a puppy to pick up. Later add a command such as "Take It."

Praise the young dog while it actually has the object in its mouth and don't be in a hurry to take it from the young dog. Allow the dog possession time. The dog deserves and needs a little time to play with the ball. After sufficient praise, distract it with a bit of food or even another ball. When the dog drops the ball, go back to training your dog.

Chase. Another aspect of Flyball is to get the dog to return to the handler as quickly as possible. When getting the ball is the primary motivation, the dog has no motivation to return to the handler. The primary motivation for Flyball should always be to quickly return to the handler. The ball should only be a secondary motivation or reinforcement. Running in the opposite direction takes advantage of the dog's natural instincts to chase. It doesn't matter that the dog drops the ball. Restrained recalls are another technique that takes advantage of the dog's natural instincts to chase.

Play games where the dog chases you. Reinforce your dog for chasing you with a game of tug of war. Let your dog win. Hide and seek can also be a great way to reinforce the bond between you and your dog.

Formal Exercises

Jump. Start with regulation Flyball jumps or the regulation obedience broad jumps. McCann vs. McCann Flyball team set more NAFA records from 1980 through 1990 than all the other teams combined. They prefer to use a long board, eight (8) inches in width, with one side elevated slightly. Regulation obedience broad jumps are ideally suited. Being wider than the Flyball jumps they discourage the dog from running around the jump. Also, having no uprights, the broad jumps tend to be less threatening.

One person restrains the dog directly in front of the jump, while the handler, on the other side entices the dog to jump pairing the command "Jump" with the reinforcement and the behavior. Once the dog is jumping readily, a second jump may be added, a third and finally a forth jump can be added. Since the dog is doing a straight line recall, there is less chance that the dog will cut the jumps than if the dog were on lead with the handler running along side. The dogs are also more self-reliant when not led over the jumps. Initially keep the jumps low. Flyball is not a high-jumping contest.

Push/Catch. Earlier we taught the dog to "Push" (hand shake) by enticing your dog to push your hand and reinforcing the behavior with a piece of food. Now, instead of food, play a game of catch combining the command "Push" with the command "Catch." Later, we will transfer the target (our hand) to the Flyball box pedal.

Sit and Release. The command "Sit" should predict positive reinforcement. Using treats as a target, elicit a response to the command "Sit" and reinforce the desired response. The command "Sit" and the reinforcement is paired with the response. Once the dog is responding, the treat is out of sight and is no longer used as a target to elicit a response, but rather simply to reinforce the response initiated by the dog on the command "Sit". Initially you only reinforce the dog's best efforts. Introduce a variable schedule of reinforcement once the behavior has been established. In the early stages be sure to release the dog after only a few moments of sitting.

A common error is to release the dog then give the dog the reinforcement. Thus, the dog is being reinforced for breaking a sit command.

Turn and Chase. With your dog facing you, either sitting or standing, give the command "Ready" and throw the ball in the opposite direction. This exercise is to introduce the quick turn.

Stay Command. Basic obedience is a prerequisite for Flyball. Once competing the dog will be off lead. The handler must be able to control his dog with verbal communication only. The dog must come when called, be steady in the midst of distractions and stay when told. Again with any new exercise start off by teaching the dog that the command "Stay" predicts a positive reinforcement. With the dog in the sit position, give the command "Stay" placing your hand in front of the dog's face. The lead should be on the dead ring, with slight tension upward. Reinforce the dog for staying if only for one or two seconds. Gradually increase the time, using the command "Stay" as a conditioned reinforcer. After the exercise give the dog the primary reinforcement. Here the command "Stay" is a conditioned reinforcer and bridge, until such time that the handler can reinforce the behavior with a primary reinforcement.

Gradually, work farther and farther from your dog, returning to your dog from time to time giving the command "Stay" paired with a piece of food. This will reinforce the conditioned reinforcer "Stay" with a primary reinforcer. If the dog breaks, give a verbal correction, "eh-eh" or "No" and start over.

