Push or Pull Stick Welding: Which Technique is Right and Why

I still remember the first coupon I ran in my welding class. The instructor dropped a plate in front of me, nodded at the machine, and said, “Lay a 6-inch bead, then show me what you got.” I ran what I thought was a beautiful pass — wide, shiny, glassy slag that peeled off in one piece. He looked at it for about two seconds, flipped it over, and pointed at the root. There was no fusion at all. The two plates were barely acquainted. He asked me one question: “Were you pushing or pulling?” I had no idea what he meant. I had been dragging the rod the way my hand felt natural, changing direction every time the puddle looked weird, and hoping for the best.

That moment is one every beginner hits, and it is why the question push or pull stick welding comes up in almost every welding forum, classroom, and shop argument. The short answer that gets passed around — “always drag when there’s slag” — is genuinely useful advice, but it is only half of the story. The full answer depends on your electrode, your position, your polarity, your amperage, the joint design, and what you want the finished weld to look like. Knowing why the drag rule works, and knowing the specific cases where you will deliberately break it, is what separates a welder who follows rules from a welder who understands the puddle.

This guide walks through the entire topic from the ground up: what push and pull mean in SMAW, how each affects penetration, slag behavior, and bead profile, which technique each common rod wants, how to read a finished bead to diagnose your mistakes, and how to practice until the right motion is automatic.

What “Push” and “Pull” Actually Mean in Stick Welding

Before anything else, the vocabulary has to be nailed down, because half the confusion in this debate comes from people using the same words for different motions.

Pull (also called drag, backhand, or trailing) means the electrode tip points back toward the finished bead, and you move the rod away from the direction the tip is angled. The molten puddle sits behind the arc, and the arc is always working into cold, unwelded metal in front of the puddle. If you are welding left to right, your rod tip points back to the left, and you travel to the right. You are, quite literally, dragging the puddle along behind you.

Push (also called forehand or leading) means the electrode tip points in the direction of travel. You are leading with the rod, and the puddle forms ahead of and underneath the electrode, with the arc energy directed at the metal you are about to weld. On that same left-to-right bead, the rod tip points right, into the unwelded joint.

There is a third motion that gets lumped into this discussion: the whip. With a whip, you arc forward out of the puddle, pause, and come back — a stepping motion rather than a steady drag. Whipping is a manipulation style that is usually performed with a trailing (drag) electrode angle, so it is not really a third direction so much as a way of pacing the drag. We will cover it in detail because it is the heart of the 6010 and 6011 technique debate.

The one-sentence rule everyone quotes

The most repeated line in welding education is: “If there’s slag, you must drag.” It rhymes because it works. Processes that generate a protective layer of slag over the finished bead — SMAW, flux-cored arc welding, submerged arc welding, electroslag welding — all behave better when the electrode trails. The reason is mechanical and thermal: when you drag, the arc digs into clean metal and the molten slag flows backward, always covering the bead that has just been deposited. When you push, the arc can blow molten slag forward into unwelded territory, where it gets trapped under the incoming weld metal and becomes a slag inclusion.

That rule is correct often enough that it should be your default setting. But “default” is not “universal,” and a welder who cannot explain the exceptions is one bad test away from being exposed. So let us look at what is physically happening in the puddle under each technique.

How Push and Pull Change the Physics of the Arc

The welding arc is not a neutral heat source — it is a directional jet of ionized gas and plasma traveling at high velocity from the electrode tip into the work. That jet has force, and it has a vector. Where you point it determines where the heat goes, where the molten metal is pushed, and how deep the fusion zone reaches.

What happens when you drag (pull)

With a trailing electrode, the arc force is directed straight into the base metal ahead of the puddle. Heat concentrates in a small area, the arc column digs, and you get a narrow, deep penetration profile. Because the arc is always pointing at cold metal, that metal has not been preheated, so the temperature gradient is steep and the fusion zone runs deeper relative to the bead width. Meanwhile the flux coating is melting behind the arc, and gravity plus the arc’s own back-pressure keep the liquid slag flowing over the solidifying bead where it belongs.

The practical consequences of dragging:

• Deeper, narrower penetration per amp of current.
• Slag stays on top of the finished bead and peels more cleanly.
• The arc and puddle are visible in front of your hand, but the finished bead is hidden behind the rod — you cannot see what you have already laid down until you chip it.
• Travel speed tends to be slower because the arc is fighting cold metal.
• The bead is typically taller and more reinforced, with a more pronounced crown.

