Why does my dog respect the fence except when a squirrel runs by?

Prey drive can override learned behavior. Work the run-through ladder in order before adding training: fresh battery, contact fit, boundary width at the recommended 12-24 feet total, correction level, and only then staged trigger sessions one level above baseline.

Will raising the correction level hurt my dog?

The static is a brief, low-energy pulse designed to interrupt, not injure. Fix contact fit before raising anything, because a blocked undercoat is the most common false reason to turn the dial up. A yelp or panic means the level is too high. Keep wear under 12 hours a day and reposition the collar every 1-2 hours.

My dog goes through the fence chasing something and won't come back. What do I do?

Never call them back across an active boundary. The collar can fire as they cross, and some systems correct on re-entry. Power off the transmitter or remove the collar, leash them, lead them back in, and reward the return. Restart at controlled-trigger sessions, and if it keeps happening run the run-through branch of our online fence troubleshooter.

Are some breeds harder to chase-proof?

Yes. Sighthounds, terriers, herding breeds, and hunting breeds carry stronger drive. Add 1-2 weeks for high-drive breeds, and never assume a sighthound is fully chase-proofed.

What if my dog still runs through at the maximum level?

Two run-throughs at maximum level with verified fit means this system is not right for this dog. Discontinue it and return to supervised or on-leash yard time while you consult a qualified positive-reinforcement trainer; some dogs' drive outruns any static correction.

How to Stop a Dog Running Through the Fence to Chase

A dog in full chase crosses a narrow correction field in a fraction of a second, and 44% of dogs using electronic containment escaped at least once in their lives in a 2017 JAVMA owner survey. If your dog holds the boundary on quiet afternoons but blasts through when a squirrel appears, that is a chase-proofing gap, not a training failure. This guide addresses it over 3-5 weeks of staged work; add 1-2 weeks for high-drive breeds. Vendors advertise "99%+ containment" without publishing comparable methodology, so plan conservatively instead.

Drive is faster than the fence. Training is what holds. A dog in full chase crosses the correction field in a fraction of a second, and a correction too weak to interrupt drive teaches your dog it's survivable: that's how run-throughs are learned. The honest baseline is the JAVMA owner survey, where 44 percent of electronically fenced dogs escaped at least once. Chase-proofing closes the gap with 3-5 weeks of staged triggers, building the skill the first fence training never taught: stopping at the line with a squirrel already running. Before any of that, work the hardware ladder: battery, contact fit, boundary width, then level.

Why Standard Training Isn't Enough#

Holding a boundary takes two separate skills:

Knowledge: flags + tone = stop. Boundary training delivers this.
Inhibition under drive: resisting prey, chase, or social drive at full activation. Chase-proofing builds this.

Skipping the second is why dogs who "knew" the fence still end up at the neighbor's house.

Rule Out Hardware Before Adding Training#

Run-throughs have a cause ladder. Work it in order, because most "training problems" die on the first three rungs (see the run-through troubleshooting guide for the full diagnostic):

Dead or weak battery. Failure is silent. Test the collar weekly with the test light at dog-neck height; swap batteries on a fixed 3-month schedule. The collar troubleshooting guide has the full battery check.
Poor skin contact. One-finger fit, contacts touching skin. On thick or double coats, part the hair or fit longer contact points. A blocked undercoat is the most common false reason to raise the level. Repositioning every 1-2 hours and keeping wear under 12 hours a day also prevents collar sores.
Boundary width too narrow. Manufacturers recommend 12-24 feet total width; a sprinting dog crosses anything narrower in a fraction of a second, taking one brief pulse on the way out. Widen the field, then re-walk the boundary with the collar and re-plant flags every 10 feet at the new beep line. The width setting and testing guide covers calibration.
Correction below arousal level. Fixed by the escalation rules in Step 4 below.
Incomplete training. Fixed by the plan itself.

One mechanism to understand before you touch the dial: a correction too low to interrupt drive teaches your dog that the pulse is survivable. After a learned run-through, the fix is full retraining plus a level increase, never just a turn of the dial.

Before Session One#

Before-Session-One Checklist

0 of 6 items complete

Boundary training complete; dog has held the fence in calm conditions for 1 week

A 30-50 ft long line (our store) (never chase-proof off leash)

Contact fit verified (one-finger test); spare contact points (our store) on hand, longer tips fitted for thick coats

Fresh battery installed and collar tested with the test light

Boundary width set to 12-24 ft total; flags standing every 10 ft at the beep line

Helper recruited and trigger props matched to your dog's trigger

The Plan#

Click steps to mark them complete

0 / 6 completed

Confirm Baseline Boundary Training Is Solid

Before day 1

Your dog must have completed all five phases of boundary training and held the boundary in calm conditions for at least one week. A dog still testing the fence on quiet days needs basic training fixed first. Chase-proofing layers on top of it, never replaces it.