Down and Release. From the sitting position use food as a target to guide your dog into the down position. Placing your free hand on his back and applying slight pressure. Again be sure to release your dog after only few moments.

Stand and Release. Again, from the sitting position use food as a target to guide your dog into the stand position, placing your free hand under its stomach. Again be sure to release your dog after a only few moments.

Retrieve. Begin to work with a long line and the retrieve. Later, once the dog has retrieved the ball, introduce a correction for the quick turn. For now, the long line is only used to guide the dog back to the handler.

Foundation Exercises

Long line recalls. First, your dog should become accustomed to a long line. With a long line on, start with a simple recall or even a restrained recall.

Later, to establish reliable responses to the command come, we introduce an escape response (correction) for a slow response. Of course, if the dog fails to respond or if all the responses are slow we would work on restrained recalls or perhaps a game of hide and seek until we have the desired response. Corrections designed to teach the dog an avoidance response should not be used to establish a response. With the introduction of an escape response, the dog learns an avoidance response. Again avoidance responses (where the dog learns to avoid the behavior that predicts negative reinforcement by responding quickly) are reinforced. Escape responses are not reinforced, but always followed immediately with another command that gives the dog an opportunity to avoid the behavior that predicts negative reinforcement.

Directed Jumping. With your dog off to one side, guide it over the jump. Initially the dog is set off to one side and the handler is near the jump. Gradually the dog is set off farther and farther away from the jump and more and more to one side. Finally, the handler begins to move farther and farther away from the jump and more and more off to the side.

About Turn. Very similar to the circle right, the about turn will help establish a quick turn. Using "Ready...Ready...Ready..." as a conditioned reinforcement and food as a primary reinforcement, circle right. Keep the food low and the circles tight.

Trigger the box. The dog should be introduced to the box gradually. Leave the box around the house where the dog will become accustomed to it. Put treats on the box or let the dog retrieve a few balls that roll near the box.

Take the dog up to an unloaded box. The idea is to entice the dog to approach the box and have a pleasant experience. Once the dog has learned to target yourhand with its paw for a treat, we transfer the target from the hand to the pedal by merely placing our hand on the pedal with the palm facing up. We can either give the dog a treat for targeting our hand (thus the pedal) or toss the dog a ball, combining the "push......catch" command we learned earlier. Later, with the dog on a short lead the handler restrains the dog while the loader targets the box with his hand. Initially, do not load the box. Focus on getting the dog to press the pedal and catch a ball tossed by the box loader. With some boxes the box loader can set the ball in the cup without loading the box. Finally, the box loader can load the box.

Most dogs learn to press the pedal without use of a target. However, those dogs may have problems learning to use a new box. Also, dogs that have been taught to run up to the box and slam into the box will have problems turning until they learn to press the pedal.

To encourage the dog to trigger the box have one person restrain the dog a short distance from the box. The ball is then tossed to the box loader, who catches the ball and immediately loads the box. Tossing the ball to the box loader heightens the dog's excitement. Once the dog is reliable on the box the handler should gradually back the dog away farther and farther from the box. Once the dog catches the ball reliably, the handler turns and takes off running in the opposite direction, enticing the dog in a game of chase. Again, don't be concerned if the dog initially drops the ball.

Eventually combine the "turn and chase" with the box. As the dog triggers the box and catches the ball, the box loader will give the command "Ready" and throw another ball in the opposite direction.

Stand for Exam. Once your dog is confident with the stand and stay command, add a distraction by having the judge examine and measure your dog. Initially, have the judge just pass his hand over the dog.

Retrieve. After you have worked for some time with a long line and the retrieve, introduce a correction for the quick turn. Initially, with the dog on a short lead (6 foot lead) pair the command Come with a correction as you back up and call your dog. If the dog responds as it should, the dog will not get a correction. The handler should be quick to reinforce the dog for the proper response; otherwise, correct the dog. After the dog has learned to avoid the behavior that causes the correction, namely responding quickly to the command "Come," toss a ball a few feet from the dog and have the dog retrieve the ball or "Take It." Once the dog has the ball give the command to "Come." Again, if the dog responds with a quick turn, the dog will not get a correction; rather, the handler should be quick to reinforce the dog. Once both the handler and dog are confident on a 6 foot lead, try it with a long line.