What happens when you push

With a leading electrode, the arc jet is directed at metal that is already being warmed by the advancing puddle, and the molten pool is pushed forward in front of the rod. Heat spreads out over a wider area. Penetration gets shallower and broader, the bead flattens, and the toes wet out wider because the puddle is being spread rather than dug.

The practical consequences of pushing:

• Shallower, wider penetration — less dig per amp.
• Flatter bead profile with smoother tie-in at the toes.
• The finished bead is behind your hand, so you can watch and correct it as you go.
• Slag is pushed ahead of the pool, where it can fold under the weld and become trapped.
• Less overall heat input into the joint for a given speed, which matters on thin material.

Side-by-Side Comparison of Push vs Pull in SMAW

The table below summarizes the practical differences you will actually observe at the bench. Treat it as a starting framework; the rod-specific sections later refine it.

Why Slag Changes Everything

To understand why stick welding leans so hard toward dragging, it helps to understand what slag is doing while you weld. The flux coating on a stick electrode burns and melts to form three things: a shielding gas envelope that keeps oxygen away from the puddle, a liquid slag that floats on top of the molten pool, and, in some formulations, alloy additions that condition the weld metal chemistry.

That liquid slag is lighter than steel and it wants to float. In a perfectly oriented drag, the arc force holds the puddle in place while the slag continuously rolls backward over the freshly deposited metal, staying on top where it can do its job of slowing the cooling rate and protecting the bead from nitrogen pick-up. The moment you flip to a push angle, the arc jet starts nudging that liquid slag forward, into the leading edge of the puddle, into the joint root, into any crack or gap you are about to fill. The incoming weld metal then caps it. What you have made is a slag inclusion — a glassy, non-metallic discontinuity inside the weld that acts exactly like a crack under load.

Slag also affects how the bead looks after cleanup. Dragged welds almost always produce a continuous slag crust that lifts in a single ribbon when tapped with a chipping hammer, and often pops loose on its own as the bead cools because the contraction rates differ. Pushed welds leave fragmented, stubborn slag that has to be dug out with a pick. If you have ever spent ten minutes picking gray flakes out of the toes of a bead, you were almost certainly pushing, or your drag angle was far too shallow.

There is a second, subtler slag effect: heat management. Because slag insulates the hot metal, a thick continuous blanket of it slows cooling. Slow cooling is desirable for 7018 low-hydrogen weld metal, which wants a controlled cool to develop its intended microstructure. This is one of the reasons different electrodes want different handling, and it is why the rod-by-rod section of this guide matters more than any single universal rule.

Travel Angle vs Work Angle: Two Angles, Two Jobs

Push versus pull only describes one of the two angles you control with a stick electrode. Welding technique actually runs on a pair of angles, and confusing them is one of the most common sources of bad beads in beginners.

The travel angle (sometimes called the work angle in loose conversation, which is wrong) is the angle between the electrode and the axis of travel — measured in the plane you are moving along. It is the forward or backward lean, and it is what determines whether you are pushing or dragging. A drag angle leans the rod back away from the direction of travel. A push angle leans it forward into the direction of travel. The electrode’s angle to the surface perpendicular is what defines push versus pull, so this is the angle that changes when you switch technique.

The work angle is the angle between the electrode and the joint surfaces themselves, measured across the joint. On a lap joint, the work angle is how much you tilt the rod toward the top plate versus the bottom plate. On a T-fillet it is how you split the 90-degree corner. Work angle controls where the metal goes — which leg of the fillet is bigger, which plate gets more heat — and it is independent of whether you are pushing or pulling.

How much angle is too much

Most stick procedures call for a drag angle somewhere between 5 and 15 degrees off vertical for flat and horizontal work. Lincoln Electric’s published technique notes for low-hydrogen electrodes recommend a 5 to 10 degree drag for 7018, with the rod pointed slightly up in the travel direction. ESAB’s technical guidance puts the general SMAW sweet spot at 5 to 15 degrees of drag for most applications. That range is not arbitrary — it is where arc force is directed deeply enough to fuse the root while still letting slag flow backward.

Past roughly 20 to 25 degrees of travel angle in either direction, the arc jet starts blowing sideways. An excessive drag angle burns the flux coating back unevenly into a sharp overhang — the defect welders call fingernailing — while an excessive push angle blows molten slag forward into the puddle, causing spatter, trapped slag, and cold lap at the toes.

What Each Technique Does to Penetration

Penetration is the depth to which the arc melts into the base metal beneath and beside the deposited filler. It is the single most important mechanical property of a structural weld, because a bead that sits on top of the plate without fusing into it will pass a visual inspection and fail a bend test.