Identify the Trigger

Days 1-3

Watch your dog at the boundary and note exactly what breaks them:

Prey: squirrels, rabbits, deer, cats
Movement: cars, joggers, kids on bikes
Social: other dogs, people walking by, the mail carrier

The training sequence is the same for every trigger. Only the staged trigger changes.

Stage Controlled Triggers

Days 4-10

With your dog inside the boundary on the long line, have a helper present the trigger outside the fence at low intensity: a ball rolled slowly, a walking jogger, a leashed dog at distance. Reward every hold.

Run three 10-15 minute sessions a day, the cadence the major manufacturers publish. Raise trigger intensity one notch only after the dog ignores the current level for 3 consecutive sessions.

Raise the Correction One Level, With Rules

During days 4-14

High-drive dogs need a stronger interrupt while drive is activated. Escalate by decision rule, not by frustration:

Start one level above baseline.
Go to a second level up only after re-checking contact fit and 2 failed sessions at the first level.
Never go beyond that without pausing the program, re-running the hardware ladder, and reading the stop rule below.
Return to baseline once the dog holds reliably.

Before you turn the dial, read what the correction actually feels like.

Read Your Dog

The right level produces subtle recognition: ears up, head tilt. A yelp or panic means it is too high, so drop back one step and re-check fit before the next session.

Real-World Exposure

Days 11-14

Once staged triggers are ignored reliably, supervise your dog through real exposures (wild squirrels, passing cars, dogs on the sidewalk) while they drag the long line. Stay close and reward every hold.

Off-Line With Backup

Weeks 3-5

Drop the long line but stay in the yard, ready to redirect calmly if you see a break building. After 2 weeks of clean off-line exposure you can step back inside, then spot-check monthly for the first season.

If They Break Through#

Never Call Your Dog Back Across an Active Boundary

The collar can fire as they cross, turning your recall cue into an unpredictable correction. Research on e-collar stress (Schalke 2007) found that unpredictable, poorly timed corrections elevate cortisol, and that association creates fear and boundary refusal.

Do this instead:

Power off the transmitter or remove the collar.
Walk out, leash your dog, and lead them back inside the boundary.
Reward the return. Never punish it, no matter how long the chase lasted.
Restart the plan at Step 3 (controlled triggers) before any unsupervised yard time.

Know your system's re-entry behavior: many correct the dog on the way back in, stranding them outside the fence after a run-through. Outbound-only correction ("run-through protection") is a purchase spec worth demanding when you buy or upgrade. If run-throughs keep happening, work the run-through branch of the fence troubleshooter.

When to Stop#

The Stop Rule

If your dog runs through twice at maximum level with verified fit (fresh battery, one-finger contact, 12-24 ft width), this system is not for this dog. Discontinue it and use supervised or on-leash yard time while you consult a qualified positive-reinforcement trainer. Some dogs' drive outruns any static correction, and repeating a failed correction only erodes welfare and trust.

Two limits of chase-proofing itself:

Sighthounds and other high-prey dogs fail often enough that sighthound rescues categorically reject electronic fences. If that is your dog, read whether an underground fence is right for your dog before investing more sessions.
The fence keeps nothing out. Loose dogs, cats, and wildlife can walk into your yard and become the trigger, and chase-proofing cannot fully solve a provocation that comes to your dog. Supervision remains part of the system.

Breed Notes#

Add 1-2 weeks to the plan for high-drive breeds.

Sighthounds (Greyhounds, Whippets, Salukis): assume permanent supervision near the boundary. Even excellent training may not override sight-prey response.
Terriers: willing to take the correction; expect to use the full escalation rule.
Herding breeds (Border Collies, Heelers): movement-triggered. Stage with bikes, joggers, and cars.
Hunting breeds (Labs, Goldens, Pointers): biddable but scent-locked. Stage with scent trails.

Need More Help?#

Check the FAQ for common questions, or send us a message. We reply within a business day.

Chase-Proofing: Stop Boundary Run-Throughs Caused by Squirrels, Cars, and Other Dogs

Tools Required

Materials Required