This exercise is to be taught independent of Flyball. The dog is to associate any correction with a slow turn, not the box, or the ball. Furthermore, by teaching the exercise independent of Flyball, a problem with quick turns in a tournament can be corrected before going into the ring without having a box and a set of jumps.

Advanced Exercises

Quick turns. Quick turns don't just happen, they must be shaped. Long line recalls (with a correction) are important in teaching quick turns, as are turn and chase, back up and call your dog, quarter turns, about turns and circle right. Once satisfied with straight recalls we can further shape the desired response - a quick turn. Turn the dog a quarter turn, leave your dog and go about 30 or 40 feet away. Call your dog. Each time you call your dog, turn your dog in small increments until the dog's back is facing you.

Satisfied that the dog is responding correctly with a quick turn, we introduce the correction. If the dog fails to respond with a quick turn, correct the dog. The dog's natural response will be to escape the correction. If the dog's response is to avoid the behavior that predicts negative reinforcement, the avoidance response is reinforced with a game of chase, or even a piece of food.

Up to this point the cue for a quick turn is the command "Come." The final stage of shaping quick turns is to teach the dog a new cue for the quick turn, namely the ball. This is done by introducing a long line retrieve. Once the dog has retrieved the ball (the new cue for a quick turn) we give the command to "Come." The dog, to avoid a correction and to be reinforced for a quick turn, will turn quickly. In time the command "Come" can be dropped and the ball becomes the cue for a quick turn. Again, when introducing escape responses the trainer must maintain an emotional balance between the desire to avoid certain behavior and the desire to be reinforced. It is a delicate balance and sometimes difficult to achieve.

Timing is critical. Before the correction is given, the dog must have time to process the command "Come" and avoid the behavior that to predicts a negative reinforcement; namely, anything but a quick turn.

Return Run. The return run is simply a restrained recall where the dog races over four jumps and chases the handler or in some cases a ball if the handler is unable to run away from the dog. The exercise is taught using backward chaining as discussed in Chapter 9. Again, the jumps need not be regulation Flyball jumps. Some trainers prefer regulation obedience broad jumps. They are excellent for the learning stages, and easier to set up, move around and store.

Outrun. The outrun is a game of chase with jumps. The dog merely chases a ball thrown by the handler. Use a tennis racket, if you can not throw a tennis ball a good distance accurately. Once the dog has been released the handler should go to the dog. Do not insist on a retrieve, as we can only shape one thing at a time.

Combine the Outrun and Return Run. Later, by combining the outrun and return run we have a retrieve with jumps but no box.

Racing side by side. Do the return run or the outrun with two dogs racing side by side.

Passing. Passing is discussed in Chapter 9.

Added Distractions. Ideally, the dog should not be introduced to distractions in the learning stages. This is not always practical or even possible. However, until the dog has learned the behavior it is not fair to correct the dog. Further, unless absolutely necessary don't put your dog in a situation where distractions are likely. Again, before you introduce distractions the command "Leave It" and "No" must predict negative reinforcement.

Ultimately, using a distraction to proof the dog is a good idea, but not until the dog has learned the exercise. Using distractions to teach the exercise is a bad idea.

In competition there will be times when another dog may drop a ball or your dog may have to retrieve a missed ball. In Flyball anything can happen and often does. In Louisville a few years back, I remember a dog coming over to our box during a race. Our third dog, Trey, had to jump over the dog to get to the pedal (and did). On the return run Trey again jumped over the dog. On another run, our fourth dog, Wendy, managed to drag another dog over four jumps with the dog's teeth firmly planted in Wendy's hind quarters. Certainly that was unusual, and similar distractions to proof the dog are not necessary. However, some distractions that the dog should be introduced to include loose tennis balls, racing side by side, switching lanes, dogs and handlers crossing the lanes (as judges and loose dogs sometimes do), barking dogs, knocked down jumps and crowd noise. Hopefully, the dog will be somewhat accustomed to these distractions as a part of interacting with other dogs, their training, socializing and preparation for new or unfamiliar environments (see new tank syndrome, Chapter 1).