When you drag, the arc column is aimed at room-temperature metal. That metal absorbs heat quickly in a concentrated spot, and the resulting melt-through runs deep. The weld pool then solidifies over that deep, narrow fusion zone. The ratio of depth to width is high. For a root pass on plate, or a keyhole root on pipe, this is exactly what you want: the arc needs to punch through the joint face and tie the two pieces together at the very bottom of the groove.

When you push, the arc is aimed at metal that the puddle has already warmed, and the leading edge of the pool is doing part of the melting before the arc arrives. Heat is distributed over a larger footprint, so the melt is wider and shallower. The depth-to-width ratio drops. On thick material that means you may not get the root fusion you need — the bead looks good on the surface and is cold at the bottom. On thin material, however, that reduced dig is a feature: pushing lets you run along 16- or 18-gauge sheet without blowing holes through it.

Research and shop testing on this are consistent. Comparative trials between forehand and backhand techniques across processes consistently show deeper penetration from the backhand (pull) direction for a given set of parameters. The magnitude varies with electrode type, current, and joint geometry, but the direction of the effect does not flip.

Penetration is not only about direction

It would be dishonest to present push versus pull as if it were the only lever on penetration. It is one of at least five, and the others can overpower it:

• Amperage — the biggest single factor. Moving from 90 A to 130 A with the same rod and direction will change penetration far more than switching from drag to push at fixed amps.
• Arc length — a long arc spreads voltage and heat, widening and shallowing the crater. A tight arc (about one electrode diameter, roughly 1/8 inch for a 1/8-inch rod) concentrates it.
• Travel speed — too fast and the arc never has time to melt the root; too slow and heat builds up, the puddle gets wide and shallow, and you risk burn-through.
• Polarity — DCEP (electrode positive) drives about two-thirds of the heat into the electrode end and gives deeper, more focused penetration on most rods. DCEN reverses the balance. AC splits it. The same drag at the same amps on DCEP and AC will not give the same depth.

Think of technique direction as the fine-tuning knob. Set your amps, arc length, and speed right first; then use push or pull to shape the final result.

Bead Profile: What You Can See From the Outside

Every technique leaves a visible signature on the finished bead, and learning to read it is how you diagnose your own work without cutting the coupon.

A dragged bead is generally narrower than it is tall, with a distinctly convex crown and well-defined toes. The ripples are tight and uniform because the puddle is small and freezing quickly behind the arc. On plate where the amps are right, the toes show a clean, feathered transition into the base metal with no visible step. If the drag angle was excessive or the amps too high, the crown gets even taller and the toes undercut — a small groove melted into the base metal right at the edge of the bead that the filler never filled.

A pushed bead is wider than it is tall, flatter, with a lower crown and toes that wrap further out onto the plate. Ripples are wider and more spread out. The flatter profile is cosmetically appealing, and on a capping pass it can look excellent. The risk is that the wide, flat bead has not been backed by matching depth — it is a wide, shallow puddle sitting on a poorly fused root. Welders call this rolling over or cold lap when the toe edges of the bead sit on the base metal instead of being washed into it.

Reading the bead: a quick diagnostic chart

Arc Length: The Hidden Third Variable

No discussion of push versus pull is complete without arc length, because direction and length interact more than beginners expect.

A stick electrode has to be held at a consistent distance from the puddle — typically about the same as the core wire diameter. For a 1/8-inch 7018, that means roughly 1/8 inch of stand-off. For a 3/32 rod, closer to 3/32. Pulling the rod further away (running a long arc) raises the voltage, widens the arc cone, and spreads heat over a larger area — which mimics some of the effects of pushing: wider, shallower, more spatter, more porosity. So a welder who thinks they are dragging but is actually running a long arc gets results that look suspiciously like a push.

Pushing makes a long arc worse, spreading heat ahead over unwelded metal. Lincoln Electric’s low-hydrogen guidance is blunt: a long arc is the most common cause of porosity in 7018 work, and it also destabilizes the slag blanket.

Rod by Rod: What Each Electrode Actually Wants

Here is where the blanket rule “always drag” starts to flex. Every electrode family has a personality formed by its flux chemistry, and the ideal travel direction and manipulation were designed around that personality. Running every rod the same way is a little like driving a truck, a sports car, and a tractor with identical technique — it will move, but it will not be right.