Building the Behavior Chain. Chaining is discussed in the next Chapter.

     Dog/Handler  	I	I	I	I	Ball


		  	I	I	I	I	Ball
     Dog/Handler 	
	
		  	I	I	I	I	Ball

     Dog/Handler			  
				

	Handler 		I	  	I		I		I	Dog/Helper	
						   

	Handler		  	I		I		I		I	Dog/Helper
	 	
						   
	Handler		  	I		I		I		I	Dog/Helper

	Handler 1						Dog 1/Helper

	I	  	I		I		I		
	
	Dog 2/Helper						Handler 2

Handler 1 and 3					    Dog 1 and 3/Helper
	I	  	I		I		I		
	
Dog 2 and 4/Helper                                  Handler 2 and 4

	Handler 1 and 3					    Dog 1 and 3/Helper


				I	  	I		I		I		
	

	Dog 2 and 4/Helper						Handler 2 and 4						 Step 1




	Handler 1 and 3					    Dog 1 and 3/Helper

				I	  	I		I		I		
	
	Dog 2 and 4/Helper						Handler 2 and 4						Step 2




	Handler 1 and 3					    Dog 1 and 3/Helper
				I	  	I		I		I		
	Dog 2 and 4/Helper						Handler 2 and 4						Step 3

Date ___________  Place _______________________ Page  ____  of ____

Comments___________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________


First Dog    	Second Dog   	Third Dog  	Fourth Dog	  Team   
Time  Notes	Time  Notes	Time  Notes	Time  Notes	  Total

____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____


   EP - Early Pass   MB - Missed Ball	  	CJ - Cuts Jump	
   FS - False Start  OS - Over Shoots Box  	KJ - Knocks Down Jump
   LS - Late Start   HB - Hesitates at Box  	NR - No Retrieve	
   LP - Late Pass    WT - Wide/Slow Turn  	DB - Drops Ball
   SL - Slow/Double  TB - Fails to Trigger 	IN - Interferes
	     Steps	  the Box 		OT - Other

Date ___________  Place _______________________ Page  ____  of ____

Comments___________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________


First Dog    	Second Dog   	Third Dog  	Fourth Dog	  Team   
Time  Notes	Time  Notes	Time  Notes	Time  Notes	  Total

____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____
____  ____     ____  ____     ____  ____     ____  ____       _____


   EP - Early Pass   MB - Missed Ball	  	CJ - Cuts Jump	
   FS - False Start  OS - Over Shoots Box  	KJ - Knocks Down Jump
   LS - Late Start   HB - Hesitates at Box  	NR - No Retrieve	
   LP - Late Pass	 WT - Wide/Slow Turn  	DB - Drops Ball
   SL - Slow/Double	 TB - Fails to Trigger 	IN - Interferes
	     Steps		   the Box 		OT - Other

	
    Dog's  Name         No of  Runs     %    Rank      Avg.   Rank
	Runs   _____  			    Time ____