E6010: the whip-and-drag workhorse

E6010 is the deep-penetration, high-cellulose-sodium rod that defines pipe root passes. Its coating generates a huge volume of gas, produces a hot, forceful, almost spattery arc (often described as “digging”), and freezes the puddle quickly. It runs on DCEP only and it does not care about rust, mill scale, or light contamination — the arc burns through it.

The text-book handling for 6010 is the whip and pause: drag the electrode forward a short distance into the joint, then whip the tip back out of the puddle and up, letting the small fast-freeze crater solidify, then return to the puddle and repeat. The whip is executed with the rod angled in the trailing (drag) direction while the motion is rapid, deliberate stepping — not a smooth push at all.

Why this works: the 6010 puddle freezes in a small window. The whip pulls heat out of the puddle long enough for it to skin over, which prevents it from running away downhill or blowing through an open root. The pause on the way in lets the arc bore to the bottom of the joint. The classic 6010 is whipped and dragged — drag in, flick out of the puddle to let it freeze, return. The whip direction itself is a drag. That control is why 6010 excels on open roots.

• Direction: whip-and-pause, always trailing (drag). Never a steady push.
• Travel angle: roughly 10–15° drag, adjusted to control the keyhole.

E6011: 6010’s AC-friendly cousin

E6011 uses a potassium-based cellulose flux, which lets it run on AC as well as DC. On a transformer buzz-box — still millions of machines in garages and farm shops — 6011 is the only way to get that deep, dirty-metal-digging arc. Its puddle behavior and handling match 6010 closely.

The recommended technique is the same whip-and-pause drag, adapted to the slightly less forceful arc. On AC you get less arc “push,” so the whip is usually a little more deliberate and the pauses a little longer.

• Direction: whip-and-pause drag on DC; steady drag acceptable on AC.
• Travel angle: 10–15° drag.
• Best use: repair work, dirty steel, farm equipment, AC-only machines.

E6013: the smooth beginner rod

E6013 is a rutile-coated rod with a gentle, quiet arc, light spatter, and shallow penetration. It is the rod most schools put in your hand first because it strikes and holds an arc with very little skill. Its slag is sticky but its puddle is forgiving, and it runs on AC or DC with either polarity, which makes it very tolerant of a machine that is set wrong.

Because its penetration is deliberately modest, 6013 works best in a steady, comfortable drag, a little slower and lower in the amps than you might expect, so the puddle has time to wet into the toes. Pushing a 6013 turns its already-shallow dig into a pure surface deposit — wide, flat, and cosmetic. On thin sheet metal, a slight push can be justified to avoid burn-through, but you are trading soundness for appearance there.

• Direction: steady drag.
• Travel angle: 10–15° drag; shallow push only for thin sheet.
• Best use: learning, sheet metal, mild-steel repairs with clean surfaces.

E7018: the low-hydrogen structural rule-setter

E7018 is the rod that most welders think of when they hear “stick welding,” and it is the one with the clearest, most documented technique. It is a low-hydrogen electrode whose flux contains iron powder; the coating burns hot, thick, and protective. Its welds are strong, ductile, and resistant to hydrogen cracking, which is why everything structural — bridges, columns, pressure vessels, code work — runs on 7018 or its cousins (7016, 8018, 10018).

The accepted handling is a flat drag at 5 to 10 degrees, with a short arc, no whipping, no weaving on root and single-run fillets. Lincoln Electric’s technique notes for low-hydrogen rods are explicit: keep a tight arc, do not whip the rod out of the puddle, hold a 5-10° drag slightly pointing up in the travel direction, and keep the coating from fingernailing at steep angles.

Why 7018 wants a clean, quiet drag instead of a whip: its puddle is bigger and more fluid than the fast-freeze rods, and it is designed to let dissolved hydrogen out of the weld metal as it cools. Whipping defeats the slow-cool blanket and stirs in slag. A smooth, consistent drag at correct heat produces the shiny, “oil-can” ribbed bead with slag that lifts off in a single ribbon.

• Direction: steady drag, 5–10°.
• Arc length: as short as you can hold.
• Manipulation: stringer; gentle weave only on multi-pass fill caps.

E7024 / iron-powder rods: the drag they were built for

E7024 is a high iron-powder rod engineered for maximum deposition rate in the flat and horizontal positions. Its coating is overloaded with iron powder that becomes additional weld metal, so it fills joints fast. It does not want to be whipped, pushed, or maneuvered — it wants to be dragged flat along the joint with the coating riding on the plate.

The published technique is a contact drag: let the flux coating just skim the work, keep a steady 10–15° drag, and let the heavy puddle bank up behind the arc. The iron powder handles the fill while the drag guarantees the root gets warmed by the digging arc. Push a 7024 and that deposition advantage disappears.

Push or Pull in Every Welding Position

Position is the other variable that bends the rule. The drag default is built around flat welding, where gravity cooperates. As soon as you rotate gravity, the puddle starts fighting you, and some positions genuinely call for pushing — not because the slag rule changed, but because the arc force against gravity changes which technique is stable.

Flat (1G / 1F)

This is the reference position. Gravity holds the puddle, slag floats cleanly, and the drag technique works text-book-perfect. Run every rod exactly as described above: 10–15° drag for general-purpose rods, 5–10° for 7018. This is where you establish your baseline settings before moving to harder positions.

Horizontal (2G / 2F)

A horizontal fillet is close to flat in feel: the drag works, with the work angle biased slightly (typically 60/40 toward the vertical plate so the weld is centered on the joint) and a drag angle of 10–15°. The puddle wants to sag down the vertical plate; a slight upward tilt of the rod plus consistent drag holds it in place. Pushing a horizontal fillet is a guaranteed recipe for a sagging, porous, flat bead with trapped slag.

Vertical-up (3G / 3F, up)

Vertical-up is where the honest exceptions live. With the puddle defying gravity, many welders discover that a slightly forward-leaning angle — a shallow push — helps prop the pool against the wall of weld metal below. ESAB’s position-specific guidance, for example, describes vertical-up progression with the electrode tilted 5–10° toward the direction of travel to support the pool against gravity.

Two distinctions matter. First, vertical-up “push” is tiny — 5–10° of forward lean, not the 15–20° aggressive push you would use on MIG sheet metal. Second, it is used mainly with the fast-freeze rods (6010, 6011) and with weaves on 7018 cover passes, where the motion is usually a deliberate triangle or crescent that both supports the puddle and keeps slag trailing. A blind, head-down straight-line push uphill will drop slag into the toe of the pass below it.

Vertical-down (3F, down)

Vertical-down is fast, and it is used for thin-gauge sheet and pipe downhand. The classic technique is exactly a drag — often a deliberate, exaggerated drag of about 15–20°, letting the puddle drain downhill and the slag trail behind. Heavy rods like 6010/6011 downhill are run with a step-drag. The 7018 is not run downhill outside of specific structural procedures, because its fluid puddle runs away from you.

Overhead (4G / 4F)

Overhead, gravity grabs the puddle the moment you stop holding it, so the discipline is all about a tight, short arc and minimal manipulation. Nearly everyone runs overhead with a drag — usually closer to perpendicular than in flat, because the rod is pushing metal up against gravity and any lean just makes the puddle drool. Keep amps at the bottom third of the rod’s range and reduce the travel angle to near zero so arc force is doing the work of holding weld metal up.

How Polarity Interacts With Your Technique

Polarity decides where the heat lives, and therefore how much effort your technique has to put into penetration.

DCEP (electrode positive, reverse polarity) puts roughly two-thirds of arc heat into the electrode end, with a strong, digging arc and deep, narrow fusion. Almost every consumable we have discussed is built for this — E6010, E6011 (also AC), E7018 (also AC in many formulations). With DCEP you can often use a shallower drag angle and a faster travel, because the arc is doing the digging.

DCEN (electrode negative, straight polarity) puts most of the heat into the work, so the parent metal melts more eagerly while the electrode melts more slowly. Penetration is shallower and broader. It is used for thin sheet, surfacing, and cast-iron repair with specific rods, and it does not appear on most common electrodes’ data sheets.

AC splits the heat and suffers an arc that extinguishes and re-ignites 120 times per second on 60 Hz supply. The puddle is less emphatic at every edge — shallower dig, broader melt. Technique on AC leans toward a slightly tighter arc and a slightly slower drag to compensate. The direction rule does not change.

The takeaway: direction is about slag control; polarity is about heat distribution. You cannot fix a polarity/penetration problem with push versus pull, and you cannot fix a slag problem with polarity. Match them correctly and the weld follows.

Common Mistakes: How Beginners (and Pros) Get Push vs Pull Wrong

Most bad stick welds are traceable to a small set of repeatable technique failures. Here are the ones that show up constantly in classes, forums, and quality reports — ranked roughly by how often they actually happen.