 1 __________________   _____  _____  _____  _____     _____  _____
 2 __________________   _____  _____  _____  _____     _____  _____
 3 __________________   _____  _____  _____  _____     _____  _____
 4 __________________   _____  _____  _____  _____     _____  _____
 5 __________________   _____  _____  _____  _____     _____  _____
 6 __________________   _____  _____  _____  _____     _____  _____
 7 __________________   _____  _____  _____  _____     _____  _____
 8 __________________   _____  _____  _____  _____     _____  _____
 9 __________________   _____  _____  _____  _____     _____  _____
10 __________________   _____  _____  _____  _____     _____  _____
11 __________________   _____  _____  _____  _____     _____  _____
12 __________________   _____  _____  _____  _____     _____  _____
13 __________________   _____  _____  _____  _____     _____  _____
14 __________________   _____  _____  _____  _____     _____  _____
15 __________________   _____  _____  _____  _____     _____  _____
16 __________________   _____  _____  _____  _____     _____  _____
17 __________________   _____  _____  _____  _____     _____  _____
18 __________________   _____  _____  _____  _____     _____  _____
19 __________________   _____  _____  _____  _____     _____  _____
20 __________________   _____  _____  _____  _____     _____  _____
21 __________________   _____  _____  _____  _____     _____  _____
22 __________________   _____  _____  _____  _____     _____  _____
23 __________________   _____  _____  _____  _____     _____  _____
24 __________________   _____  _____  _____  _____     _____  _____

					Team statistics

Team1 Team2 Team3 Team4

_____ _____ _____ _____ Runs < _____ seconds/total number of runs
_____ _____ _____ _____ % (Number of runs/total number of runs)
_____ _____ _____ _____ Average time (runs under ______ seconds) 
_____ _____ _____ _____ early passes  
_____ _____ _____ _____ missed balls
_____ _____ _____ _____ times the jumps were cut 	
_____ _____ _____ _____ false starts
_____ _____ _____ _____ times a dog over shot the box
_____ _____ _____ _____ times a jump was knocked down
_____ _____ _____ _____ late starts
_____ _____ _____ _____ times a dog hesitates at the box
_____ _____ _____ _____ times a dog fails to retrieve
_____ _____ _____ _____ late passes	
_____ _____ _____ _____ wide or slow turns
_____ _____ _____ _____ times a dog dropped the ball
_____ _____ _____ _____ times dog was slow/double stepped 
_____ _____ _____ _____ times dog fails to trigger the box
_____ _____ _____ _____ times a dog interfered with the other team
_____ _____ _____ _____ Other __________________________________

	
    Dog's Name		    No of  Runs     %    Rank      Avg.   Rank
		 		    Runs   <____  			     Time
                                                       <____

 1 __________________   _____  _____  _____  _____     _____  _____
 2 __________________   _____  _____  _____  _____     _____  _____
 3 __________________   _____  _____  _____  _____     _____  _____
 4 __________________   _____  _____  _____  _____     _____  _____
 5 __________________   _____  _____  _____  _____     _____  _____
 6 __________________   _____  _____  _____  _____     _____  _____
 7 __________________   _____  _____  _____  _____     _____  _____
 8 __________________   _____  _____  _____  _____     _____  _____
 9 __________________   _____  _____  _____  _____     _____  _____
10 __________________   _____  _____  _____  _____     _____  _____
11 __________________   _____  _____  _____  _____     _____  _____
12 __________________   _____  _____  _____  _____     _____  _____
13 __________________   _____  _____  _____  _____     _____  _____
14 __________________   _____  _____  _____  _____     _____  _____
15 __________________   _____  _____  _____  _____     _____  _____
16 __________________   _____  _____  _____  _____     _____  _____
17 __________________   _____  _____  _____  _____     _____  _____
18 __________________   _____  _____  _____  _____     _____  _____
19 __________________   _____  _____  _____  _____     _____  _____
20 __________________   _____  _____  _____  _____     _____  _____
21 __________________   _____  _____  _____  _____     _____  _____
22 __________________   _____  _____  _____  _____     _____  _____
23 __________________   _____  _____  _____  _____     _____  _____
24 __________________   _____  _____  _____  _____     _____  _____

					Team statistics

Team1 Team2 Team3 Team4

_____ _____ _____ _____ Runs < _____ seconds/total number of runs
_____ _____ _____ _____ % (Number of runs/total number of runs)
_____ _____ _____ _____ Average time (runs under ______ seconds) 
_____ _____ _____ _____ early passes  
_____ _____ _____ _____ missed balls
_____ _____ _____ _____ times the jumps were cut 	
_____ _____ _____ _____ false starts
_____ _____ _____ __