1. Drifting into a push without noticing

You start the bead with a perfect drag and end it pushing. This is the most common failure of all, and it is almost never deliberate. As the rod burns down, welders unconsciously rotate their wrist forward or backwards, or they lean into the work, and the angle flips. The bead starts tall and clean and gradually goes wide, flat, and slag-choked by the last inch. If your beads are consistently a good start with a bad ending, suspect your angle is drifting rather than your settings.

Fix: check your angle at the halfway point of every bead for a week. Better yet, weld a bead, stop, and look at your hand position before you move the coupon. Rebuilding angle discipline is faster than re-diagnosing the puddle.

2. Applying MIG instincts to stick

Welders who learned MIG first (or who watch a lot of MIG video) have push burned into muscle memory. When they pick up a stick, they naturally push, and nothing about stick looks wrong until they chip a weld full of slag. The mental shortcut is real: in MIG, push for looks on sheet and pull for penetration on plate; in stick, drag for almost everything, always. If you came from MIG, say it out loud three times before every practice session until the muscle memory overwrites.

3. “Whipping” a low-hydrogen rod

The whip is a fantastic tool for 6010 and 6011. It is actively wrong for 7018. When welders inherit the whip from their 6010 practice and apply it to 7018, the result is a rough, lumpy bead with slag pinched into the valleys between whips and, frequently, porosity where the long arc let air in. Whip the fast-freeze rods; drag the low-hydrogen rods. No exceptions.

4. Fighting cold-lap on thin stock by pushing harder

When a pushed bead won’t fuse on sheet metal, the natural response is to push harder or faster to “flatten it out.” That makes cold lap worse. The correct response is the opposite of intuitive: slow down, tighten the arc, and let the puddle dwell long enough to wash both toes in — while keeping the slag trailing. On very thin stock, drop the rod size (3/32 instead of 1/8) and reduce amps before you change direction.

5. Ignoring the amperage and blaming technique

Low amps produce a sluggish puddle that looks like it needs a push to “get going.” High amps produce a runny, wide, undercut puddle that looks like it needs a drag to “slow down.” Beginners spiral between pushing and pulling to fix what is actually a current problem. Set the amps from the rod manufacturer’s chart for your thickness, run one test bead, and then use direction as a fine-tune — not the other way around.

How to Diagnose a Bead: Deciding Push or Pull After the Fact

Every pair of eyes can learn to look at a finished bead and correctly reconstruct whether it was pushed or dragged. This is the skill that makes the “flip it over” exercise possible.

If the bead is tall and narrow, uniformly rippled, with clean toes and one-piece slag, it was dragged correctly. Ripple spacing tells you the speed and the rod: tight ripples with 6010 mean a proper whip cadence; wide, even, “shingled” ripples on 7018 mean a steady drag at proper heat.

If the bead is flat and wide, sits over a cold-looking root, or trails a residue of gray specks at the toes, it was pushed, or the angle flipped partway. A bead that is flat at the end and tall at the start was dragged at the start and pushed at the end — angle drift.

If the bead is tall but the edges have deep grooves, the drag angle was too steep or the amps were too high — the arc blew the toes out and never filled them. This one can look similar to a push result at a glance, but the giveaway is the undercut groove shape.

If the bead is missing, rough, and intermittently fused, the travel was too fast, the angle was unstable, or the arc length was bouncing. Direction is the least of your problems.

If there is a glassy, gray streak running lengthwise through the weld face, that is trapped slag and it means the slag got ahead of the puddle at some point — a classic push signature, or a drag angle collapsed to near zero mid-bead.

The most instructive exercise in all of stick welding is also the cheapest: run six beads on one coupon, three dragged and three pushed, using the same rod and amps. Number them. Chip and wire-brush them. Try to sort them purely by appearance, then flip the coupon and look at the roots. You will never watch an instructor’s demo the same way again.

Practice Drills That Rewire Muscle Memory

Reading about angles changes nothing until your hands agree. The following drills are ordered to stack — each one assumes the previous one is automatic. Run them on a clean scrap plate at least 1/4 inch (6 mm) thick so burn-through is never the variable.

Drill 1: The dry-run angle hold

Hold a rod in your usual grip, no power on. Set a 15° drag by laying a carpenter’s bevel gauge or a simple paper template next to the joint, freeze your wrist, and trace the joint from end to end on a dead machine. Then do it with a 5° drag, then a 20° push, then a 5° push. Do 10 clean passes in each; you are rewiring the reference positions without wasting rod.

Drill 2: Six-inch straight drags with 7018

Set your machine per the rod chart for your plate thickness (for 1/8-inch 7018 on 1/4-inch plate, start around 110–120 A DCEP). Weld six-inch beads in a steady 5–10° drag, tight arc, no whip. The entire bead should be uniform — same height, same width, same ripple spacing for all six inches. Any bead that gives you a start, middle, or end that differs is telling you something real.

Drill 3: The 6010 whip cadence

On the same coupon, run 6010 at the low-mid of its range with a whip-and-pause: a short forward drag, an upward flick out of the puddle (the bead’s edge is your lid so the crater just skins), a brief pause, back in. The result at the end of the plate should look like a tight stack of overlapping coins. If it looks like beads of a necklace — separated, round-topped drops — you whipped too far out. If it looks like a continuous wavy line, you did not leave the puddle.

Drill 4: The A/B test

Weld pairs: one bead dragged, one pushed, same settings, every time. Chip both, look at toes, slag crust, and root side. Your brain needs to see the two results next to each other dozens of times before the feedback loop becomes automatic at the machine.

Practical Tips From the Bench

Push vs Pull Compared With Other Processes

A lot of the confusion around “push or pull stick welding” leaks in from tutorial videos about other processes, so it is worth locking down where stick fits relative to its neighbors.

Alternatives to other wire processes? See good versus bad MIG welds, how to use a wire welder, and MIG welding history.

Flux-cored (FCAW). Like stick, flux-cored generates slag, so the drag rule returns. Self-shielded wires in particular are dragged at 10–15° to keep slag behind; pushing a self-shielded wire produces porosity almost instantly.

TIG (GTAW). TIG has no slag and no consumable electrode angle to speak of — the torch is held nearly vertical and the technique question becomes how to feed the filler wire. The push/pull vocabulary basically does not apply; whatever lean you use is for visibility, not physics.

The mental rule that ties it all together: if the process makes slag, drag it. That single sentence resolves 80 percent of the push-or-pull arguments you will ever hear.

Health, Safety, and the Ones Who Have To Stand Next To You

A practical article about technique would be incomplete without the safety note, because “push harder into a bad angle” is also how people weld in a hurry and get hurt.

The immediate hazards around stick welding are stable and well documented: ultraviolet and infrared arc radiation that can burn eyes and skin in seconds; hot slag that spalls off beads and rolls into boot tops; flying scale when you chip slag with a hammer; and the position injuries that come from welding overhead while angling your body into joints for hours.

Welding fume is the one that sneaks up on you. SMAW fume contains oxides of the metals in the coating and base metal — including manganese — and exposure depends heavily on your distance from the arc and your ventilation. Dragging keeps your head out of the plume a little better than leaning in to inspect a push, which is one more quiet argument for good fundamental posture. Run fume extraction or strong local ventilation, keep your hood lens clean, and never weld plated, painted, or galvanized metal without respiratory protection and coating removal.

Electrically, SMAW is unforgiving: it is a live circuit with your body in the middle if you mishandle the holder. Insulated gloves, dry clothing, and an insulated stand or dry boots are basics. Give every joint, every weld, and every bead the respect the process demands.

Push or Pull Stick Welding: The FAQ Everyone Asks

Should I push or pull when stick welding?

Drag (pull). Stick welding makes slag, and dragging keeps that slag trailing behind the arc on top of the finished bead. Exceptions are narrow and position-specific: a subtle 5–10° forward tilt on vertical-up passes with fast-freeze rods, and occasionally a shallow push on thin sheet to avoid burn-through. If you take nothing else from this article, default to drag.

What angle should I drag a stick rod?

For most rods in flat and horizontal positions, 10–15° back from vertical is the standard guidance. For E7018, the recommended drag is tighter — about 5–10° — with the shortest arc you can hold. Overhead, drop the degree closer to vertical. Source manufacturers’ technique notes agree on roughly these numbers for the common rods.

Does 7018 burn better being dragged or pushed?

Dragged. E7018 is a low-hydrogen rod whose technique is documented in manufacturer education materials: a short arc, no whipping, 5–10° of drag, and no weaving on root passes and single-pass fillets. Its puddle is fluid and large, and the drag lets the protective slag blanket float and cool it slowly. Pushing invites trapped slag and porosity.

Do you drag or whip a 6010?

Both, in sequence. The classic 6010 shed is a whip-and-pause executed with a trailing rod: drag in, flick out of the puddle to let it freeze, return. The whip direction itself is a drag; the motion is the “manipulation.” For straight-line pipe root passes, some welders find a steady drag works fine at the right amps, but the whip is the standard for controlling the fast-freeze puddle and the keyhole.

Why does flux-cored wire get dragged but MIG gets pushed?

Because flux-cored generates slag and MIG does not. The slag-trailing rule exists to keep slag out of the solidifying weld. MIG’s gas shield does that job without slag, so its push/pull choice is free to be decided on penetration and bead cosmetics instead.

Can pushing ever be better in stick welding?

Yes, in narrow cases: vertical-up with a slight forward lean supports the pool against gravity, and very thin sheet metal can tolerate a shallow push to keep heat input down.

My beads are flat and wide with gray specks at the toes. What is wrong?

You are pushing (or your drag angle collapsed to zero mid-pass) and slag is running ahead of the puddle and getting trapped. Confirm by checking the root side for shallow fusion. Re-establish a firm drag, tighten the arc, and slow down slightly; if it persists, raise amps modestly.

Why do my 7018 welds have a low, flat, wide profile even when I drag?

Amp selection or travel speed is pulling the puddle wide. Common causes: amps too high for the rod size, travel too slow, or a long arc. The low-hydrogen technique wants modest amps, a tight arc, and a steady pace so the bead crowns correctly. Try dropping 10 A and increasing travel speed slightly.

Is push or pull better for the root pass on pipe?

For an open root, the working positions call for 6010/6011 with a whip-and-pause that trails the puddle — effectively drag-in, flick-back.

I learned MIG first. How do I break the push habit?

Practice the dry-run angle holds from the drills above, run the A/B comparison drill with the same rod, and keep the rule on a sticky note at the machine: slag → drag.

What does it mean to say a rod “runs downhill” vs “uphill”?

That is travel direction relative to gravity, not push versus pull. Thin-gauge vertical-down work is dragged downhill at speed with fast-freeze rods. Structural vertical-up work is dragged uphill, with the puddle held in place by arc force and small weaves. Both are drags; the position is what changes.

Does my handedness change the push/pull rule?

No. The rule is defined by where the electrode points relative to the direction of travel, not your body. A left-handed dragger traveling right-to-left is still dragging. What changes is which hand holds the rod and how you brace, but the physics and the slag rule are identical.

The Quick-Reference Cheat Sheet

Here is the whole article condensed to a single table taped to the lid of your welding box.

The Bottom Line: Rules You Can Keep for Life

Big picture time. After all the exceptions, angles, rods, and positions, the whole lesson collapses into four sentences you can carry into any shop in the world.

1. If there’s slag, you drag. SMAW, FCAW, SAW — the slag-trailing rule is universal across them. Pushing a stick rod traps slag, cuts penetration, and flattens the bead into a cosmetic lie.

2. Direction is the fine-tune, not the fix. Amps, arc length, travel speed, and polarity shape a weld more than your hand angle. Set the process right first, then clean up the last 10 percent with technique.

3. Every rod has its mood. 6010 and 6011 want a whip-and-pause drag; 7018 wants a quiet, tight, 5–10° drag with no whipping; iron-powder rods want their coating skimmed along the plate.

4. The bead is the teacher. Tall and rippled with clean toes means you dragged right. Wide and flat with gray toes means your angle collapsed. Flip the coupon and let the root side be the final judge — what the surface looks like when you are pushing is not what the joint needs.

The day my instructor flipped my first coupon over, he was not being cruel. He was testing whether I could see what the process physically demanded versus what my hand felt comfortable doing. Once you have felt the difference between a drag that digs and a push that smears, you will never weld the same way again — and when some beginner in your shop asks you push or pull stick welding, you will have the whole answer ready.

Authoritative Sources (Outbound References)

These technical references support the electrode techniques, aluminum comparisons, angle recommendations, and safety guidance covered in this article.

• Lincoln Electric — “Low Hydrogen SMAW Electrodes: Techniques for Success”
• Lincoln Electric — “Stick Electrode: Welding Procedures and Techniques”
• ESAB — “Stick Welding Electrode Angle Control (SMAW Guide)”
• American Welding Society — AWS Standards & Certification Information
• U.S. Occupational Safety and Health Administration — “Welding, Cutting, and Brazing” (OSHA 1910 Subpart Q)
• Centers for Disease Control and Prevention NIOSH — “Welding and Manganese: Potential Neurologic Effects”
• CWB Group — “Stick vs. Push-Pull MIG on Aluminum: What You Need to Know”

Always consult the manufacturer’s data sheet for the exact electrode you are running and the applicable code (AWS D1.1 and similar) for your jurisdiction before welding structural components. When in doubt, weld a test coupon and bend it before you trust a new rod or a new technique.